Tolto un controllo inutile sul nome che finisce con .pdf

Change get backend method for Lattice and run tests

.github/FUNDING.yml

@@ -0,0 +1 @@
+open_collective: camelot

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,57 @@
+---
+name: Bug report
+about: Please follow this template to submit bug reports.
+title: ''
+labels: bug
+assignees: ''
+
+---
+
+<!-- Please read the filing issues section of the contributor's guide first: https://camelot-py.readthedocs.io/en/master/dev/contributing.html -->
+
+**Describe the bug**
+
+<!-- A clear and concise description of what the bug is. -->
+
+**Steps to reproduce the bug**
+
+<!-- Steps used to install `camelot`:
+1. Add step here (you can add more steps too) -->
+
+<!-- Steps to be used to reproduce behavior:
+1. Add step here (you can add more steps too) -->
+
+**Expected behavior**
+
+<!-- A clear and concise description of what you expected to happen. -->
+
+**Code**
+
+<!-- Add the Camelot code snippet that you used. -->
+
+```
+import camelot
+
+# add your code here
+```
+
+**PDF**
+
+<!-- Add the PDF file that you want to extract tables from. -->
+
+**Screenshots**
+
+<!-- If applicable, add screenshots to help explain your problem. -->
+
+**Environment**
+
+- OS: [e.g. macOS]
+- Python version:
+- Numpy version:
+- OpenCV version:
+- Ghostscript version:
+- Camelot version:
+
+**Additional context**
+
+<!-- Add any other context about the problem here. -->

.github/workflows/tests.yml

@@ -0,0 +1,44 @@
+name: tests
+
+on: [pull_request]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: [3.6, 3.7, 3.8]
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install camelot with dependencies
+        run: |
+          make install
+      - name: Test with pytest
+        run: |
+          make test
+
+  test_latest:
+    name: Test on ${{ matrix.os }} with Python 3.9
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest, windows-latest]
+        python-version: [3.9]
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install camelot with dependencies
+        run: |
+          make install
+      - name: Test with pytest
+        run: |
+          make test

.gitignore

@@ -1,3 +1,4 @@
+fontconfig/
 __pycache__/
 *.py[cod]
 *.so
@@ -12,5 +13,8 @@ coverage.xml
 .pytest_cache/
 _build/
 
+.venv/
+htmlcov/
+
 # vscode
 .vscode

.readthedocs.yml

@@ -0,0 +1,27 @@
+# .readthedocs.yml
+# Read the Docs configuration file
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+# Required
+version: 2
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/conf.py
+
+# Build documentation with MkDocs
+#mkdocs:
+#  configuration: mkdocs.yml
+
+# Optionally build your docs in additional formats such as PDF
+formats:
+  - pdf
+
+# Optionally set the version of Python and requirements required to build your docs
+python:
+  version: 3.8
+  install:
+    - method: pip
+      path: .
+      extra_requirements:
+        - dev

.travis.yml

@@ -1,20 +0,0 @@
-sudo: false
-language: python
-cache: pip
-python:
-  - "2.7"
-  - "3.5"
-  - "3.6"
-matrix:
-  include:
-    - python: 3.7
-      dist: xenial
-      sudo: true
-before_install:
-  - sudo apt-get install python-tk python3-tk ghostscript
-install:
-  - pip install ".[dev]"
-script:
-  - pytest --verbose --cov-config .coveragerc --cov-report term --cov-report xml --cov=camelot tests
-after_success:
-  - codecov --verbose

CONTRIBUTING.md

@@ -16,14 +16,14 @@ As the [Requests Code Of Conduct](http://docs.python-requests.org/en/master/dev/
 
 ## Your first contribution
 
-A great way to start contributing to Camelot is to pick an issue tagged with the [help wanted](https://github.com/socialcopsdev/camelot/labels/help%20wanted) tag or the [good first issue](https://github.com/socialcopsdev/camelot/labels/good%20first%20issue) tag. If you're unable to find a good first issue, feel free to contact the maintainer.
+A great way to start contributing to Camelot is to pick an issue tagged with the [help wanted](https://github.com/camelot-dev/camelot/labels/help%20wanted) tag or the [good first issue](https://github.com/camelot-dev/camelot/labels/good%20first%20issue) tag. If you're unable to find a good first issue, feel free to contact the maintainer.
 
 ## Setting up a development environment
 
 To install the dependencies needed for development, you can use pip:
 
 <pre>
-$ pip install camelot-py[dev]
+$ pip install "camelot-py[dev]"
 </pre>
 
 Alternatively, you can clone the project repository, and install using pip:
@@ -36,7 +36,7 @@ $ pip install ".[dev]"
 
 ### Submit a pull request
 
-The preferred workflow for contributing to Camelot is to fork the [project repository](https://github.com/socialcopsdev/camelot) on GitHub, clone, develop on a branch and then finally submit a pull request. Here are the steps:
+The preferred workflow for contributing to Camelot is to fork the [project repository](https://github.com/camelot-dev/camelot) on GitHub, clone, develop on a branch and then finally submit a pull request. Here are the steps:
 
 1. Fork the project repository. Click on the ‘Fork’ button near the top of the page. This creates a copy of the code under your account on the GitHub.
 
@@ -106,7 +106,7 @@ The function docstrings are written using the [numpydoc](https://numpydoc.readth
 
 ## Filing Issues
 
-We use [GitHub issues](https://github.com/socialcopsdev/camelot/issues) to keep track of all issues and pull requests. Before opening an issue (which asks a question or reports a bug), please use GitHub search to look for existing issues (both open and closed) that may be similar.
+We use [GitHub issues](https://github.com/camelot-dev/camelot/issues) to keep track of all issues and pull requests. Before opening an issue (which asks a question or reports a bug), please use GitHub search to look for existing issues (both open and closed) that may be similar.
 
 ### Questions
 

HISTORY.md

@@ -1,6 +1,233 @@
 Release History
 ===============
 
+master
+------
+
+0.10.1 (2021-07-11)
+------------------
+
+- Change extra requirements from `cv` to `base`. You can use `pip install "camelot-py[base]"` to install everything required to run camelot.
+
+0.10.0 (2021-07-11)
+------------------
+
+**Improvements**
+
+- Add support for multiple image conversion backends. [#198](https://github.com/camelot-dev/camelot/pull/198) and [#253](https://github.com/camelot-dev/camelot/pull/253) by Vinayak Mehta.
+- Add markdown export format. [#222](https://github.com/camelot-dev/camelot/pull/222/) by [Lucas Cimon](https://github.com/Lucas-C).
+
+**Documentation**
+
+- Add faq section. [#216](https://github.com/camelot-dev/camelot/pull/216) by [Stefano Fiorucci](https://github.com/anakin87).
+
+0.9.0 (2021-06-15)
+------------------
+
+**Bugfixes**
+
+- Fix use of resolution argument to generate image with ghostscript. [#231](https://github.com/camelot-dev/camelot/pull/231) by [Tiago Samaha Cordeiro](https://github.com/tiagosamaha).
+- [#15](https://github.com/camelot-dev/camelot/issues/15) Fix duplicate strings being assigned to the same cell. [#206](https://github.com/camelot-dev/camelot/pull/206) by [Eduardo Gonzalez Lopez de Murillas](https://github.com/edugonza).
+- Save plot when filename is specified. [#121](https://github.com/camelot-dev/camelot/pull/121) by [Jens Diemer](https://github.com/jedie).
+- Close file streams explicitly. [#202](https://github.com/camelot-dev/camelot/pull/202) by [Martin Abente Lahaye](https://github.com/tchx84).
+- Use correct re.sub signature. [#186](https://github.com/camelot-dev/camelot/pull/186) by [pevisscher](https://github.com/pevisscher).
+- [#183](https://github.com/camelot-dev/camelot/issues/183) Fix UnicodeEncodeError when using Stream flavor by adding encoding kwarg to `to_html`. [#188](https://github.com/camelot-dev/camelot/pull/188) by [Stefano Fiorucci](https://github.com/anakin87).
+- [#179](https://github.com/camelot-dev/camelot/issues/179) Fix `max() arg is an empty sequence` error on PDFs with blank pages. [#189](https://github.com/camelot-dev/camelot/pull/189) by Vinayak Mehta.
+
+**Improvements**
+
+- Add `line_overlap` and `boxes_flow` to `LAParams`. [#219](https://github.com/camelot-dev/camelot/pull/219) by [Arnie97](https://github.com/Arnie97).
+- [Add bug report template.](https://github.com/camelot-dev/camelot/commit/0a3944e54d133b701edfe9c7546ff11289301ba8)
+- Move from [Travis to GitHub Actions](https://github.com/camelot-dev/camelot/pull/241).
+- Update `.readthedocs.yml` and [remove requirements.txt](https://github.com/camelot-dev/camelot/commit/7ab5db39d07baa4063f975e9e00f6073340e04c1#diff-cde814ef2f549dc093f5b8fc533b7e8f47e7b32a8081e0760e57d5c25a1139d9)
+
+**Documentation**
+
+- [#193](https://github.com/camelot-dev/camelot/issues/193) Add better checks to confirm proper installation of ghostscript. [#196](https://github.com/camelot-dev/camelot/pull/196) by [jimhall](https://github.com/jimhall).
+- Update `advanced.rst` plotting examples. [#119](https://github.com/camelot-dev/camelot/pull/119) by [Jens Diemer](https://github.com/jedie).
+
+0.8.2 (2020-07-27)
+------------------
+
+* Revert the changes in `0.8.1`.
+
+0.8.1 (2020-07-21)
+------------------
+
+**Bugfixes**
+
+* [#169](https://github.com/camelot-dev/camelot/issues/169) Fix import error caused by `pdfminer.six==20200720`. [#171](https://github.com/camelot-dev/camelot/pull/171) by Vinayak Mehta.
+
+0.8.0 (2020-05-24)
+------------------
+
+**Improvements**
+
+* Drop Python 2 support!
+    * Remove Python 2.7 and 3.5 support.
+    * Replace all instances of `.format` with f-strings.
+    * Remove all `__future__` imports.
+    * Fix HTTP 403 forbidden exception in read_pdf(url) and remove Python 2 urllib support.
+    * Fix test data.
+
+**Bugfixes**
+
+* Fix library discovery on Windows. [#32](https://github.com/camelot-dev/camelot/pull/32) by [KOLANICH](https://github.com/KOLANICH).
+* Fix calling convention of callback functions. [#34](https://github.com/camelot-dev/camelot/pull/34) by [KOLANICH](https://github.com/KOLANICH).
+
+0.7.3 (2019-07-07)
+------------------
+
+**Improvements**
+
+* Camelot now follows the Black code style! [#1](https://github.com/camelot-dev/camelot/pull/1) and [#3](https://github.com/camelot-dev/camelot/pull/3).
+
+**Bugfixes**
+
+* Fix Click.HelpFormatter monkey-patch. [#5](https://github.com/camelot-dev/camelot/pull/5) by [Dimiter Naydenov](https://github.com/dimitern).
+* Fix strip_text argument getting ignored. [#4](https://github.com/camelot-dev/camelot/pull/4) by [Dimiter Naydenov](https://github.com/dimitern).
+* [#25](https://github.com/camelot-dev/camelot/issues/25) edge_tol skipped in read_pdf. [#26](https://github.com/camelot-dev/camelot/pull/26) by Vinayak Mehta.
+* Fix pytest deprecation warning. [#2](https://github.com/camelot-dev/camelot/pull/2) by Vinayak Mehta.
+* [#293](https://github.com/socialcopsdev/camelot/issues/293) Split text ignores all text to the right of last cut. [#294](https://github.com/socialcopsdev/camelot/pull/294) by Vinayak Mehta.
+* [#277](https://github.com/socialcopsdev/camelot/issues/277) Sort TableList by order of tables in PDF. [#283](https://github.com/socialcopsdev/camelot/pull/283) by [Sym Roe](https://github.com/symroe).
+* [#312](https://github.com/socialcopsdev/camelot/issues/312) `table_regions` throws `ValueError` when `flavor='stream'`. [#332](https://github.com/socialcopsdev/camelot/pull/332) by Vinayak Mehta.
+
+0.7.2 (2019-01-10)
+------------------
+
+**Bugfixes**
+
+* [#245](https://github.com/socialcopsdev/camelot/issues/245) Fix AttributeError for encrypted files. [#251](https://github.com/socialcopsdev/camelot/pull/251) by Yatin Taluja.
+
+
+0.7.1 (2019-01-06)
+------------------
+
+**Bugfixes**
+
+* Move ghostscript import to inside the function so Anaconda builds don't fail.
+
+0.7.0 (2019-01-05)
+------------------
+
+**Improvements**
+
+* [#240](https://github.com/socialcopsdev/camelot/issues/209) Add support to analyze only certain page regions to look for tables. [#243](https://github.com/socialcopsdev/camelot/pull/243) by Vinayak Mehta.
+    * You can use `table_regions` in `read_pdf()` to specify approximate page regions which may contain tables.
+    * Kwarg `line_size_scaling` is now called `line_scale`.
+* [#212](https://github.com/socialcopsdev/camelot/issues/212) Add support to export as sqlite database. [#244](https://github.com/socialcopsdev/camelot/pull/244) by Vinayak Mehta.
+* [#239](https://github.com/socialcopsdev/camelot/issues/239) Raise warning if PDF is image-based. [#240](https://github.com/socialcopsdev/camelot/pull/240) by Vinayak Mehta.
+
+**Documentation**
+
+* Remove mention of old mesh kwarg from docs. [#241](https://github.com/socialcopsdev/camelot/pull/241) by [fte10kso](https://github.com/fte10kso).
+
+**Note**: The python wrapper to Ghostscript's C API is now vendorized under the `ext` module. This was done due to unavailability of the [ghostscript](https://pypi.org/project/ghostscript/) package on Anaconda. The code should be removed after we submit a recipe for it to conda-forge. With this release, the user doesn't need to ensure that the Ghostscript executable is available on the PATH variable.
+
+0.6.0 (2018-12-24)
+------------------
+
+**Improvements**
+
+* [#91](https://github.com/socialcopsdev/camelot/issues/91) Add support to read from url. [#236](https://github.com/socialcopsdev/camelot/pull/236) by Vinayak Mehta.
+* [#229](https://github.com/socialcopsdev/camelot/issues/229), [#230](https://github.com/socialcopsdev/camelot/issues/230) and [#233](https://github.com/socialcopsdev/camelot/issues/233) New configuration parameters. [#234](https://github.com/socialcopsdev/camelot/pull/234) by Vinayak Mehta.
+    * `strip_text`: To define characters that should be stripped from each string.
+    * `edge_tol`: Tolerance parameter for extending textedges vertically.
+    * `resolution`: Resolution used for PDF to PNG conversion.
+    * Check out the [advanced docs](https://camelot-py.readthedocs.io/en/master/user/advanced.html#strip-characters-from-text) for usage details.
+* [#170](https://github.com/socialcopsdev/camelot/issues/170) Add option to pass pdfminer layout kwargs. [#232](https://github.com/socialcopsdev/camelot/pull/232) by Vinayak Mehta.
+    * Keyword arguments for [pdfminer.layout.LAParams](https://github.com/euske/pdfminer/blob/master/pdfminer/layout.py#L33) can now be passed using `layout_kwargs` in `read_pdf()`.
+    * The `margins` keyword argument in `read_pdf()` is now deprecated.
+
+0.5.0 (2018-12-13)
+------------------
+
+**Improvements**
+
+* [#207](https://github.com/socialcopsdev/camelot/issues/207) Add a plot type for Stream text edges and detected table areas. [#224](https://github.com/socialcopsdev/camelot/pull/224) by Vinayak Mehta.
+* [#204](https://github.com/socialcopsdev/camelot/issues/204) `suppress_warnings` is now called `suppress_stdout`. [#225](https://github.com/socialcopsdev/camelot/pull/225) by Vinayak Mehta.
+
+**Bugfixes**
+
+* [#217](https://github.com/socialcopsdev/camelot/issues/217) Fix IndexError when scale is large.
+* [#105](https://github.com/socialcopsdev/camelot/issues/105), [#192](https://github.com/socialcopsdev/camelot/issues/192) and [#215](https://github.com/socialcopsdev/camelot/issues/215) in [#227](https://github.com/socialcopsdev/camelot/pull/227) by Vinayak Mehta.
+
+**Documentation**
+
+* Add pdfplumber comparison and update Tabula (stream) comparison. Check out the [wiki page](https://github.com/socialcopsdev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools).
+
+0.4.1 (2018-12-05)
+------------------
+
+**Bugfixes**
+
+* Add chardet to `install_requires` to fix [#210](https://github.com/socialcopsdev/camelot/issues/210). More details in [pdfminer.six#213](https://github.com/pdfminer/pdfminer.six/issues/213).
+
+0.4.0 (2018-11-23)
+------------------
+
+**Improvements**
+
+* [#102](https://github.com/socialcopsdev/camelot/issues/102) Detect tables automatically when Stream is used. [#206](https://github.com/socialcopsdev/camelot/pull/206) Add implementation of Anssi Nurminen's table detection algorithm by Vinayak Mehta.
+
+0.3.2 (2018-11-04)
+------------------
+
+**Improvements**
+
+* [#186](https://github.com/socialcopsdev/camelot/issues/186) Add `_bbox` attribute to table. [#193](https://github.com/socialcopsdev/camelot/pull/193) by Vinayak Mehta.
+    * You can use `table._bbox` to get coordinates of the detected table.
+
+0.3.1 (2018-11-02)
+------------------
+
+**Improvements**
+
+* Matplotlib is now an optional requirement. [#190](https://github.com/socialcopsdev/camelot/pull/190) by Vinayak Mehta.
+    * You can install it using `$ pip install camelot-py[plot]`.
+* [#127](https://github.com/socialcopsdev/camelot/issues/127) Add tests for plotting. Coverage is now at 87%! [#179](https://github.com/socialcopsdev/camelot/pull/179) by [Suyash Behera](https://github.com/Suyash458).
+
+0.3.0 (2018-10-28)
+------------------
+
+**Improvements**
+
+* [#162](https://github.com/socialcopsdev/camelot/issues/162) Add password keyword argument. [#180](https://github.com/socialcopsdev/camelot/pull/180) by [rbares](https://github.com/rbares).
+    * An encrypted PDF can now be decrypted by passing `password='<PASSWORD>'` to `read_pdf` or `--password <PASSWORD>` to the command-line interface. (Limited encryption algorithm support from PyPDF2.)
+* [#139](https://github.com/socialcopsdev/camelot/issues/139) Add suppress_warnings keyword argument. [#155](https://github.com/socialcopsdev/camelot/pull/155) by [Jonathan Lloyd](https://github.com/jonathanlloyd).
+    * Warnings raised by Camelot can now be suppressed by passing `suppress_warnings=True` to `read_pdf` or `--quiet` to the command-line interface.
+* [#154](https://github.com/socialcopsdev/camelot/issues/154) The CLI can now be run using `python -m`. Try `python -m camelot --help`. [#159](https://github.com/socialcopsdev/camelot/pull/159) by [Parth P Panchal](https://github.com/pqrth).
+* [#165](https://github.com/socialcopsdev/camelot/issues/165) Rename `table_area` to `table_areas`. [#171](https://github.com/socialcopsdev/camelot/pull/171) by [Parth P Panchal](https://github.com/pqrth).
+
+**Bugfixes**
+
+* Raise error if the ghostscript executable is not on the PATH variable. [#166](https://github.com/socialcopsdev/camelot/pull/166) by Vinayak Mehta.
+* Convert filename to lowercase to check for PDF extension. [#169](https://github.com/socialcopsdev/camelot/pull/169) by [Vinicius Mesel](https://github.com/vmesel).
+
+**Files**
+
+* [#114](https://github.com/socialcopsdev/camelot/issues/114) Add Makefile and make codecov run only once. [#132](https://github.com/socialcopsdev/camelot/pull/132) by [Vaibhav Mule](https://github.com/vaibhavmule).
+* Add .editorconfig. [#151](https://github.com/socialcopsdev/camelot/pull/151) by [KOLANICH](https://github.com/KOLANICH).
+* Downgrade numpy version from 1.15.2 to 1.13.3.
+* Add requirements.txt for readthedocs.
+
+**Documentation**
+
+* Add "Using conda" section to installation instructions.
+* Add readthedocs badge.
+
+0.2.3 (2018-10-08)
+------------------
+
+* Remove hard dependencies on requirements versions.
+
+0.2.2 (2018-10-08)
+------------------
+
+**Bugfixes**
+
+* Move opencv-python to extra\_requires. [#134](https://github.com/socialcopsdev/camelot/pull/134) by Vinayak Mehta.
+
 0.2.1 (2018-10-05)
 ------------------
 
@@ -51,4 +278,4 @@ Release History
 0.1.0 (2018-09-24)
 ------------------
 
-* Birth!
+* Rebirth!

LICENSE

@@ -1,6 +1,7 @@
 MIT License
 
-Copyright (c) 2018 Peeply Private Ltd (Singapore)
+Copyright (c) 2019-2021 Camelot Developers
+Copyright (c) 2018-2019 Peeply Private Ltd (Singapore)
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

MANIFEST.in

@@ -1 +1 @@
-include MANIFEST.in README.md HISTORY.md LICENSE requirements.txt requirements-dev.txt setup.py setup.cfg
+include MANIFEST.in README.md HISTORY.md LICENSE setup.py setup.cfg

Makefile

@@ -0,0 +1,28 @@
+.PHONY: docs
+INSTALL :=
+UNAME_S := $(shell uname -s)
+ifeq ($(UNAME_S),Linux)
+	INSTALL := @sudo apt install python-tk python3-tk ghostscript
+else ifeq ($(UNAME_S),Darwin)
+	INSTALL := @brew install tcl-tk ghostscript
+else
+	INSTALL := @echo "Please install tk and ghostscript"
+endif
+
+install:
+	$(INSTALL)
+	pip install --upgrade pip
+	pip install ".[dev]"
+
+test:
+	pytest --verbose --cov-config .coveragerc --cov-report term --cov-report xml --cov=camelot --mpl
+
+docs:
+	cd docs && make html
+	@echo "\033[95m\n\nBuild successful! View the docs homepage at docs/_build/html/index.html.\n\033[0m"
+
+publish:
+	pip install twine
+	python setup.py sdist
+	twine upload dist/*
+	rm -fr build dist .egg camelot_py.egg-info

README.md

@@ -1,24 +1,28 @@
 <p align="center">
-   <img src="https://raw.githubusercontent.com/socialcopsdev/camelot/master/docs/_static/camelot.png" width="200">
+   <img src="https://raw.githubusercontent.com/camelot-dev/camelot/master/docs/_static/camelot.png" width="200">
 </p>
 
 # Camelot: PDF Table Extraction for Humans
 
-[![Build Status](https://travis-ci.org/socialcopsdev/camelot.svg?branch=master)](https://travis-ci.org/socialcopsdev/camelot) [![codecov.io](https://codecov.io/github/socialcopsdev/camelot/badge.svg?branch=master&service=github)](https://codecov.io/github/socialcopsdev/camelot?branch=master)
+[![tests](https://github.com/camelot-dev/camelot/actions/workflows/tests.yml/badge.svg)](https://github.com/camelot-dev/camelot/actions/workflows/tests.yml) [![Documentation Status](https://readthedocs.org/projects/camelot-py/badge/?version=master)](https://camelot-py.readthedocs.io/en/master/)
- [![image](https://img.shields.io/pypi/v/camelot-py.svg)](https://pypi.org/project/camelot-py/) [![image](https://img.shields.io/pypi/l/camelot-py.svg)](https://pypi.org/project/camelot-py/) [![image](https://img.shields.io/pypi/pyversions/camelot-py.svg)](https://pypi.org/project/camelot-py/)
+ [![codecov.io](https://codecov.io/github/camelot-dev/camelot/badge.svg?branch=master&service=github)](https://codecov.io/github/camelot-dev/camelot?branch=master)
+ [![image](https://img.shields.io/pypi/v/camelot-py.svg)](https://pypi.org/project/camelot-py/) [![image](https://img.shields.io/pypi/l/camelot-py.svg)](https://pypi.org/project/camelot-py/) [![image](https://img.shields.io/pypi/pyversions/camelot-py.svg)](https://pypi.org/project/camelot-py/) [![Gitter chat](https://badges.gitter.im/camelot-dev/Lobby.png)](https://gitter.im/camelot-dev/Lobby)
+[![image](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/ambv/black)
 
-**Camelot** is a Python library that makes it easy for *anyone* to extract tables from PDF files!
+**Camelot** is a Python library that can help you extract tables from PDFs!
+
+**Note:** You can also check out [Excalibur](https://github.com/camelot-dev/excalibur), the web interface to Camelot!
 
 ---
 
-**Here's how you can extract tables from PDF files.** Check out the PDF used in this example [here](https://github.com/socialcopsdev/camelot/blob/master/docs/_static/pdf/foo.pdf).
+**Here's how you can extract tables from PDFs.** You can check out the PDF used in this example [here](https://github.com/camelot-dev/camelot/blob/master/docs/_static/pdf/foo.pdf).
 
 <pre>
 >>> import camelot
 >>> tables = camelot.read_pdf('foo.pdf')
 >>> tables
 &lt;TableList n=1&gt;
->>> tables.export('foo.csv', f='csv', compress=True) # json, excel, html
+>>> tables.export('foo.csv', f='csv', compress=True) # json, excel, html, markdown, sqlite
 >>> tables[0]
 &lt;Table shape=(7, 7)&gt;
 >>> tables[0].parsing_report
@@ -28,7 +32,7 @@
     'order': 1,
     'page': 1
 }
->>> tables[0].to_csv('foo.csv') # to_json, to_excel, to_html
+>>> tables[0].to_csv('foo.csv') # to_json, to_excel, to_html, to_markdown, to_sqlite
 >>> tables[0].df # get a pandas DataFrame!
 </pre>
 
@@ -41,80 +45,73 @@
 | 2032_2     | 0.17      | 57.8          | 21.7%                | 0.3%            | 2.7%            | 1.2%           |
 | 4171_1     | 0.07      | 173.9         | 58.1%                | 1.6%            | 2.1%            | 0.5%           |
 
-There's a [command-line interface](https://camelot-py.readthedocs.io/en/latest/user/cli.html) too!
+Camelot also comes packaged with a [command-line interface](https://camelot-py.readthedocs.io/en/master/user/cli.html)!
 
 **Note:** Camelot only works with text-based PDFs and not scanned documents. (As Tabula [explains](https://github.com/tabulapdf/tabula#why-tabula), "If you can click and drag to select text in your table in a PDF viewer, then your PDF is text-based".)
 
+You can check out some frequently asked questions [here](https://camelot-py.readthedocs.io/en/master/user/faq.html).
+
 ## Why Camelot?
 
-- **You are in control.**: Unlike other libraries and tools which either give a nice output or fail miserably (with no in-between), Camelot gives you the power to tweak table extraction. (This is important since everything in the real world, including PDF table extraction, is fuzzy.)
+- **Configurability**: Camelot gives you control over the table extraction process with [tweakable settings](https://camelot-py.readthedocs.io/en/master/user/advanced.html).
-- *Bad* tables can be discarded based on **metrics** like accuracy and whitespace, without ever having to manually look at each table.
+- **Metrics**: You can discard bad tables based on metrics like accuracy and whitespace, without having to manually look at each table.
-- Each table is a **pandas DataFrame**, which seamlessly integrates into [ETL and data analysis workflows](https://gist.github.com/vinayak-mehta/e5949f7c2410a0e12f25d3682dc9e873).
+- **Output**: Each table is extracted into a **pandas DataFrame**, which seamlessly integrates into [ETL and data analysis workflows](https://gist.github.com/vinayak-mehta/e5949f7c2410a0e12f25d3682dc9e873). You can also export tables to multiple formats, which include CSV, JSON, Excel, HTML, Markdown, and Sqlite.
-- **Export** to multiple formats, including JSON, Excel and HTML.
 
-See [comparison with other PDF table extraction libraries and tools](https://github.com/socialcopsdev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools).
+See [comparison with similar libraries and tools](https://github.com/camelot-dev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools).
+
+## Support the development
+
+If Camelot has helped you, please consider supporting its development with a one-time or monthly donation [on OpenCollective](https://opencollective.com/camelot).
 
 ## Installation
 
-After [installing the dependencies](https://camelot-py.readthedocs.io/en/latest/user/install.html) ([tk](https://packages.ubuntu.com/trusty/python-tk) and [ghostscript](https://www.ghostscript.com/)), you can simply use pip to install Camelot:
+### Using conda
+
+The easiest way to install Camelot is with [conda](https://conda.io/docs/), which is a package manager and environment management system for the [Anaconda](http://docs.continuum.io/anaconda/) distribution.
 
 <pre>
-$ pip install camelot-py
+$ conda install -c conda-forge camelot-py
 </pre>
 
-### Alternatively
+### Using pip
 
-After [installing the dependencies](https://camelot-py.readthedocs.io/en/latest/user/install.html), clone the repo using:
+After [installing the dependencies](https://camelot-py.readthedocs.io/en/master/user/install-deps.html) ([tk](https://packages.ubuntu.com/bionic/python/python-tk) and [ghostscript](https://www.ghostscript.com/)), you can also just use pip to install Camelot:
 
 <pre>
-$ git clone https://www.github.com/socialcopsdev/camelot
+$ pip install "camelot-py[base]"
+</pre>
+
+### From the source code
+
+After [installing the dependencies](https://camelot-py.readthedocs.io/en/master/user/install.html#using-pip), clone the repo using:
+
+<pre>
+$ git clone https://www.github.com/camelot-dev/camelot
 </pre>
 
 and install Camelot using pip:
 
 <pre>
 $ cd camelot
-$ pip install .
+$ pip install ".[base]"
 </pre>
 
-**Note:** Use a [virtualenv](https://virtualenv.pypa.io/en/stable/) if you don't want to affect your global Python installation.
-
 ## Documentation
 
-Great documentation is available at [http://camelot-py.readthedocs.io/](http://camelot-py.readthedocs.io/).
+The documentation is available at [http://camelot-py.readthedocs.io/](http://camelot-py.readthedocs.io/).
 
-## Development
+## Wrappers
 
-The [Contributor's Guide](https://camelot-py.readthedocs.io/en/latest/dev/contributing.html) has detailed information about contributing code, documentation, tests and more. We've included some basic information in this README.
+- [camelot-php](https://github.com/randomstate/camelot-php) provides a [PHP](https://www.php.net/) wrapper on Camelot.
 
-### Source code
+## Contributing
 
-You can check the latest sources with:
+The [Contributor's Guide](https://camelot-py.readthedocs.io/en/master/dev/contributing.html) has detailed information about contributing issues, documentation, code, and tests.
-
-<pre>
-$ git clone https://www.github.com/socialcopsdev/camelot
-</pre>
-
-### Setting up a development environment
-
-You can install the development dependencies easily, using pip:
-
-<pre>
-$ pip install camelot-py[dev]
-</pre>
-
-### Testing
-
-After installation, you can run tests using:
-
-<pre>
-$ python setup.py test
-</pre>
 
 ## Versioning
 
-Camelot uses [Semantic Versioning](https://semver.org/). For the available versions, see the tags on this repository. For the changelog, you can check out [HISTORY.md](https://github.com/socialcopsdev/camelot/blob/master/HISTORY.md).
+Camelot uses [Semantic Versioning](https://semver.org/). For the available versions, see the tags on this repository. For the changelog, you can check out [HISTORY.md](https://github.com/camelot-dev/camelot/blob/master/HISTORY.md).
 
 ## License
 
-This project is licensed under the MIT License, see the [LICENSE](https://github.com/socialcopsdev/camelot/blob/master/LICENSE) file for details.
+This project is licensed under the MIT License, see the [LICENSE](https://github.com/camelot-dev/camelot/blob/master/LICENSE) file for details.

camelot/__init__.py

@@ -4,14 +4,18 @@ import logging
 
 from .__version__ import __version__
 from .io import read_pdf
+from .plotting import PlotMethods
 
 
 # set up logging
-logger = logging.getLogger('camelot')
+logger = logging.getLogger("camelot")
 
-format_string = '%(asctime)s - %(levelname)s - %(message)s'
+format_string = "%(asctime)s - %(levelname)s - %(message)s"
-formatter = logging.Formatter(format_string, datefmt='%Y-%m-%dT%H:%M:%S')
+formatter = logging.Formatter(format_string, datefmt="%Y-%m-%dT%H:%M:%S")
 handler = logging.StreamHandler()
 handler.setFormatter(formatter)
 
 logger.addHandler(handler)
+
+# instantiate plot method
+plot = PlotMethods()

camelot/__main__.py

@@ -0,0 +1,14 @@
+# -*- coding: utf-8 -*-
+
+
+__all__ = ("main",)
+
+
+def main():
+    from camelot.cli import cli
+
+    cli()
+
+
+if __name__ == "__main__":
+    main()

camelot/__version__.py

@@ -1,11 +1,23 @@
 # -*- coding: utf-8 -*-
 
-VERSION = (0, 2, 1)
+VERSION = (0, 10, 1)
+PRERELEASE = None  # alpha, beta or rc
+REVISION = None
 
-__title__ = 'camelot-py'
+
-__description__ = 'PDF Table Extraction for Humans.'
+def generate_version(version, prerelease=None, revision=None):
-__url__ = 'http://camelot-py.readthedocs.io/'
+    version_parts = [".".join(map(str, version))]
-__version__ = '.'.join(map(str, VERSION))
+    if prerelease is not None:
-__author__ = 'Vinayak Mehta'
+        version_parts.append(f"-{prerelease}")
-__author_email__ = 'vmehta94@gmail.com'
+    if revision is not None:
-__license__ = 'MIT License'
+        version_parts.append(f".{revision}")
+    return "".join(version_parts)
+
+
+__title__ = "camelot-py"
+__description__ = "PDF Table Extraction for Humans."
+__url__ = "http://camelot-py.readthedocs.io/"
+__version__ = generate_version(VERSION, prerelease=PRERELEASE, revision=REVISION)
+__author__ = "Vinayak Mehta"
+__author_email__ = "vmehta94@gmail.com"
+__license__ = "MIT License"

camelot/backends/__init__.py

@@ -0,0 +1,3 @@
+# -*- coding: utf-8 -*-
+
+from .image_conversion import ImageConversionBackend

camelot/backends/ghostscript_backend.py

@@ -0,0 +1,47 @@
+# -*- coding: utf-8 -*-
+
+import sys
+import ctypes
+from ctypes.util import find_library
+
+
+def installed_posix():
+    library = find_library("gs")
+    return library is not None
+
+
+def installed_windows():
+    library = find_library(
+        "".join(("gsdll", str(ctypes.sizeof(ctypes.c_voidp) * 8), ".dll"))
+    )
+    return library is not None
+
+
+class GhostscriptBackend(object):
+    def installed(self):
+        if sys.platform in ["linux", "darwin"]:
+            return installed_posix()
+        elif sys.platform == "win32":
+            return installed_windows()
+        else:
+            return installed_posix()
+
+    def convert(self, pdf_path, png_path, resolution=300):
+        if not self.installed():
+            raise OSError(
+                "Ghostscript is not installed. You can install it using the instructions"
+                " here: https://camelot-py.readthedocs.io/en/master/user/install-deps.html"
+            )
+
+        import ghostscript
+
+        gs_command = [
+            "gs",
+            "-q",
+            "-sDEVICE=png16m",
+            "-o",
+            png_path,
+            f"-r{resolution}",
+            pdf_path,
+        ]
+        ghostscript.Ghostscript(*gs_command)

camelot/backends/image_conversion.py

@@ -0,0 +1,40 @@
+# -*- coding: utf-8 -*-
+
+from .poppler_backend import PopplerBackend
+from .ghostscript_backend import GhostscriptBackend
+
+BACKENDS = {"poppler": PopplerBackend, "ghostscript": GhostscriptBackend}
+
+
+class ImageConversionBackend(object):
+    def __init__(self, backend="poppler", use_fallback=True):
+        if backend not in BACKENDS.keys():
+            raise ValueError(f"Image conversion backend '{backend}' not supported")
+
+        self.backend = backend
+        self.use_fallback = use_fallback
+        self.fallbacks = list(filter(lambda x: x != backend, BACKENDS.keys()))
+
+    def convert(self, pdf_path, png_path):
+        try:
+            converter = BACKENDS[self.backend]()
+            converter.convert(pdf_path, png_path)
+        except Exception as e:
+            import sys
+
+            if self.use_fallback:
+                for fallback in self.fallbacks:
+                    try:
+                        converter = BACKENDS[fallback]()
+                        converter.convert(pdf_path, png_path)
+                    except Exception as e:
+                        raise type(e)(
+                            str(e) + f" with image conversion backend '{fallback}'"
+                        ).with_traceback(sys.exc_info()[2])
+                        continue
+                    else:
+                        break
+            else:
+                raise type(e)(
+                    str(e) + f" with image conversion backend '{self.backend}'"
+                ).with_traceback(sys.exc_info()[2])

camelot/backends/poppler_backend.py

@@ -0,0 +1,22 @@
+# -*- coding: utf-8 -*-
+
+import shutil
+import subprocess
+
+
+class PopplerBackend(object):
+    def convert(self, pdf_path, png_path):
+        pdftopng_executable = shutil.which("pdftopng")
+        if pdftopng_executable is None:
+            raise OSError(
+                "pdftopng is not installed. You can install it using the 'pip install pdftopng' command."
+            )
+
+        pdftopng_command = [pdftopng_executable, pdf_path, png_path]
+
+        try:
+            subprocess.check_output(
+                " ".join(pdftopng_command), stderr=subprocess.STDOUT, shell=True
+            )
+        except subprocess.CalledProcessError as e:
+            raise ValueError(e.output)

camelot/cli.py

@@ -4,11 +4,17 @@ import logging
 
 import click
 
-from . import __version__
+try:
-from .io import read_pdf
+    import matplotlib.pyplot as plt
+except ImportError:
+    _HAS_MPL = False
+else:
+    _HAS_MPL = True
+
+from . import __version__, read_pdf, plot
 
 
-logger = logging.getLogger('camelot')
+logger = logging.getLogger("camelot")
 logger.setLevel(logging.INFO)
 
 
@@ -23,21 +29,49 @@ class Config(object):
 pass_config = click.make_pass_decorator(Config)
 
 
-@click.group()
+@click.group(name="camelot")
 @click.version_option(version=__version__)
-@click.option('-p', '--pages', default='1', help='Comma-separated page numbers.'
+@click.option("-q", "--quiet", is_flag=False, help="Suppress logs and warnings.")
-              ' Example: 1,3,4 or 1,4-end.')
+@click.option(
-@click.option('-o', '--output', help='Output file path.')
+    "-p",
-@click.option('-f', '--format',
+    "--pages",
-              type=click.Choice(['csv', 'json', 'excel', 'html']),
+    default="1",
-              help='Output file format.')
+    help="Comma-separated page numbers." " Example: 1,3,4 or 1,4-end or all.",
-@click.option('-z', '--zip', is_flag=True, help='Create ZIP archive.')
+)
-@click.option('-split', '--split_text', is_flag=True,
+@click.option("-pw", "--password", help="Password for decryption.")
-              help='Split text that spans across multiple cells.')
+@click.option("-o", "--output", help="Output file path.")
-@click.option('-flag', '--flag_size', is_flag=True, help='Flag text based on'
+@click.option(
-              ' font size. Useful to detect super/subscripts.')
+    "-f",
-@click.option('-M', '--margins', nargs=3, default=(1.0, 0.5, 0.1),
+    "--format",
-              help='PDFMiner char_margin, line_margin and word_margin.')
+    type=click.Choice(["csv", "excel", "html", "json", "markdown", "sqlite"]),
+    help="Output file format.",
+)
+@click.option("-z", "--zip", is_flag=True, help="Create ZIP archive.")
+@click.option(
+    "-split",
+    "--split_text",
+    is_flag=True,
+    help="Split text that spans across multiple cells.",
+)
+@click.option(
+    "-flag",
+    "--flag_size",
+    is_flag=True,
+    help="Flag text based on" " font size. Useful to detect super/subscripts.",
+)
+@click.option(
+    "-strip",
+    "--strip_text",
+    help="Characters that should be stripped from a string before"
+    " assigning it to a cell.",
+)
+@click.option(
+    "-M",
+    "--margins",
+    nargs=3,
+    default=(1.0, 0.5, 0.1),
+    help="PDFMiner char_margin, line_margin and word_margin.",
+)
 @click.pass_context
 def cli(ctx, *args, **kwargs):
     """Camelot: PDF Table Extraction for Humans"""
@@ -46,111 +80,225 @@ def cli(ctx, *args, **kwargs):
         ctx.obj.set_config(key, value)
 
 
-@cli.command('lattice')
+@cli.command("lattice")
-@click.option('-T', '--table_area', default=[], multiple=True,
+@click.option(
-              help='Table areas to process. Example: x1,y1,x2,y2'
+    "-R",
-              ' where x1, y1 -> left-top and x2, y2 -> right-bottom.')
+    "--table_regions",
-@click.option('-back', '--process_background', is_flag=True,
+    default=[],
-              help='Process background lines.')
+    multiple=True,
-@click.option('-scale', '--line_size_scaling', default=15,
+    help="Page regions to analyze. Example: x1,y1,x2,y2"
-              help='Line size scaling factor. The larger the value,'
+    " where x1, y1 -> left-top and x2, y2 -> right-bottom.",
-              ' the smaller the detected lines.')
+)
-@click.option('-copy', '--copy_text', default=[], type=click.Choice(['h', 'v']),
+@click.option(
-              multiple=True, help='Direction in which text in a spanning cell'
+    "-T",
-              ' will be copied over.')
+    "--table_areas",
-@click.option('-shift', '--shift_text', default=['l', 't'],
+    default=[],
-              type=click.Choice(['', 'l', 'r', 't', 'b']), multiple=True,
+    multiple=True,
-              help='Direction in which text in a spanning cell will flow.')
+    help="Table areas to process. Example: x1,y1,x2,y2"
-@click.option('-l', '--line_close_tol', default=2,
+    " where x1, y1 -> left-top and x2, y2 -> right-bottom.",
-              help='Tolerance parameter used to merge close vertical'
+)
-              ' and horizontal lines.')
+@click.option(
-@click.option('-j', '--joint_close_tol', default=2,
+    "-back", "--process_background", is_flag=True, help="Process background lines."
-              help='Tolerance parameter used to decide whether'
+)
-              ' the detected lines and points lie close to each other.')
+@click.option(
-@click.option('-block', '--threshold_blocksize', default=15,
+    "-scale",
-              help='For adaptive thresholding, size of a pixel'
+    "--line_scale",
-              ' neighborhood that is used to calculate a threshold value for'
+    default=15,
-              ' the pixel. Example: 3, 5, 7, and so on.')
+    help="Line size scaling factor. The larger the value,"
-@click.option('-const', '--threshold_constant', default=-2,
+    " the smaller the detected lines.",
-              help='For adaptive thresholding, constant subtracted'
+)
-              ' from the mean or weighted mean. Normally, it is positive but'
+@click.option(
-              ' may be zero or negative as well.')
+    "-copy",
-@click.option('-I', '--iterations', default=0,
+    "--copy_text",
-              help='Number of times for erosion/dilation will be applied.')
+    default=[],
-@click.option('-plot', '--plot_type',
+    type=click.Choice(["h", "v"]),
-              type=click.Choice(['text', 'table', 'contour', 'joint', 'line']),
+    multiple=True,
-              help='Plot geometry found on PDF page, for debugging.')
+    help="Direction in which text in a spanning cell" " will be copied over.",
-@click.argument('filepath', type=click.Path(exists=True))
+)
+@click.option(
+    "-shift",
+    "--shift_text",
+    default=["l", "t"],
+    type=click.Choice(["", "l", "r", "t", "b"]),
+    multiple=True,
+    help="Direction in which text in a spanning cell will flow.",
+)
+@click.option(
+    "-l",
+    "--line_tol",
+    default=2,
+    help="Tolerance parameter used to merge close vertical" " and horizontal lines.",
+)
+@click.option(
+    "-j",
+    "--joint_tol",
+    default=2,
+    help="Tolerance parameter used to decide whether"
+    " the detected lines and points lie close to each other.",
+)
+@click.option(
+    "-block",
+    "--threshold_blocksize",
+    default=15,
+    help="For adaptive thresholding, size of a pixel"
+    " neighborhood that is used to calculate a threshold value for"
+    " the pixel. Example: 3, 5, 7, and so on.",
+)
+@click.option(
+    "-const",
+    "--threshold_constant",
+    default=-2,
+    help="For adaptive thresholding, constant subtracted"
+    " from the mean or weighted mean. Normally, it is positive but"
+    " may be zero or negative as well.",
+)
+@click.option(
+    "-I",
+    "--iterations",
+    default=0,
+    help="Number of times for erosion/dilation will be applied.",
+)
+@click.option(
+    "-res",
+    "--resolution",
+    default=300,
+    help="Resolution used for PDF to PNG conversion.",
+)
+@click.option(
+    "-plot",
+    "--plot_type",
+    type=click.Choice(["text", "grid", "contour", "joint", "line"]),
+    help="Plot elements found on PDF page for visual debugging.",
+)
+@click.argument("filepath", type=click.Path(exists=True))
 @pass_config
 def lattice(c, *args, **kwargs):
     """Use lines between text to parse the table."""
     conf = c.config
-    pages = conf.pop('pages')
+    pages = conf.pop("pages")
-    output = conf.pop('output')
+    output = conf.pop("output")
-    f = conf.pop('format')
+    f = conf.pop("format")
-    compress = conf.pop('zip')
+    compress = conf.pop("zip")
-    plot_type = kwargs.pop('plot_type')
+    quiet = conf.pop("quiet")
-    filepath = kwargs.pop('filepath')
+    plot_type = kwargs.pop("plot_type")
+    filepath = kwargs.pop("filepath")
     kwargs.update(conf)
 
-    table_area = list(kwargs['table_area'])
+    table_regions = list(kwargs["table_regions"])
-    kwargs['table_area'] = None if not table_area else table_area
+    kwargs["table_regions"] = None if not table_regions else table_regions
-    copy_text = list(kwargs['copy_text'])
+    table_areas = list(kwargs["table_areas"])
-    kwargs['copy_text'] = None if not copy_text else copy_text
+    kwargs["table_areas"] = None if not table_areas else table_areas
-    kwargs['shift_text'] = list(kwargs['shift_text'])
+    copy_text = list(kwargs["copy_text"])
+    kwargs["copy_text"] = None if not copy_text else copy_text
+    kwargs["shift_text"] = list(kwargs["shift_text"])
 
-    tables = read_pdf(filepath, pages=pages, flavor='lattice', **kwargs)
-    click.echo('Found {} tables'.format(tables.n))
     if plot_type is not None:
-        for table in tables:
+        if not _HAS_MPL:
-            table.plot(plot_type)
+            raise ImportError("matplotlib is required for plotting.")
     else:
         if output is None:
-            raise click.UsageError('Please specify output file path using --output')
+            raise click.UsageError("Please specify output file path using --output")
         if f is None:
-            raise click.UsageError('Please specify output file format using --format')
+            raise click.UsageError("Please specify output file format using --format")
+
+    tables = read_pdf(
+        filepath, pages=pages, flavor="lattice", suppress_stdout=quiet, **kwargs
+    )
+    click.echo(f"Found {tables.n} tables")
+    if plot_type is not None:
+        for table in tables:
+            plot(table, kind=plot_type)
+            plt.show()
+    else:
         tables.export(output, f=f, compress=compress)
 
 
-@cli.command('stream')
+@cli.command("stream")
-@click.option('-T', '--table_area', default=[], multiple=True,
+@click.option(
-              help='Table areas to process. Example: x1,y1,x2,y2'
+    "-R",
-              ' where x1, y1 -> left-top and x2, y2 -> right-bottom.')
+    "--table_regions",
-@click.option('-C', '--columns', default=[], multiple=True,
+    default=[],
-              help='X coordinates of column separators.')
+    multiple=True,
-@click.option('-r', '--row_close_tol', default=2, help='Tolerance parameter'
+    help="Page regions to analyze. Example: x1,y1,x2,y2"
-              ' used to combine text vertically, to generate rows.')
+    " where x1, y1 -> left-top and x2, y2 -> right-bottom.",
-@click.option('-c', '--col_close_tol', default=0, help='Tolerance parameter'
+)
-              ' used to combine text horizontally, to generate columns.')
+@click.option(
-@click.option('-plot', '--plot_type',
+    "-T",
-              type=click.Choice(['text', 'table']),
+    "--table_areas",
-              help='Plot geometry found on PDF page for debugging.')
+    default=[],
-@click.argument('filepath', type=click.Path(exists=True))
+    multiple=True,
+    help="Table areas to process. Example: x1,y1,x2,y2"
+    " where x1, y1 -> left-top and x2, y2 -> right-bottom.",
+)
+@click.option(
+    "-C",
+    "--columns",
+    default=[],
+    multiple=True,
+    help="X coordinates of column separators.",
+)
+@click.option(
+    "-e",
+    "--edge_tol",
+    default=50,
+    help="Tolerance parameter" " for extending textedges vertically.",
+)
+@click.option(
+    "-r",
+    "--row_tol",
+    default=2,
+    help="Tolerance parameter" " used to combine text vertically, to generate rows.",
+)
+@click.option(
+    "-c",
+    "--column_tol",
+    default=0,
+    help="Tolerance parameter"
+    " used to combine text horizontally, to generate columns.",
+)
+@click.option(
+    "-plot",
+    "--plot_type",
+    type=click.Choice(["text", "grid", "contour", "textedge"]),
+    help="Plot elements found on PDF page for visual debugging.",
+)
+@click.argument("filepath", type=click.Path(exists=True))
 @pass_config
 def stream(c, *args, **kwargs):
     """Use spaces between text to parse the table."""
     conf = c.config
-    pages = conf.pop('pages')
+    pages = conf.pop("pages")
-    output = conf.pop('output')
+    output = conf.pop("output")
-    f = conf.pop('format')
+    f = conf.pop("format")
-    compress = conf.pop('zip')
+    compress = conf.pop("zip")
-    plot_type = kwargs.pop('plot_type')
+    quiet = conf.pop("quiet")
-    filepath = kwargs.pop('filepath')
+    plot_type = kwargs.pop("plot_type")
+    filepath = kwargs.pop("filepath")
     kwargs.update(conf)
 
-    table_area = list(kwargs['table_area'])
+    table_regions = list(kwargs["table_regions"])
-    kwargs['table_area'] = None if not table_area else table_area
+    kwargs["table_regions"] = None if not table_regions else table_regions
-    columns = list(kwargs['columns'])
+    table_areas = list(kwargs["table_areas"])
-    kwargs['columns'] = None if not columns else columns
+    kwargs["table_areas"] = None if not table_areas else table_areas
+    columns = list(kwargs["columns"])
+    kwargs["columns"] = None if not columns else columns
 
-    tables = read_pdf(filepath, pages=pages, flavor='stream', **kwargs)
-    click.echo('Found {} tables'.format(tables.n))
     if plot_type is not None:
-        for table in tables:
+        if not _HAS_MPL:
-            table.plot(plot_type)
+            raise ImportError("matplotlib is required for plotting.")
     else:
         if output is None:
-            raise click.UsageError('Please specify output file path using --output')
+            raise click.UsageError("Please specify output file path using --output")
         if f is None:
-            raise click.UsageError('Please specify output file format using --format')
+            raise click.UsageError("Please specify output file format using --format")
+
+    tables = read_pdf(
+        filepath, pages=pages, flavor="stream", suppress_stdout=quiet, **kwargs
+    )
+    click.echo(f"Found {tables.n} tables")
+    if plot_type is not None:
+        for table in tables:
+            plot(table, kind=plot_type)
+            plt.show()
+    else:
         tables.export(output, f=f, compress=compress)

camelot/core.py

@@ -1,13 +1,231 @@
 # -*- coding: utf-8 -*-
 
 import os
+import sqlite3
 import zipfile
 import tempfile
+from itertools import chain
+from operator import itemgetter
 
 import numpy as np
 import pandas as pd
 
-from .plotting import *
+
+# minimum number of vertical textline intersections for a textedge
+# to be considered valid
+TEXTEDGE_REQUIRED_ELEMENTS = 4
+# padding added to table area on the left, right and bottom
+TABLE_AREA_PADDING = 10
+
+
+class TextEdge(object):
+    """Defines a text edge coordinates relative to a left-bottom
+    origin. (PDF coordinate space)
+
+    Parameters
+    ----------
+    x : float
+        x-coordinate of the text edge.
+    y0 : float
+        y-coordinate of bottommost point.
+    y1 : float
+        y-coordinate of topmost point.
+    align : string, optional (default: 'left')
+        {'left', 'right', 'middle'}
+
+    Attributes
+    ----------
+    intersections: int
+        Number of intersections with horizontal text rows.
+    is_valid: bool
+        A text edge is valid if it intersections with at least
+        TEXTEDGE_REQUIRED_ELEMENTS horizontal text rows.
+
+    """
+
+    def __init__(self, x, y0, y1, align="left"):
+        self.x = x
+        self.y0 = y0
+        self.y1 = y1
+        self.align = align
+        self.intersections = 0
+        self.is_valid = False
+
+    def __repr__(self):
+        x = round(self.x, 2)
+        y0 = round(self.y0, 2)
+        y1 = round(self.y1, 2)
+        return (
+            f"<TextEdge x={x} y0={y0} y1={y1} align={self.align} valid={self.is_valid}>"
+        )
+
+    def update_coords(self, x, y0, edge_tol=50):
+        """Updates the text edge's x and bottom y coordinates and sets
+        the is_valid attribute.
+        """
+        if np.isclose(self.y0, y0, atol=edge_tol):
+            self.x = (self.intersections * self.x + x) / float(self.intersections + 1)
+            self.y0 = y0
+            self.intersections += 1
+            # a textedge is valid only if it extends uninterrupted
+            # over a required number of textlines
+            if self.intersections > TEXTEDGE_REQUIRED_ELEMENTS:
+                self.is_valid = True
+
+
+class TextEdges(object):
+    """Defines a dict of left, right and middle text edges found on
+    the PDF page. The dict has three keys based on the alignments,
+    and each key's value is a list of camelot.core.TextEdge objects.
+    """
+
+    def __init__(self, edge_tol=50):
+        self.edge_tol = edge_tol
+        self._textedges = {"left": [], "right": [], "middle": []}
+
+    @staticmethod
+    def get_x_coord(textline, align):
+        """Returns the x coordinate of a text row based on the
+        specified alignment.
+        """
+        x_left = textline.x0
+        x_right = textline.x1
+        x_middle = x_left + (x_right - x_left) / 2.0
+        x_coord = {"left": x_left, "middle": x_middle, "right": x_right}
+        return x_coord[align]
+
+    def find(self, x_coord, align):
+        """Returns the index of an existing text edge using
+        the specified x coordinate and alignment.
+        """
+        for i, te in enumerate(self._textedges[align]):
+            if np.isclose(te.x, x_coord, atol=0.5):
+                return i
+        return None
+
+    def add(self, textline, align):
+        """Adds a new text edge to the current dict."""
+        x = self.get_x_coord(textline, align)
+        y0 = textline.y0
+        y1 = textline.y1
+        te = TextEdge(x, y0, y1, align=align)
+        self._textedges[align].append(te)
+
+    def update(self, textline):
+        """Updates an existing text edge in the current dict."""
+        for align in ["left", "right", "middle"]:
+            x_coord = self.get_x_coord(textline, align)
+            idx = self.find(x_coord, align)
+            if idx is None:
+                self.add(textline, align)
+            else:
+                self._textedges[align][idx].update_coords(
+                    x_coord, textline.y0, edge_tol=self.edge_tol
+                )
+
+    def generate(self, textlines):
+        """Generates the text edges dict based on horizontal text
+        rows.
+        """
+        for tl in textlines:
+            if len(tl.get_text().strip()) > 1:  # TODO: hacky
+                self.update(tl)
+
+    def get_relevant(self):
+        """Returns the list of relevant text edges (all share the same
+        alignment) based on which list intersects horizontal text rows
+        the most.
+        """
+        intersections_sum = {
+            "left": sum(
+                te.intersections for te in self._textedges["left"] if te.is_valid
+            ),
+            "right": sum(
+                te.intersections for te in self._textedges["right"] if te.is_valid
+            ),
+            "middle": sum(
+                te.intersections for te in self._textedges["middle"] if te.is_valid
+            ),
+        }
+
+        # TODO: naive
+        # get vertical textedges that intersect maximum number of
+        # times with horizontal textlines
+        relevant_align = max(intersections_sum.items(), key=itemgetter(1))[0]
+        return self._textedges[relevant_align]
+
+    def get_table_areas(self, textlines, relevant_textedges):
+        """Returns a dict of interesting table areas on the PDF page
+        calculated using relevant text edges.
+        """
+
+        def pad(area, average_row_height):
+            x0 = area[0] - TABLE_AREA_PADDING
+            y0 = area[1] - TABLE_AREA_PADDING
+            x1 = area[2] + TABLE_AREA_PADDING
+            # add a constant since table headers can be relatively up
+            y1 = area[3] + average_row_height * 5
+            return (x0, y0, x1, y1)
+
+        # sort relevant textedges in reading order
+        relevant_textedges.sort(key=lambda te: (-te.y0, te.x))
+
+        table_areas = {}
+        for te in relevant_textedges:
+            if te.is_valid:
+                if not table_areas:
+                    table_areas[(te.x, te.y0, te.x, te.y1)] = None
+                else:
+                    found = None
+                    for area in table_areas:
+                        # check for overlap
+                        if te.y1 >= area[1] and te.y0 <= area[3]:
+                            found = area
+                            break
+                    if found is None:
+                        table_areas[(te.x, te.y0, te.x, te.y1)] = None
+                    else:
+                        table_areas.pop(found)
+                        updated_area = (
+                            found[0],
+                            min(te.y0, found[1]),
+                            max(found[2], te.x),
+                            max(found[3], te.y1),
+                        )
+                        table_areas[updated_area] = None
+
+        # extend table areas based on textlines that overlap
+        # vertically. it's possible that these textlines were
+        # eliminated during textedges generation since numbers and
+        # chars/words/sentences are often aligned differently.
+        # drawback: table areas that have paragraphs on their sides
+        # will include the paragraphs too.
+        sum_textline_height = 0
+        for tl in textlines:
+            sum_textline_height += tl.y1 - tl.y0
+            found = None
+            for area in table_areas:
+                # check for overlap
+                if tl.y0 >= area[1] and tl.y1 <= area[3]:
+                    found = area
+                    break
+            if found is not None:
+                table_areas.pop(found)
+                updated_area = (
+                    min(tl.x0, found[0]),
+                    min(tl.y0, found[1]),
+                    max(found[2], tl.x1),
+                    max(found[3], tl.y1),
+                )
+                table_areas[updated_area] = None
+        average_textline_height = sum_textline_height / float(len(textlines))
+
+        # add some padding to table areas
+        table_areas_padded = {}
+        for area in table_areas:
+            table_areas_padded[pad(area, average_textline_height)] = None
+
+        return table_areas_padded
 
 
 class Cell(object):
@@ -67,11 +285,14 @@ class Cell(object):
         self.bottom = False
         self.hspan = False
         self.vspan = False
-        self._text = ''
+        self._text = ""
 
     def __repr__(self):
-        return '<Cell x1={} y1={} x2={} y2={}>'.format(
+        x1 = round(self.x1)
-            round(self.x1, 2), round(self.y1, 2), round(self.x2, 2), round(self.y2, 2))
+        y1 = round(self.y1)
+        x2 = round(self.x2)
+        y2 = round(self.y2)
+        return f"<Cell x1={x1} y1={y1} x2={x2} y2={y2}>"
 
     @property
     def text(self):
@@ -79,12 +300,11 @@ class Cell(object):
 
     @text.setter
     def text(self, t):
-        self._text = ''.join([self._text, t])
+        self._text = "".join([self._text, t])
 
     @property
     def bound(self):
-        """The number of sides on which the cell is bounded.
+        """The number of sides on which the cell is bounded."""
-        """
         return self.top + self.bottom + self.left + self.right
 
 
@@ -116,11 +336,11 @@ class Table(object):
         PDF page number.
 
     """
+
     def __init__(self, cols, rows):
         self.cols = cols
         self.rows = rows
-        self.cells = [[Cell(c[0], r[1], c[1], r[0])
+        self.cells = [[Cell(c[0], r[1], c[1], r[0]) for c in cols] for r in rows]
-                       for c in cols] for r in rows]
         self.df = None
         self.shape = (0, 0)
         self.accuracy = 0
@@ -129,12 +349,18 @@ class Table(object):
         self.page = None
 
     def __repr__(self):
-        return '<{} shape={}>'.format(self.__class__.__name__, self.shape)
+        return f"<{self.__class__.__name__} shape={self.shape}>"
+
+    def __lt__(self, other):
+        if self.page == other.page:
+            if self.order < other.order:
+                return True
+        if self.page < other.page:
+            return True
 
     @property
     def data(self):
-        """Returns two-dimensional list of strings in table.
+        """Returns two-dimensional list of strings in table."""
-        """
         d = []
         for row in self.cells:
             d.append([cell.text.strip() for cell in row])
@@ -147,22 +373,21 @@ class Table(object):
         """
         # pretty?
         report = {
-            'accuracy': round(self.accuracy, 2),
+            "accuracy": round(self.accuracy, 2),
-            'whitespace': round(self.whitespace, 2),
+            "whitespace": round(self.whitespace, 2),
-            'order': self.order,
+            "order": self.order,
-            'page': self.page
+            "page": self.page,
         }
         return report
 
     def set_all_edges(self):
-        """Sets all table edges to True.
+        """Sets all table edges to True."""
-        """
         for row in self.cells:
             for cell in row:
                 cell.left = cell.right = cell.top = cell.bottom = True
         return self
 
-    def set_edges(self, vertical, horizontal, joint_close_tol=2):
+    def set_edges(self, vertical, horizontal, joint_tol=2):
         """Sets a cell's edges to True depending on whether the cell's
         coordinates overlap with the line's coordinates within a
         tolerance.
@@ -178,12 +403,21 @@ class Table(object):
         for v in vertical:
             # find closest x coord
             # iterate over y coords and find closest start and end points
-            i = [i for i, t in enumerate(self.cols)
+            i = [
-                 if np.isclose(v[0], t[0], atol=joint_close_tol)]
+                i
-            j = [j for j, t in enumerate(self.rows)
+                for i, t in enumerate(self.cols)
-                 if np.isclose(v[3], t[0], atol=joint_close_tol)]
+                if np.isclose(v[0], t[0], atol=joint_tol)
-            k = [k for k, t in enumerate(self.rows)
+            ]
-                 if np.isclose(v[1], t[0], atol=joint_close_tol)]
+            j = [
+                j
+                for j, t in enumerate(self.rows)
+                if np.isclose(v[3], t[0], atol=joint_tol)
+            ]
+            k = [
+                k
+                for k, t in enumerate(self.rows)
+                if np.isclose(v[1], t[0], atol=joint_tol)
+            ]
             if not j:
                 continue
             J = j[0]
@@ -229,12 +463,21 @@ class Table(object):
         for h in horizontal:
             # find closest y coord
             # iterate over x coords and find closest start and end points
-            i = [i for i, t in enumerate(self.rows)
+            i = [
-                 if np.isclose(h[1], t[0], atol=joint_close_tol)]
+                i
-            j = [j for j, t in enumerate(self.cols)
+                for i, t in enumerate(self.rows)
-                 if np.isclose(h[0], t[0], atol=joint_close_tol)]
+                if np.isclose(h[1], t[0], atol=joint_tol)
-            k = [k for k, t in enumerate(self.cols)
+            ]
-                 if np.isclose(h[2], t[0], atol=joint_close_tol)]
+            j = [
+                j
+                for j, t in enumerate(self.cols)
+                if np.isclose(h[0], t[0], atol=joint_tol)
+            ]
+            k = [
+                k
+                for k, t in enumerate(self.cols)
+                if np.isclose(h[2], t[0], atol=joint_tol)
+            ]
             if not j:
                 continue
             J = j[0]
@@ -251,7 +494,7 @@ class Table(object):
                         self.cells[L][J].top = True
                         J += 1
             elif i == []:  # only bottom edge
-                I = len(self.rows) - 1
+                L = len(self.rows) - 1
                 if k:
                     K = k[0]
                     while J < K:
@@ -280,8 +523,7 @@ class Table(object):
         return self
 
     def set_border(self):
-        """Sets table border edges to True.
+        """Sets table border edges to True."""
-        """
         for r in range(len(self.rows)):
             self.cells[r][0].left = True
             self.cells[r][len(self.cols) - 1].right = True
@@ -321,33 +563,6 @@ class Table(object):
                     cell.hspan = True
         return self
 
-    def plot(self, geometry_type):
-        """Plot geometry found on PDF page based on geometry_type
-        specified, useful for debugging and playing with different
-        parameters to get the best output.
-
-        Parameters
-        ----------
-        geometry_type : str
-            The geometry type for which a plot should be generated.
-            Can be 'text', 'table', 'contour', 'joint', 'line'
-
-        """
-        if self.flavor == 'stream' and geometry_type in ['contour', 'joint', 'line']:
-            raise NotImplementedError("{} cannot be plotted with flavor='stream'".format(
-                                       geometry_type))
-
-        if geometry_type == 'text':
-            plot_text(self._text)
-        elif geometry_type == 'table':
-            plot_table(self)
-        elif geometry_type == 'contour':
-            plot_contour(self._image)
-        elif geometry_type == 'joint':
-            plot_joint(self._image)
-        elif geometry_type == 'line':
-            plot_line(self._segments)
-
     def to_csv(self, path, **kwargs):
         """Writes Table to a comma-separated values (csv) file.
 
@@ -359,12 +574,7 @@ class Table(object):
             Output filepath.
 
         """
-        kw = {
+        kw = {"encoding": "utf-8", "index": False, "header": False, "quoting": 1}
-            'encoding': 'utf-8',
-            'index': False,
-            'header': False,
-            'quoting': 1
-        }
         kw.update(kwargs)
         self.df.to_csv(path, **kw)
 
@@ -379,12 +589,10 @@ class Table(object):
             Output filepath.
 
         """
-        kw = {
+        kw = {"orient": "records"}
-            'orient': 'records'
-        }
         kw.update(kwargs)
         json_string = self.df.to_json(**kw)
-        with open(path, 'w') as f:
+        with open(path, "w") as f:
             f.write(json_string)
 
     def to_excel(self, path, **kwargs):
@@ -399,8 +607,8 @@ class Table(object):
 
         """
         kw = {
-            'sheet_name': 'page-{}-table-{}'.format(self.page, self.order),
+            "sheet_name": f"page-{self.page}-table-{self.order}",
-            'encoding': 'utf-8'
+            "encoding": "utf-8",
         }
         kw.update(kwargs)
         writer = pd.ExcelWriter(path)
@@ -419,9 +627,43 @@ class Table(object):
 
         """
         html_string = self.df.to_html(**kwargs)
-        with open(path, 'w') as f:
+        with open(path, "w", encoding="utf-8") as f:
             f.write(html_string)
 
+    def to_markdown(self, path, **kwargs):
+        """Writes Table to a Markdown file.
+
+        For kwargs, check :meth:`pandas.DataFrame.to_markdown`.
+
+        Parameters
+        ----------
+        path : str
+            Output filepath.
+
+        """
+        md_string = self.df.to_markdown(**kwargs)
+        with open(path, "w", encoding="utf-8") as f:
+            f.write(md_string)
+
+    def to_sqlite(self, path, **kwargs):
+        """Writes Table to sqlite database.
+
+        For kwargs, check :meth:`pandas.DataFrame.to_sql`.
+
+        Parameters
+        ----------
+        path : str
+            Output filepath.
+
+        """
+        kw = {"if_exists": "replace", "index": False}
+        kw.update(kwargs)
+        conn = sqlite3.connect(path)
+        table_name = f"page-{self.page}-table-{self.order}"
+        self.df.to_sql(table_name, conn, **kw)
+        conn.commit()
+        conn.close()
+
 
 class TableList(object):
     """Defines a list of camelot.core.Table objects. Each table can
@@ -433,12 +675,12 @@ class TableList(object):
         Number of tables in the list.
 
     """
+
     def __init__(self, tables):
         self._tables = tables
 
     def __repr__(self):
-        return '<{} n={}>'.format(
+        return f"<{self.__class__.__name__} n={self.n}>"
-            self.__class__.__name__, self.n)
 
     def __len__(self):
         return len(self._tables)
@@ -448,37 +690,35 @@ class TableList(object):
 
     @staticmethod
     def _format_func(table, f):
-        return getattr(table, 'to_{}'.format(f))
+        return getattr(table, f"to_{f}")
 
     @property
     def n(self):
         return len(self)
 
     def _write_file(self, f=None, **kwargs):
-        dirname = kwargs.get('dirname')
+        dirname = kwargs.get("dirname")
-        root = kwargs.get('root')
+        root = kwargs.get("root")
-        ext = kwargs.get('ext')
+        ext = kwargs.get("ext")
         for table in self._tables:
-            filename = os.path.join('{}-page-{}-table-{}{}'.format(
+            filename = f"{root}-page-{table.page}-table-{table.order}{ext}"
-                                    root, table.page, table.order, ext))
             filepath = os.path.join(dirname, filename)
             to_format = self._format_func(table, f)
             to_format(filepath)
 
     def _compress_dir(self, **kwargs):
-        path = kwargs.get('path')
+        path = kwargs.get("path")
-        dirname = kwargs.get('dirname')
+        dirname = kwargs.get("dirname")
-        root = kwargs.get('root')
+        root = kwargs.get("root")
-        ext = kwargs.get('ext')
+        ext = kwargs.get("ext")
-        zipname = os.path.join(os.path.dirname(path), root) + '.zip'
+        zipname = os.path.join(os.path.dirname(path), root) + ".zip"
-        with zipfile.ZipFile(zipname, 'w', allowZip64=True) as z:
+        with zipfile.ZipFile(zipname, "w", allowZip64=True) as z:
             for table in self._tables:
-                filename = os.path.join('{}-page-{}-table-{}{}'.format(
+                filename = f"{root}-page-{table.page}-table-{table.order}{ext}"
-                                        root, table.page, table.order, ext))
                 filepath = os.path.join(dirname, filename)
                 z.write(filepath, os.path.basename(filepath))
 
-    def export(self, path, f='csv', compress=False):
+    def export(self, path, f="csv", compress=False):
         """Exports the list of tables to specified file format.
 
         Parameters
@@ -486,7 +726,7 @@ class TableList(object):
         path : str
             Output filepath.
         f : str
-            File format. Can be csv, json, excel and html.
+            File format. Can be csv, excel, html, json, markdown or sqlite.
         compress : bool
             Whether or not to add files to a ZIP archive.
 
@@ -497,25 +737,28 @@ class TableList(object):
         if compress:
             dirname = tempfile.mkdtemp()
 
-        kwargs = {
+        kwargs = {"path": path, "dirname": dirname, "root": root, "ext": ext}
-            'path': path,
-            'dirname': dirname,
-            'root': root,
-            'ext': ext
-        }
 
-        if f in ['csv', 'json', 'html']:
+        if f in ["csv", "html", "json", "markdown"]:
             self._write_file(f=f, **kwargs)
             if compress:
                 self._compress_dir(**kwargs)
-        elif f == 'excel':
+        elif f == "excel":
             filepath = os.path.join(dirname, basename)
             writer = pd.ExcelWriter(filepath)
             for table in self._tables:
-                sheet_name = 'page-{}-table-{}'.format(table.page, table.order)
+                sheet_name = f"page-{table.page}-table-{table.order}"
-                table.df.to_excel(writer, sheet_name=sheet_name, encoding='utf-8')
+                table.df.to_excel(writer, sheet_name=sheet_name, encoding="utf-8")
             writer.save()
             if compress:
-                zipname = os.path.join(os.path.dirname(path), root) + '.zip'
+                zipname = os.path.join(os.path.dirname(path), root) + ".zip"
-                with zipfile.ZipFile(zipname, 'w', allowZip64=True) as z:
+                with zipfile.ZipFile(zipname, "w", allowZip64=True) as z:
+                    z.write(filepath, os.path.basename(filepath))
+        elif f == "sqlite":
+            filepath = os.path.join(dirname, basename)
+            for table in self._tables:
+                table.to_sqlite(filepath)
+            if compress:
+                zipname = os.path.join(os.path.dirname(path), root) + ".zip"
+                with zipfile.ZipFile(zipname, "w", allowZip64=True) as z:
                     z.write(filepath, os.path.basename(filepath))

camelot/handlers.py

@@ -1,13 +1,20 @@
 # -*- coding: utf-8 -*-
 
 import os
+import sys
 
 from PyPDF2 import PdfFileReader, PdfFileWriter
 
 from .core import TableList
 from .parsers import Stream, Lattice
-from .utils import (TemporaryDirectory, get_page_layout, get_text_objects,
+from .utils import (
-                    get_rotation)
+    TemporaryDirectory,
+    get_page_layout,
+    get_text_objects,
+    get_rotation,
+    is_url,
+    download_url,
+)
 
 
 class PDFHandler(object):
@@ -17,29 +24,41 @@ class PDFHandler(object):
 
     Parameters
     ----------
-    filename : str
+    filepath : str
-        Path to PDF file.
+        Filepath or URL of the PDF file.
     pages : str, optional (default: '1')
         Comma-separated page numbers.
-        Example: 1,3,4 or 1,4-end.
+        Example: '1,3,4' or '1,4-end' or 'all'.
+    password : str, optional (default: None)
+        Password for decryption.
 
     """
-    def __init__(self, filename, pages='1'):
-        self.filename = filename
-        if not self.filename.endswith('.pdf'):
-            raise NotImplementedError("File format not supported")
-        self.pages = self._get_pages(self.filename, pages)
 
-    def _get_pages(self, filename, pages):
+    def __init__(self, filepath, pages="1", password=None):
+        if is_url(filepath):
+            filepath = download_url(filepath)
+        self.filepath = filepath
+        # if not filepath.lower().endswith(".pdf"):
+        #     raise NotImplementedError("File format not supported")
+
+        if password is None:
+            self.password = ""
+        else:
+            self.password = password
+            if sys.version_info[0] < 3:
+                self.password = self.password.encode("ascii")
+        self.pages = self._get_pages(pages)
+
+    def _get_pages(self, pages):
         """Converts pages string to list of ints.
 
         Parameters
         ----------
-        filename : str
+        filepath : str
-            Path to PDF file.
+            Filepath or URL of the PDF file.
         pages : str, optional (default: '1')
             Comma-separated page numbers.
-            Example: 1,3,4 or 1,4-end.
+            Example: '1,3,4' or '1,4-end' or 'all'.
 
         Returns
         -------
@@ -48,73 +67,84 @@ class PDFHandler(object):
 
         """
         page_numbers = []
-        if pages == '1':
+
-            page_numbers.append({'start': 1, 'end': 1})
+        if pages == "1":
+            page_numbers.append({"start": 1, "end": 1})
         else:
-            infile = PdfFileReader(open(filename, 'rb'), strict=False)
+            with open(self.filepath, "rb") as f:
-            if pages == 'all':
+                infile = PdfFileReader(f, strict=False)
-                page_numbers.append({'start': 1, 'end': infile.getNumPages()})
+
-            else:
+                if infile.isEncrypted:
-                for r in pages.split(','):
+                    infile.decrypt(self.password)
-                    if '-' in r:
+
-                        a, b = r.split('-')
+                if pages == "all":
-                        if b == 'end':
+                    page_numbers.append({"start": 1, "end": infile.getNumPages()})
-                            b = infile.getNumPages()
+                else:
-                        page_numbers.append({'start': int(a), 'end': int(b)})
+                    for r in pages.split(","):
-                    else:
+                        if "-" in r:
-                        page_numbers.append({'start': int(r), 'end': int(r)})
+                            a, b = r.split("-")
+                            if b == "end":
+                                b = infile.getNumPages()
+                            page_numbers.append({"start": int(a), "end": int(b)})
+                        else:
+                            page_numbers.append({"start": int(r), "end": int(r)})
+
         P = []
         for p in page_numbers:
-            P.extend(range(p['start'], p['end'] + 1))
+            P.extend(range(p["start"], p["end"] + 1))
         return sorted(set(P))
 
-    def _save_page(self, filename, page, temp):
+    def _save_page(self, filepath, page, temp):
         """Saves specified page from PDF into a temporary directory.
 
         Parameters
         ----------
-        filename : str
+        filepath : str
-            Path to PDF file.
+            Filepath or URL of the PDF file.
         page : int
             Page number.
         temp : str
             Tmp directory.
 
         """
-        with open(filename, 'rb') as fileobj:
+        with open(filepath, "rb") as fileobj:
             infile = PdfFileReader(fileobj, strict=False)
             if infile.isEncrypted:
-                infile.decrypt('')
+                infile.decrypt(self.password)
-            fpath = os.path.join(temp, 'page-{0}.pdf'.format(page))
+            fpath = os.path.join(temp, f"page-{page}.pdf")
             froot, fext = os.path.splitext(fpath)
             p = infile.getPage(page - 1)
             outfile = PdfFileWriter()
             outfile.addPage(p)
-            with open(fpath, 'wb') as f:
+            with open(fpath, "wb") as f:
                 outfile.write(f)
             layout, dim = get_page_layout(fpath)
             # fix rotated PDF
-            lttextlh = get_text_objects(layout, ltype="lh")
+            chars = get_text_objects(layout, ltype="char")
-            lttextlv = get_text_objects(layout, ltype="lv")
+            horizontal_text = get_text_objects(layout, ltype="horizontal_text")
-            ltchar = get_text_objects(layout, ltype="char")
+            vertical_text = get_text_objects(layout, ltype="vertical_text")
-            rotation = get_rotation(lttextlh, lttextlv, ltchar)
+            rotation = get_rotation(chars, horizontal_text, vertical_text)
-            if rotation != '':
+            if rotation != "":
-                fpath_new = ''.join([froot.replace('page', 'p'), '_rotated', fext])
+                fpath_new = "".join([froot.replace("page", "p"), "_rotated", fext])
                 os.rename(fpath, fpath_new)
-                infile = PdfFileReader(open(fpath_new, 'rb'), strict=False)
+                instream = open(fpath_new, "rb")
+                infile = PdfFileReader(instream, strict=False)
                 if infile.isEncrypted:
-                    infile.decrypt('')
+                    infile.decrypt(self.password)
                 outfile = PdfFileWriter()
                 p = infile.getPage(0)
-                if rotation == 'anticlockwise':
+                if rotation == "anticlockwise":
                     p.rotateClockwise(90)
-                elif rotation == 'clockwise':
+                elif rotation == "clockwise":
                     p.rotateCounterClockwise(90)
                 outfile.addPage(p)
-                with open(fpath, 'wb') as f:
+                with open(fpath, "wb") as f:
                     outfile.write(f)
+                instream.close()
 
-    def parse(self, flavor='lattice', **kwargs):
+    def parse(
+        self, flavor="lattice", suppress_stdout=False, layout_kwargs={}, **kwargs
+    ):
         """Extracts tables by calling parser.get_tables on all single
         page PDFs.
 
@@ -123,6 +153,10 @@ class PDFHandler(object):
         flavor : str (default: 'lattice')
             The parsing method to use ('lattice' or 'stream').
             Lattice is used by default.
+        suppress_stdout : str (default: False)
+            Suppress logs and warnings.
+        layout_kwargs : dict, optional (default: {})
+            A dict of `pdfminer.layout.LAParams <https://github.com/euske/pdfminer/blob/master/pdfminer/layout.py#L33>`_ kwargs.
         kwargs : dict
             See camelot.read_pdf kwargs.
 
@@ -130,19 +164,17 @@ class PDFHandler(object):
         -------
         tables : camelot.core.TableList
             List of tables found in PDF.
-        geometry : camelot.core.GeometryList
-            List of geometry objects (contours, lines, joints) found
-            in PDF.
 
         """
         tables = []
         with TemporaryDirectory() as tempdir:
             for p in self.pages:
-                self._save_page(self.filename, p, tempdir)
+                self._save_page(self.filepath, p, tempdir)
-            pages = [os.path.join(tempdir, 'page-{0}.pdf'.format(p))
+            pages = [os.path.join(tempdir, f"page-{p}.pdf") for p in self.pages]
-                     for p in self.pages]
+            parser = Lattice(**kwargs) if flavor == "lattice" else Stream(**kwargs)
-            parser = Lattice(**kwargs) if flavor == 'lattice' else Stream(**kwargs)
             for p in pages:
-                t = parser.extract_tables(p)
+                t = parser.extract_tables(
+                    p, suppress_stdout=suppress_stdout, layout_kwargs=layout_kwargs
+                )
                 tables.extend(t)
-        return TableList(tables)
+        return TableList(sorted(tables))

camelot/image_processing.py

@@ -1,7 +1,5 @@
 # -*- coding: utf-8 -*-
 
-from __future__ import division
-
 import cv2
 import numpy as np
 
@@ -39,16 +37,23 @@ def adaptive_threshold(imagename, process_background=False, blocksize=15, c=-2):
 
     if process_background:
         threshold = cv2.adaptiveThreshold(
-            gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
+            gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, blocksize, c
-            cv2.THRESH_BINARY, blocksize, c)
+        )
     else:
         threshold = cv2.adaptiveThreshold(
-            np.invert(gray), 255,
+            np.invert(gray),
-            cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, blocksize, c)
+            255,
+            cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
+            cv2.THRESH_BINARY,
+            blocksize,
+            c,
+        )
     return img, threshold
 
 
-def find_lines(threshold, direction='horizontal', line_size_scaling=15, iterations=0):
+def find_lines(
+    threshold, regions=None, direction="horizontal", line_scale=15, iterations=0
+):
     """Finds horizontal and vertical lines by applying morphological
     transformations on an image.
 
@@ -56,9 +61,13 @@ def find_lines(threshold, direction='horizontal', line_size_scaling=15, iteratio
     ----------
     threshold : object
         numpy.ndarray representing the thresholded image.
+    regions : list, optional (default: None)
+        List of page regions that may contain tables of the form x1,y1,x2,y2
+        where (x1, y1) -> left-top and (x2, y2) -> right-bottom
+        in image coordinate space.
     direction : string, optional (default: 'horizontal')
         Specifies whether to find vertical or horizontal lines.
-    line_size_scaling : int, optional (default: 15)
+    line_scale : int, optional (default: 15)
         Factor by which the page dimensions will be divided to get
         smallest length of lines that should be detected.
 
@@ -82,15 +91,21 @@ def find_lines(threshold, direction='horizontal', line_size_scaling=15, iteratio
     """
     lines = []
 
-    if direction == 'vertical':
+    if direction == "vertical":
-        size = threshold.shape[0] // line_size_scaling
+        size = threshold.shape[0] // line_scale
         el = cv2.getStructuringElement(cv2.MORPH_RECT, (1, size))
-    elif direction == 'horizontal':
+    elif direction == "horizontal":
-        size = threshold.shape[1] // line_size_scaling
+        size = threshold.shape[1] // line_scale
         el = cv2.getStructuringElement(cv2.MORPH_RECT, (size, 1))
     elif direction is None:
-        raise ValueError("Specify direction as either 'vertical' or"
+        raise ValueError("Specify direction as either 'vertical' or 'horizontal'")
-                         " 'horizontal'")
+
+    if regions is not None:
+        region_mask = np.zeros(threshold.shape)
+        for region in regions:
+            x, y, w, h = region
+            region_mask[y : y + h, x : x + w] = 1
+        threshold = np.multiply(threshold, region_mask)
 
     threshold = cv2.erode(threshold, el)
     threshold = cv2.dilate(threshold, el)
@@ -98,25 +113,27 @@ def find_lines(threshold, direction='horizontal', line_size_scaling=15, iteratio
 
     try:
         _, contours, _ = cv2.findContours(
-            threshold, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+            threshold.astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+        )
     except ValueError:
         # for opencv backward compatibility
         contours, _ = cv2.findContours(
-            threshold, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+            threshold.astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+        )
 
     for c in contours:
         x, y, w, h = cv2.boundingRect(c)
         x1, x2 = x, x + w
         y1, y2 = y, y + h
-        if direction == 'vertical':
+        if direction == "vertical":
             lines.append(((x1 + x2) // 2, y2, (x1 + x2) // 2, y1))
-        elif direction == 'horizontal':
+        elif direction == "horizontal":
             lines.append((x1, (y1 + y2) // 2, x2, (y1 + y2) // 2))
 
     return dmask, lines
 
 
-def find_table_contours(vertical, horizontal):
+def find_contours(vertical, horizontal):
     """Finds table boundaries using OpenCV's findContours.
 
     Parameters
@@ -138,11 +155,14 @@ def find_table_contours(vertical, horizontal):
 
     try:
         __, contours, __ = cv2.findContours(
-            mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+            mask.astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+        )
     except ValueError:
         # for opencv backward compatibility
         contours, __ = cv2.findContours(
-            mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+            mask.astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+        )
+    # sort in reverse based on contour area and use first 10 contours
     contours = sorted(contours, key=cv2.contourArea, reverse=True)[:10]
 
     cont = []
@@ -153,7 +173,7 @@ def find_table_contours(vertical, horizontal):
     return cont
 
 
-def find_table_joints(contours, vertical, horizontal):
+def find_joints(contours, vertical, horizontal):
     """Finds joints/intersections present inside each table boundary.
 
     Parameters
@@ -176,18 +196,20 @@ def find_table_joints(contours, vertical, horizontal):
         and (x2, y2) -> rt in image coordinate space.
 
     """
-    joints = np.bitwise_and(vertical, horizontal)
+    joints = np.multiply(vertical, horizontal)
     tables = {}
     for c in contours:
         x, y, w, h = c
         roi = joints[y : y + h, x : x + w]
         try:
             __, jc, __ = cv2.findContours(
-                roi, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE)
+                roi.astype(np.uint8), cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE
+            )
         except ValueError:
             # for opencv backward compatibility
             jc, __ = cv2.findContours(
-                roi, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE)
+                roi.astype(np.uint8), cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE
+            )
         if len(jc) <= 4:  # remove contours with less than 4 joints
             continue
         joint_coords = []

camelot/io.py

@@ -1,10 +1,20 @@
 # -*- coding: utf-8 -*-
 
+import warnings
+
 from .handlers import PDFHandler
 from .utils import validate_input, remove_extra
 
 
-def read_pdf(filepath, pages='1', flavor='lattice', **kwargs):
+def read_pdf(
+    filepath,
+    pages="1",
+    password=None,
+    flavor="lattice",
+    suppress_stdout=False,
+    layout_kwargs={},
+    **kwargs
+):
     """Read PDF and return extracted tables.
 
     Note: kwargs annotated with ^ can only be used with flavor='stream'
@@ -13,14 +23,20 @@ def read_pdf(filepath, pages='1', flavor='lattice', **kwargs):
     Parameters
     ----------
     filepath : str
-        Path to PDF file.
+        Filepath or URL of the PDF file.
     pages : str, optional (default: '1')
         Comma-separated page numbers.
-        Example: 1,3,4 or 1,4-end.
+        Example: '1,3,4' or '1,4-end' or 'all'.
+    password : str, optional (default: None)
+        Password for decryption.
     flavor : str (default: 'lattice')
         The parsing method to use ('lattice' or 'stream').
         Lattice is used by default.
-    table_area : list, optional (default: None)
+    suppress_stdout : bool, optional (default: True)
+        Print all logs and warnings.
+    layout_kwargs : dict, optional (default: {})
+        A dict of `pdfminer.layout.LAParams <https://github.com/euske/pdfminer/blob/master/pdfminer/layout.py#L33>`_ kwargs.
+    table_areas : list, optional (default: None)
         List of table area strings of the form x1,y1,x2,y2
         where (x1, y1) -> left-top and (x2, y2) -> right-bottom
         in PDF coordinate space.
@@ -32,15 +48,18 @@ def read_pdf(filepath, pages='1', flavor='lattice', **kwargs):
     flag_size : bool, optional (default: False)
         Flag text based on font size. Useful to detect
         super/subscripts. Adds <s></s> around flagged text.
-    row_close_tol^ : int, optional (default: 2)
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
+    row_tol^ : int, optional (default: 2)
         Tolerance parameter used to combine text vertically,
         to generate rows.
-    col_close_tol^ : int, optional (default: 0)
+    column_tol^ : int, optional (default: 0)
         Tolerance parameter used to combine text horizontally,
         to generate columns.
     process_background* : bool, optional (default: False)
         Process background lines.
-    line_size_scaling* : int, optional (default: 15)
+    line_scale* : int, optional (default: 15)
         Line size scaling factor. The larger the value the smaller
         the detected lines. Making it very large will lead to text
         being detected as lines.
@@ -51,10 +70,10 @@ def read_pdf(filepath, pages='1', flavor='lattice', **kwargs):
     shift_text* : list, optional (default: ['l', 't'])
         {'l', 'r', 't', 'b'}
         Direction in which text in a spanning cell will flow.
-    line_close_tol* : int, optional (default: 2)
+    line_tol* : int, optional (default: 2)
         Tolerance parameter used to merge close vertical and horizontal
         lines.
-    joint_close_tol* : int, optional (default: 2)
+    joint_tol* : int, optional (default: 2)
         Tolerance parameter used to decide whether the detected lines
         and points lie close to each other.
     threshold_blocksize* : int, optional (default: 15)
@@ -71,22 +90,30 @@ def read_pdf(filepath, pages='1', flavor='lattice', **kwargs):
         Number of times for erosion/dilation is applied.
 
         For more information, refer `OpenCV's dilate <https://docs.opencv.org/2.4/modules/imgproc/doc/filtering.html#dilate>`_.
-    margins : tuple
+    resolution* : int, optional (default: 300)
-        PDFMiner char_margin, line_margin and word_margin.
+        Resolution used for PDF to PNG conversion.
-
-        For more information, refer `PDFMiner docs <https://euske.github.io/pdfminer/>`_.
 
     Returns
     -------
     tables : camelot.core.TableList
 
     """
-    if flavor not in ['lattice', 'stream']:
+    if flavor not in ["lattice", "stream"]:
-        raise NotImplementedError("Unknown flavor specified."
+        raise NotImplementedError(
-                                  " Use either 'lattice' or 'stream'")
+            "Unknown flavor specified." " Use either 'lattice' or 'stream'"
+        )
 
-    validate_input(kwargs, flavor=flavor)
+    with warnings.catch_warnings():
-    p = PDFHandler(filepath, pages)
+        if suppress_stdout:
-    kwargs = remove_extra(kwargs, flavor=flavor)
+            warnings.simplefilter("ignore")
-    tables = p.parse(flavor=flavor, **kwargs)
+
-    return tables
+        validate_input(kwargs, flavor=flavor)
+        p = PDFHandler(filepath, pages=pages, password=password)
+        kwargs = remove_extra(kwargs, flavor=flavor)
+        tables = p.parse(
+            flavor=flavor,
+            suppress_stdout=suppress_stdout,
+            layout_kwargs=layout_kwargs,
+            **kwargs
+        )
+        return tables

camelot/parsers/base.py

@@ -6,16 +6,15 @@ from ..utils import get_page_layout, get_text_objects
 
 
 class BaseParser(object):
-    """Defines a base parser.
+    """Defines a base parser."""
-    """
+
-    def _generate_layout(self, filename):
+    def _generate_layout(self, filename, layout_kwargs):
         self.filename = filename
-        self.layout, self.dimensions = get_page_layout(
+        self.layout_kwargs = layout_kwargs
-            self.filename,
+        self.layout, self.dimensions = get_page_layout(filename, **layout_kwargs)
-            char_margin=self.char_margin,
+        self.images = get_text_objects(self.layout, ltype="image")
-            line_margin=self.line_margin,
+        self.horizontal_text = get_text_objects(self.layout, ltype="horizontal_text")
-            word_margin=self.word_margin)
+        self.vertical_text = get_text_objects(self.layout, ltype="vertical_text")
-        self.horizontal_text = get_text_objects(self.layout, ltype="lh")
-        self.vertical_text = get_text_objects(self.layout, ltype="lv")
         self.pdf_width, self.pdf_height = self.dimensions
         self.rootname, __ = os.path.splitext(self.filename)
+        self.imagename = "".join([self.rootname, ".png"])

camelot/parsers/lattice.py

@@ -1,25 +1,37 @@
 # -*- coding: utf-8 -*-
 
-from __future__ import division
 import os
+import sys
 import copy
+import locale
 import logging
 import warnings
-import subprocess
 
 import numpy as np
 import pandas as pd
 
 from .base import BaseParser
 from ..core import Table
-from ..utils import (scale_image, scale_pdf, segments_in_bbox, text_in_bbox,
+from ..utils import (
-                     merge_close_lines, get_table_index, compute_accuracy,
+    scale_image,
-                     compute_whitespace)
+    scale_pdf,
-from ..image_processing import (adaptive_threshold, find_lines,
+    segments_in_bbox,
-                                find_table_contours, find_table_joints)
+    text_in_bbox,
+    merge_close_lines,
+    get_table_index,
+    compute_accuracy,
+    compute_whitespace,
+)
+from ..image_processing import (
+    adaptive_threshold,
+    find_lines,
+    find_contours,
+    find_joints,
+)
+from ..backends.image_conversion import BACKENDS
 
 
-logger = logging.getLogger('camelot')
+logger = logging.getLogger("camelot")
 
 
 class Lattice(BaseParser):
@@ -28,13 +40,17 @@ class Lattice(BaseParser):
 
     Parameters
     ----------
-    table_area : list, optional (default: None)
+    table_regions : list, optional (default: None)
+        List of page regions that may contain tables of the form x1,y1,x2,y2
+        where (x1, y1) -> left-top and (x2, y2) -> right-bottom
+        in PDF coordinate space.
+    table_areas : list, optional (default: None)
         List of table area strings of the form x1,y1,x2,y2
         where (x1, y1) -> left-top and (x2, y2) -> right-bottom
         in PDF coordinate space.
     process_background : bool, optional (default: False)
         Process background lines.
-    line_size_scaling : int, optional (default: 15)
+    line_scale : int, optional (default: 15)
         Line size scaling factor. The larger the value the smaller
         the detected lines. Making it very large will lead to text
         being detected as lines.
@@ -50,10 +66,13 @@ class Lattice(BaseParser):
     flag_size : bool, optional (default: False)
         Flag text based on font size. Useful to detect
         super/subscripts. Adds <s></s> around flagged text.
-    line_close_tol : int, optional (default: 2)
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
+    line_tol : int, optional (default: 2)
         Tolerance parameter used to merge close vertical and horizontal
         lines.
-    joint_close_tol : int, optional (default: 2)
+    joint_tol : int, optional (default: 2)
         Tolerance parameter used to decide whether the detected lines
         and points lie close to each other.
     threshold_blocksize : int, optional (default: 15)
@@ -70,30 +89,77 @@ class Lattice(BaseParser):
         Number of times for erosion/dilation is applied.
 
         For more information, refer `OpenCV's dilate <https://docs.opencv.org/2.4/modules/imgproc/doc/filtering.html#dilate>`_.
-    margins : tuple
+    resolution : int, optional (default: 300)
-        PDFMiner char_margin, line_margin and word_margin.
+        Resolution used for PDF to PNG conversion.
-
-        For more information, refer `PDFMiner docs <https://euske.github.io/pdfminer/>`_.
 
     """
-    def __init__(self, table_area=None, process_background=False,
+
-                 line_size_scaling=15, copy_text=None, shift_text=['l', 't'],
+    def __init__(
-                 split_text=False, flag_size=False, line_close_tol=2,
+        self,
-                 joint_close_tol=2, threshold_blocksize=15, threshold_constant=-2,
+        table_regions=None,
-                 iterations=0, margins=(1.0, 0.5, 0.1), **kwargs):
+        table_areas=None,
-        self.table_area = table_area
+        process_background=False,
+        line_scale=15,
+        copy_text=None,
+        shift_text=["l", "t"],
+        split_text=False,
+        flag_size=False,
+        strip_text="",
+        line_tol=2,
+        joint_tol=2,
+        threshold_blocksize=15,
+        threshold_constant=-2,
+        iterations=0,
+        resolution=300,
+        backend="ghostscript",
+        **kwargs,
+    ):
+        self.table_regions = table_regions
+        self.table_areas = table_areas
         self.process_background = process_background
-        self.line_size_scaling = line_size_scaling
+        self.line_scale = line_scale
         self.copy_text = copy_text
         self.shift_text = shift_text
         self.split_text = split_text
         self.flag_size = flag_size
-        self.line_close_tol = line_close_tol
+        self.strip_text = strip_text
-        self.joint_close_tol = joint_close_tol
+        self.line_tol = line_tol
+        self.joint_tol = joint_tol
         self.threshold_blocksize = threshold_blocksize
         self.threshold_constant = threshold_constant
         self.iterations = iterations
-        self.char_margin, self.line_margin, self.word_margin = margins
+        self.resolution = resolution
+        self.backend = Lattice._get_backend(backend)
+
+    @staticmethod
+    def _get_backend(backend):
+        def implements_convert():
+            methods = [
+                method for method in dir(backend) if method.startswith("__") is False
+            ]
+            return "convert" in methods
+
+        if isinstance(backend, str):
+            if backend not in BACKENDS.keys():
+                raise NotImplementedError(
+                    f"Unknown backend '{backend}' specified. Please use either 'poppler' or 'ghostscript'."
+                )
+
+            if backend == "ghostscript":
+                warnings.warn(
+                    "'ghostscript' will be replaced by 'poppler' as the default image conversion"
+                    " backend in v0.12.0. You can try out 'poppler' with backend='poppler'.",
+                    DeprecationWarning,
+                )
+
+            return BACKENDS[backend]()
+        else:
+            if not implements_convert():
+                raise NotImplementedError(
+                    f"'{backend}' must implement a 'convert' method"
+                )
+
+            return backend
 
     @staticmethod
     def _reduce_index(t, idx, shift_text):
@@ -121,19 +187,19 @@ class Lattice(BaseParser):
         indices = []
         for r_idx, c_idx, text in idx:
             for d in shift_text:
-                if d == 'l':
+                if d == "l":
                     if t.cells[r_idx][c_idx].hspan:
                         while not t.cells[r_idx][c_idx].left:
                             c_idx -= 1
-                if d == 'r':
+                if d == "r":
                     if t.cells[r_idx][c_idx].hspan:
                         while not t.cells[r_idx][c_idx].right:
                             c_idx += 1
-                if d == 't':
+                if d == "t":
                     if t.cells[r_idx][c_idx].vspan:
                         while not t.cells[r_idx][c_idx].top:
                             r_idx -= 1
-                if d == 'b':
+                if d == "b":
                     if t.cells[r_idx][c_idx].vspan:
                         while not t.cells[r_idx][c_idx].bottom:
                             r_idx += 1
@@ -162,54 +228,37 @@ class Lattice(BaseParser):
             if f == "h":
                 for i in range(len(t.cells)):
                     for j in range(len(t.cells[i])):
-                        if t.cells[i][j].text.strip() == '':
+                        if t.cells[i][j].text.strip() == "":
                             if t.cells[i][j].hspan and not t.cells[i][j].left:
                                 t.cells[i][j].text = t.cells[i][j - 1].text
             elif f == "v":
                 for i in range(len(t.cells)):
                     for j in range(len(t.cells[i])):
-                        if t.cells[i][j].text.strip() == '':
+                        if t.cells[i][j].text.strip() == "":
                             if t.cells[i][j].vspan and not t.cells[i][j].top:
                                 t.cells[i][j].text = t.cells[i - 1][j].text
         return t
 
-    def _generate_image(self):
-        # TODO: hacky, get rid of ghostscript #96
-        def get_platform():
-            import platform
-
-            info = {
-                'system': platform.system().lower(),
-                'machine': platform.machine().lower()
-            }
-            return info
-
-        self.imagename = ''.join([self.rootname, '.png'])
-        gs_call = [
-            '-q',
-            '-sDEVICE=png16m',
-            '-o',
-            self.imagename,
-            '-r600',
-            self.filename
-        ]
-        info = get_platform()
-        if info['system'] == 'windows':
-            bit = info['machine'][-2:]
-            gs_call.insert(0, 'gswin{}c.exe'.format(bit))
-        else:
-            if 'ghostscript' in subprocess.check_output(['gs', '-version']).decode('utf-8').lower():
-                gs_call.insert(0, 'gs')
-            else:
-                gs_call.insert(0, "gsc")
-        subprocess.call(
-            gs_call, stdout=open(os.devnull, 'w'),
-            stderr=subprocess.STDOUT)
-
     def _generate_table_bbox(self):
+        def scale_areas(areas):
+            scaled_areas = []
+            for area in areas:
+                x1, y1, x2, y2 = area.split(",")
+                x1 = float(x1)
+                y1 = float(y1)
+                x2 = float(x2)
+                y2 = float(y2)
+                x1, y1, x2, y2 = scale_pdf((x1, y1, x2, y2), image_scalers)
+                scaled_areas.append((x1, y1, abs(x2 - x1), abs(y2 - y1)))
+            return scaled_areas
+
         self.image, self.threshold = adaptive_threshold(
-            self.imagename, process_background=self.process_background,
+            self.imagename,
-            blocksize=self.threshold_blocksize, c=self.threshold_constant)
+            process_background=self.process_background,
+            blocksize=self.threshold_blocksize,
+            c=self.threshold_constant,
+        )
+
         image_width = self.image.shape[1]
         image_height = self.image.shape[0]
         image_width_scaler = image_width / float(self.pdf_width)
@@ -219,85 +268,110 @@ class Lattice(BaseParser):
         image_scalers = (image_width_scaler, image_height_scaler, self.pdf_height)
         pdf_scalers = (pdf_width_scaler, pdf_height_scaler, image_height)
 
-        vertical_mask, vertical_segments = find_lines(
+        if self.table_areas is None:
-            self.threshold, direction='vertical',
+            regions = None
-            line_size_scaling=self.line_size_scaling, iterations=self.iterations)
+            if self.table_regions is not None:
-        horizontal_mask, horizontal_segments = find_lines(
+                regions = scale_areas(self.table_regions)
-            self.threshold, direction='horizontal',
-            line_size_scaling=self.line_size_scaling, iterations=self.iterations)
 
-        if self.table_area is not None:
+            vertical_mask, vertical_segments = find_lines(
-            areas = []
+                self.threshold,
-            for area in self.table_area:
+                regions=regions,
-                x1, y1, x2, y2 = area.split(",")
+                direction="vertical",
-                x1 = float(x1)
+                line_scale=self.line_scale,
-                y1 = float(y1)
+                iterations=self.iterations,
-                x2 = float(x2)
+            )
-                y2 = float(y2)
+            horizontal_mask, horizontal_segments = find_lines(
-                x1, y1, x2, y2 = scale_pdf((x1, y1, x2, y2), image_scalers)
+                self.threshold,
-                areas.append((x1, y1, abs(x2 - x1), abs(y2 - y1)))
+                regions=regions,
-            table_bbox = find_table_joints(areas, vertical_mask, horizontal_mask)
+                direction="horizontal",
+                line_scale=self.line_scale,
+                iterations=self.iterations,
+            )
+
+            contours = find_contours(vertical_mask, horizontal_mask)
+            table_bbox = find_joints(contours, vertical_mask, horizontal_mask)
         else:
-            contours = find_table_contours(vertical_mask, horizontal_mask)
+            vertical_mask, vertical_segments = find_lines(
-            table_bbox = find_table_joints(contours, vertical_mask, horizontal_mask)
+                self.threshold,
+                direction="vertical",
+                line_scale=self.line_scale,
+                iterations=self.iterations,
+            )
+            horizontal_mask, horizontal_segments = find_lines(
+                self.threshold,
+                direction="horizontal",
+                line_scale=self.line_scale,
+                iterations=self.iterations,
+            )
+
+            areas = scale_areas(self.table_areas)
+            table_bbox = find_joints(areas, vertical_mask, horizontal_mask)
 
         self.table_bbox_unscaled = copy.deepcopy(table_bbox)
 
         self.table_bbox, self.vertical_segments, self.horizontal_segments = scale_image(
-            table_bbox, vertical_segments, horizontal_segments, pdf_scalers)
+            table_bbox, vertical_segments, horizontal_segments, pdf_scalers
+        )
 
     def _generate_columns_and_rows(self, table_idx, tk):
         # select elements which lie within table_bbox
         t_bbox = {}
         v_s, h_s = segments_in_bbox(
-            tk, self.vertical_segments, self.horizontal_segments)
+            tk, self.vertical_segments, self.horizontal_segments
-        t_bbox['horizontal'] = text_in_bbox(tk, self.horizontal_text)
+        )
-        t_bbox['vertical'] = text_in_bbox(tk, self.vertical_text)
+        t_bbox["horizontal"] = text_in_bbox(tk, self.horizontal_text)
-        self.t_bbox = t_bbox
+        t_bbox["vertical"] = text_in_bbox(tk, self.vertical_text)
 
-        for direction in t_bbox:
+        t_bbox["horizontal"].sort(key=lambda x: (-x.y0, x.x0))
-            t_bbox[direction].sort(key=lambda x: (-x.y0, x.x0))
+        t_bbox["vertical"].sort(key=lambda x: (x.x0, -x.y0))
+
+        self.t_bbox = t_bbox
 
         cols, rows = zip(*self.table_bbox[tk])
         cols, rows = list(cols), list(rows)
         cols.extend([tk[0], tk[2]])
         rows.extend([tk[1], tk[3]])
         # sort horizontal and vertical segments
-        cols = merge_close_lines(
+        cols = merge_close_lines(sorted(cols), line_tol=self.line_tol)
-            sorted(cols), line_close_tol=self.line_close_tol)
+        rows = merge_close_lines(sorted(rows, reverse=True), line_tol=self.line_tol)
-        rows = merge_close_lines(
-            sorted(rows, reverse=True), line_close_tol=self.line_close_tol)
         # make grid using x and y coord of shortlisted rows and cols
-        cols = [(cols[i], cols[i + 1])
+        cols = [(cols[i], cols[i + 1]) for i in range(0, len(cols) - 1)]
-                for i in range(0, len(cols) - 1)]
+        rows = [(rows[i], rows[i + 1]) for i in range(0, len(rows) - 1)]
-        rows = [(rows[i], rows[i + 1])
-                for i in range(0, len(rows) - 1)]
 
         return cols, rows, v_s, h_s
 
     def _generate_table(self, table_idx, cols, rows, **kwargs):
-        v_s = kwargs.get('v_s')
+        v_s = kwargs.get("v_s")
-        h_s = kwargs.get('h_s')
+        h_s = kwargs.get("h_s")
         if v_s is None or h_s is None:
-            raise ValueError('No segments found on {}'.format(self.rootname))
+            raise ValueError("No segments found on {}".format(self.rootname))
 
         table = Table(cols, rows)
         # set table edges to True using ver+hor lines
-        table = table.set_edges(v_s, h_s, joint_close_tol=self.joint_close_tol)
+        table = table.set_edges(v_s, h_s, joint_tol=self.joint_tol)
         # set table border edges to True
         table = table.set_border()
         # set spanning cells to True
         table = table.set_span()
 
         pos_errors = []
-        for direction in self.t_bbox:
+        # TODO: have a single list in place of two directional ones?
+        # sorted on x-coordinate based on reading order i.e. LTR or RTL
+        for direction in ["vertical", "horizontal"]:
             for t in self.t_bbox[direction]:
                 indices, error = get_table_index(
-                    table, t, direction, split_text=self.split_text,
+                    table,
-                    flag_size=self.flag_size)
+                    t,
+                    direction,
+                    split_text=self.split_text,
+                    flag_size=self.flag_size,
+                    strip_text=self.strip_text,
+                )
                 if indices[:2] != (-1, -1):
                     pos_errors.append(error)
-                    indices = Lattice._reduce_index(table, indices, shift_text=self.shift_text)
+                    indices = Lattice._reduce_index(
+                        table, indices, shift_text=self.shift_text
+                    )
                     for r_idx, c_idx, text in indices:
                         table.cells[r_idx][c_idx].text = text
         accuracy = compute_accuracy([[100, pos_errors]])
@@ -310,11 +384,11 @@ class Lattice(BaseParser):
         table.shape = table.df.shape
 
         whitespace = compute_whitespace(data)
-        table.flavor = 'lattice'
+        table.flavor = "lattice"
         table.accuracy = accuracy
         table.whitespace = whitespace
         table.order = table_idx + 1
-        table.page = int(os.path.basename(self.rootname).replace('page-', ''))
+        table.page = int(os.path.basename(self.rootname).replace("page-", ""))
 
         # for plotting
         _text = []
@@ -323,27 +397,39 @@ class Lattice(BaseParser):
         table._text = _text
         table._image = (self.image, self.table_bbox_unscaled)
         table._segments = (self.vertical_segments, self.horizontal_segments)
+        table._textedges = None
 
         return table
 
-    def extract_tables(self, filename):
+    def extract_tables(self, filename, suppress_stdout=False, layout_kwargs={}):
-        self._generate_layout(filename)
+        self._generate_layout(filename, layout_kwargs)
-        logger.info('Processing {}'.format(os.path.basename(self.rootname)))
+        if not suppress_stdout:
+            logger.info("Processing {}".format(os.path.basename(self.rootname)))
 
         if not self.horizontal_text:
-            warnings.warn("No tables found on {}".format(
+            if self.images:
-                os.path.basename(self.rootname)))
+                warnings.warn(
+                    "{} is image-based, camelot only works on"
+                    " text-based pages.".format(os.path.basename(self.rootname))
+                )
+            else:
+                warnings.warn(
+                    "No tables found on {}".format(os.path.basename(self.rootname))
+                )
             return []
 
-        self._generate_image()
+        self.backend.convert(self.filename, self.imagename)
+
         self._generate_table_bbox()
 
         _tables = []
         # sort tables based on y-coord
-        for table_idx, tk in enumerate(sorted(
+        for table_idx, tk in enumerate(
-                self.table_bbox.keys(), key=lambda x: x[1], reverse=True)):
+            sorted(self.table_bbox.keys(), key=lambda x: x[1], reverse=True)
+        ):
             cols, rows, v_s, h_s = self._generate_columns_and_rows(table_idx, tk)
             table = self._generate_table(table_idx, cols, rows, v_s=v_s, h_s=h_s)
+            table._bbox = tk
             _tables.append(table)
 
         return _tables

camelot/parsers/stream.py

@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 
-from __future__ import division
 import os
 import logging
 import warnings
@@ -9,12 +8,11 @@ import numpy as np
 import pandas as pd
 
 from .base import BaseParser
-from ..core import Table
+from ..core import TextEdges, Table
-from ..utils import (text_in_bbox, get_table_index, compute_accuracy,
+from ..utils import text_in_bbox, get_table_index, compute_accuracy, compute_whitespace
-                     compute_whitespace)
 
 
-logger = logging.getLogger('camelot')
+logger = logging.getLogger("camelot")
 
 
 class Stream(BaseParser):
@@ -26,7 +24,11 @@ class Stream(BaseParser):
 
     Parameters
     ----------
-    table_area : list, optional (default: None)
+    table_regions : list, optional (default: None)
+        List of page regions that may contain tables of the form x1,y1,x2,y2
+        where (x1, y1) -> left-top and (x2, y2) -> right-bottom
+        in PDF coordinate space.
+    table_areas : list, optional (default: None)
         List of table area strings of the form x1,y1,x2,y2
         where (x1, y1) -> left-top and (x2, y2) -> right-bottom
         in PDF coordinate space.
@@ -38,29 +40,43 @@ class Stream(BaseParser):
     flag_size : bool, optional (default: False)
         Flag text based on font size. Useful to detect
         super/subscripts. Adds <s></s> around flagged text.
-    row_close_tol : int, optional (default: 2)
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
+    edge_tol : int, optional (default: 50)
+        Tolerance parameter for extending textedges vertically.
+    row_tol : int, optional (default: 2)
         Tolerance parameter used to combine text vertically,
         to generate rows.
-    col_close_tol : int, optional (default: 0)
+    column_tol : int, optional (default: 0)
         Tolerance parameter used to combine text horizontally,
         to generate columns.
-    margins : tuple, optional (default: (1.0, 0.5, 0.1))
-        PDFMiner char_margin, line_margin and word_margin.
-
-        For more information, refer `PDFMiner docs <https://euske.github.io/pdfminer/>`_.
 
     """
-    def __init__(self, table_area=None, columns=None, split_text=False,
+
-                 flag_size=False, row_close_tol=2, col_close_tol=0,
+    def __init__(
-                 margins=(1.0, 0.5, 0.1), **kwargs):
+        self,
-        self.table_area = table_area
+        table_regions=None,
+        table_areas=None,
+        columns=None,
+        split_text=False,
+        flag_size=False,
+        strip_text="",
+        edge_tol=50,
+        row_tol=2,
+        column_tol=0,
+        **kwargs,
+    ):
+        self.table_regions = table_regions
+        self.table_areas = table_areas
         self.columns = columns
         self._validate_columns()
         self.split_text = split_text
         self.flag_size = flag_size
-        self.row_close_tol = row_close_tol
+        self.strip_text = strip_text
-        self.col_close_tol = col_close_tol
+        self.edge_tol = edge_tol
-        self.char_margin, self.line_margin, self.word_margin = margins
+        self.row_tol = row_tol
+        self.column_tol = column_tol
 
     @staticmethod
     def _text_bbox(t_bbox):
@@ -86,7 +102,7 @@ class Stream(BaseParser):
         return text_bbox
 
     @staticmethod
-    def _group_rows(text, row_close_tol=2):
+    def _group_rows(text, row_tol=2):
         """Groups PDFMiner text objects into rows vertically
         within a tolerance.
 
@@ -94,7 +110,7 @@ class Stream(BaseParser):
         ----------
         text : list
             List of PDFMiner text objects.
-        row_close_tol : int, optional (default: 2)
+        row_tol : int, optional (default: 2)
 
         Returns
         -------
@@ -105,22 +121,25 @@ class Stream(BaseParser):
         row_y = 0
         rows = []
         temp = []
+
         for t in text:
             # is checking for upright necessary?
             # if t.get_text().strip() and all([obj.upright for obj in t._objs if
             # type(obj) is LTChar]):
             if t.get_text().strip():
-                if not np.isclose(row_y, t.y0, atol=row_close_tol):
+                if not np.isclose(row_y, t.y0, atol=row_tol):
                     rows.append(sorted(temp, key=lambda t: t.x0))
                     temp = []
                     row_y = t.y0
                 temp.append(t)
+
         rows.append(sorted(temp, key=lambda t: t.x0))
-        __ = rows.pop(0)  # hacky
+        if len(rows) > 1:
+            __ = rows.pop(0)  # TODO: hacky
         return rows
 
     @staticmethod
-    def _merge_columns(l, col_close_tol=0):
+    def _merge_columns(l, column_tol=0):
         """Merges column boundaries horizontally if they overlap
         or lie within a tolerance.
 
@@ -128,7 +147,7 @@ class Stream(BaseParser):
         ----------
         l : list
             List of column x-coordinate tuples.
-        col_close_tol : int, optional (default: 0)
+        column_tol : int, optional (default: 0)
 
         Returns
         -------
@@ -142,17 +161,18 @@ class Stream(BaseParser):
                 merged.append(higher)
             else:
                 lower = merged[-1]
-                if col_close_tol >= 0:
+                if column_tol >= 0:
-                    if (higher[0] <= lower[1] or
+                    if higher[0] <= lower[1] or np.isclose(
-                            np.isclose(higher[0], lower[1], atol=col_close_tol)):
+                        higher[0], lower[1], atol=column_tol
+                    ):
                         upper_bound = max(lower[1], higher[1])
                         lower_bound = min(lower[0], higher[0])
                         merged[-1] = (lower_bound, upper_bound)
                     else:
                         merged.append(higher)
-                elif col_close_tol < 0:
+                elif column_tol < 0:
                     if higher[0] <= lower[1]:
-                        if np.isclose(higher[0], lower[1], atol=abs(col_close_tol)):
+                        if np.isclose(higher[0], lower[1], atol=abs(column_tol)):
                             merged.append(higher)
                         else:
                             upper_bound = max(lower[1], higher[1])
@@ -179,17 +199,18 @@ class Stream(BaseParser):
             List of continuous row y-coordinate tuples.
 
         """
-        row_mids = [sum([(t.y0 + t.y1) / 2 for t in r]) / len(r)
+        row_mids = [
-                    if len(r) > 0 else 0 for r in rows_grouped]
+            sum([(t.y0 + t.y1) / 2 for t in r]) / len(r) if len(r) > 0 else 0
+            for r in rows_grouped
+        ]
         rows = [(row_mids[i] + row_mids[i - 1]) / 2 for i in range(1, len(row_mids))]
         rows.insert(0, text_y_max)
         rows.append(text_y_min)
-        rows = [(rows[i], rows[i + 1])
+        rows = [(rows[i], rows[i + 1]) for i in range(0, len(rows) - 1)]
-                for i in range(0, len(rows) - 1)]
         return rows
 
     @staticmethod
-    def _add_columns(cols, text, row_close_tol):
+    def _add_columns(cols, text, row_tol):
         """Adds columns to existing list by taking into account
         the text that lies outside the current column x-coordinates.
 
@@ -208,10 +229,11 @@ class Stream(BaseParser):
 
         """
         if text:
-            text = Stream._group_rows(text, row_close_tol=row_close_tol)
+            text = Stream._group_rows(text, row_tol=row_tol)
             elements = [len(r) for r in text]
-            new_cols = [(t.x0, t.x1)
+            new_cols = [
-                        for r in text if len(r) == max(elements) for t in r]
+                (t.x0, t.x1) for r in text if len(r) == max(elements) for t in r
+            ]
             cols.extend(Stream._merge_columns(sorted(new_cols)))
         return cols
 
@@ -236,42 +258,80 @@ class Stream(BaseParser):
         cols = [(cols[i][0] + cols[i - 1][1]) / 2 for i in range(1, len(cols))]
         cols.insert(0, text_x_min)
         cols.append(text_x_max)
-        cols = [(cols[i], cols[i + 1])
+        cols = [(cols[i], cols[i + 1]) for i in range(0, len(cols) - 1)]
-                for i in range(0, len(cols) - 1)]
         return cols
 
     def _validate_columns(self):
-        if self.table_area is not None and self.columns is not None:
+        if self.table_areas is not None and self.columns is not None:
-            if len(self.table_area) != len(self.columns):
+            if len(self.table_areas) != len(self.columns):
-                raise ValueError("Length of table_area and columns"
+                raise ValueError("Length of table_areas and columns" " should be equal")
-                                 " should be equal")
+
+    def _nurminen_table_detection(self, textlines):
+        """A general implementation of the table detection algorithm
+        described by Anssi Nurminen's master's thesis.
+        Link: https://dspace.cc.tut.fi/dpub/bitstream/handle/123456789/21520/Nurminen.pdf?sequence=3
+
+        Assumes that tables are situated relatively far apart
+        vertically.
+        """
+        # TODO: add support for arabic text #141
+        # sort textlines in reading order
+        textlines.sort(key=lambda x: (-x.y0, x.x0))
+        textedges = TextEdges(edge_tol=self.edge_tol)
+        # generate left, middle and right textedges
+        textedges.generate(textlines)
+        # select relevant edges
+        relevant_textedges = textedges.get_relevant()
+        self.textedges.extend(relevant_textedges)
+        # guess table areas using textlines and relevant edges
+        table_bbox = textedges.get_table_areas(textlines, relevant_textedges)
+        # treat whole page as table area if no table areas found
+        if not len(table_bbox):
+            table_bbox = {(0, 0, self.pdf_width, self.pdf_height): None}
+
+        return table_bbox
 
     def _generate_table_bbox(self):
-        if self.table_area is not None:
+        self.textedges = []
+        if self.table_areas is None:
+            hor_text = self.horizontal_text
+            if self.table_regions is not None:
+                # filter horizontal text
+                hor_text = []
+                for region in self.table_regions:
+                    x1, y1, x2, y2 = region.split(",")
+                    x1 = float(x1)
+                    y1 = float(y1)
+                    x2 = float(x2)
+                    y2 = float(y2)
+                    region_text = text_in_bbox((x1, y2, x2, y1), self.horizontal_text)
+                    hor_text.extend(region_text)
+            # find tables based on nurminen's detection algorithm
+            table_bbox = self._nurminen_table_detection(hor_text)
+        else:
             table_bbox = {}
-            for area in self.table_area:
+            for area in self.table_areas:
                 x1, y1, x2, y2 = area.split(",")
                 x1 = float(x1)
                 y1 = float(y1)
                 x2 = float(x2)
                 y2 = float(y2)
                 table_bbox[(x1, y2, x2, y1)] = None
-        else:
-            table_bbox = {(0, 0, self.pdf_width, self.pdf_height): None}
         self.table_bbox = table_bbox
 
     def _generate_columns_and_rows(self, table_idx, tk):
         # select elements which lie within table_bbox
         t_bbox = {}
-        t_bbox['horizontal'] = text_in_bbox(tk, self.horizontal_text)
+        t_bbox["horizontal"] = text_in_bbox(tk, self.horizontal_text)
-        t_bbox['vertical'] = text_in_bbox(tk, self.vertical_text)
+        t_bbox["vertical"] = text_in_bbox(tk, self.vertical_text)
+
+        t_bbox["horizontal"].sort(key=lambda x: (-x.y0, x.x0))
+        t_bbox["vertical"].sort(key=lambda x: (x.x0, -x.y0))
+
         self.t_bbox = t_bbox
 
-        for direction in self.t_bbox:
-            self.t_bbox[direction].sort(key=lambda x: (-x.y0, x.x0))
-
         text_x_min, text_y_min, text_x_max, text_y_max = self._text_bbox(self.t_bbox)
-        rows_grouped = self._group_rows(self.t_bbox['horizontal'], row_close_tol=self.row_close_tol)
+        rows_grouped = self._group_rows(self.t_bbox["horizontal"], row_tol=self.row_tol)
         rows = self._join_rows(rows_grouped, text_y_max, text_y_min)
         elements = [len(r) for r in rows_grouped]
 
@@ -280,43 +340,74 @@ class Stream(BaseParser):
             # take (0, pdf_width) by default
             # similar to else condition
             # len can't be 1
-            cols = self.columns[table_idx].split(',')
+            cols = self.columns[table_idx].split(",")
             cols = [float(c) for c in cols]
             cols.insert(0, text_x_min)
             cols.append(text_x_max)
             cols = [(cols[i], cols[i + 1]) for i in range(0, len(cols) - 1)]
         else:
-            ncols = max(set(elements), key=elements.count)
+            # calculate mode of the list of number of elements in
-            if ncols == 1:
+            # each row to guess the number of columns
-                warnings.warn("No tables found on {}".format(
+            if not len(elements):
-                    os.path.basename(self.rootname)))
+                cols = [(text_x_min, text_x_max)]
-            cols = [(t.x0, t.x1) for r in rows_grouped if len(r) == ncols for t in r]
+            else:
-            cols = self._merge_columns(sorted(cols), col_close_tol=self.col_close_tol)
+                ncols = max(set(elements), key=elements.count)
-            inner_text = []
+                if ncols == 1:
-            for i in range(1, len(cols)):
+                    # if mode is 1, the page usually contains not tables
-                left = cols[i - 1][1]
+                    # but there can be cases where the list can be skewed,
-                right = cols[i][0]
+                    # try to remove all 1s from list in this case and
-                inner_text.extend([t for direction in self.t_bbox
+                    # see if the list contains elements, if yes, then use
-                                     for t in self.t_bbox[direction]
+                    # the mode after removing 1s
-                                     if t.x0 > left and t.x1 < right])
+                    elements = list(filter(lambda x: x != 1, elements))
-            outer_text = [t for direction in self.t_bbox
+                    if len(elements):
+                        ncols = max(set(elements), key=elements.count)
+                    else:
+                        warnings.warn(f"No tables found in table area {table_idx + 1}")
+                cols = [
+                    (t.x0, t.x1) for r in rows_grouped if len(r) == ncols for t in r
+                ]
+                cols = self._merge_columns(sorted(cols), column_tol=self.column_tol)
+                inner_text = []
+                for i in range(1, len(cols)):
+                    left = cols[i - 1][1]
+                    right = cols[i][0]
+                    inner_text.extend(
+                        [
+                            t
+                            for direction in self.t_bbox
                             for t in self.t_bbox[direction]
-                            if t.x0 > cols[-1][1] or t.x1 < cols[0][0]]
+                            if t.x0 > left and t.x1 < right
-            inner_text.extend(outer_text)
+                        ]
-            cols = self._add_columns(cols, inner_text, self.row_close_tol)
+                    )
-            cols = self._join_columns(cols, text_x_min, text_x_max)
+                outer_text = [
+                    t
+                    for direction in self.t_bbox
+                    for t in self.t_bbox[direction]
+                    if t.x0 > cols[-1][1] or t.x1 < cols[0][0]
+                ]
+                inner_text.extend(outer_text)
+                cols = self._add_columns(cols, inner_text, self.row_tol)
+                cols = self._join_columns(cols, text_x_min, text_x_max)
 
         return cols, rows
 
     def _generate_table(self, table_idx, cols, rows, **kwargs):
         table = Table(cols, rows)
         table = table.set_all_edges()
+
         pos_errors = []
-        for direction in self.t_bbox:
+        # TODO: have a single list in place of two directional ones?
+        # sorted on x-coordinate based on reading order i.e. LTR or RTL
+        for direction in ["vertical", "horizontal"]:
             for t in self.t_bbox[direction]:
                 indices, error = get_table_index(
-                    table, t, direction, split_text=self.split_text,
+                    table,
-                    flag_size=self.flag_size)
+                    t,
+                    direction,
+                    split_text=self.split_text,
+                    flag_size=self.flag_size,
+                    strip_text=self.strip_text,
+                )
                 if indices[:2] != (-1, -1):
                     pos_errors.append(error)
                     for r_idx, c_idx, text in indices:
@@ -328,11 +419,11 @@ class Stream(BaseParser):
         table.shape = table.df.shape
 
         whitespace = compute_whitespace(data)
-        table.flavor = 'stream'
+        table.flavor = "stream"
         table.accuracy = accuracy
         table.whitespace = whitespace
         table.order = table_idx + 1
-        table.page = int(os.path.basename(self.rootname).replace('page-', ''))
+        table.page = int(os.path.basename(self.rootname).replace("page-", ""))
 
         # for plotting
         _text = []
@@ -341,26 +432,37 @@ class Stream(BaseParser):
         table._text = _text
         table._image = None
         table._segments = None
+        table._textedges = self.textedges
 
         return table
 
-    def extract_tables(self, filename):
+    def extract_tables(self, filename, suppress_stdout=False, layout_kwargs={}):
-        self._generate_layout(filename)
+        self._generate_layout(filename, layout_kwargs)
-        logger.info('Processing {}'.format(os.path.basename(self.rootname)))
+        base_filename = os.path.basename(self.rootname)
+
+        if not suppress_stdout:
+            logger.info(f"Processing {base_filename}")
 
         if not self.horizontal_text:
-            warnings.warn("No tables found on {}".format(
+            if self.images:
-                os.path.basename(self.rootname)))
+                warnings.warn(
+                    f"{base_filename} is image-based, camelot only works on"
+                    " text-based pages."
+                )
+            else:
+                warnings.warn(f"No tables found on {base_filename}")
             return []
 
         self._generate_table_bbox()
 
         _tables = []
         # sort tables based on y-coord
-        for table_idx, tk in enumerate(sorted(
+        for table_idx, tk in enumerate(
-                self.table_bbox.keys(), key=lambda x: x[1], reverse=True)):
+            sorted(self.table_bbox.keys(), key=lambda x: x[1], reverse=True)
+        ):
             cols, rows = self._generate_columns_and_rows(table_idx, tk)
             table = self._generate_table(table_idx, cols, rows)
+            table._bbox = tk
             _tables.append(table)
 
         return _tables

camelot/plotting.py

@@ -1,108 +1,225 @@
-import cv2
+# -*- coding: utf-8 -*-
-import matplotlib.pyplot as plt
+
-import matplotlib.patches as patches
+try:
+    import matplotlib.pyplot as plt
+    import matplotlib.patches as patches
+except ImportError:
+    _HAS_MPL = False
+else:
+    _HAS_MPL = True
 
 
-def plot_text(text):
+class PlotMethods(object):
-    """Generates a plot for all text present on the PDF page.
+    def __call__(self, table, kind="text", filename=None):
+        """Plot elements found on PDF page based on kind
+        specified, useful for debugging and playing with different
+        parameters to get the best output.
 
-    Parameters
+        Parameters
-    ----------
+        ----------
-    text : list
+        table: camelot.core.Table
+            A Camelot Table.
+        kind : str, optional (default: 'text')
+            {'text', 'grid', 'contour', 'joint', 'line'}
+            The element type for which a plot should be generated.
+        filepath: str, optional (default: None)
+            Absolute path for saving the generated plot.
 
-    """
+        Returns
-    fig = plt.figure()
+        -------
-    ax = fig.add_subplot(111, aspect='equal')
+        fig : matplotlib.fig.Figure
-    xs, ys = [], []
+
-    for t in text:
+        """
-        xs.extend([t[0], t[2]])
+        if not _HAS_MPL:
-        ys.extend([t[1], t[3]])
+            raise ImportError("matplotlib is required for plotting.")
-        ax.add_patch(
+
-            patches.Rectangle(
+        if table.flavor == "lattice" and kind in ["textedge"]:
-                (t[0], t[1]),
+            raise NotImplementedError(f"Lattice flavor does not support kind='{kind}'")
-                t[2] - t[0],
+        elif table.flavor == "stream" and kind in ["joint", "line"]:
-                t[3] - t[1]
+            raise NotImplementedError(f"Stream flavor does not support kind='{kind}'")
+
+        plot_method = getattr(self, kind)
+        fig = plot_method(table)
+
+        if filename is not None:
+            fig.savefig(filename)
+            return None
+
+        return fig
+
+    def text(self, table):
+        """Generates a plot for all text elements present
+        on the PDF page.
+
+        Parameters
+        ----------
+        table : camelot.core.Table
+
+        Returns
+        -------
+        fig : matplotlib.fig.Figure
+
+        """
+        fig = plt.figure()
+        ax = fig.add_subplot(111, aspect="equal")
+        xs, ys = [], []
+        for t in table._text:
+            xs.extend([t[0], t[2]])
+            ys.extend([t[1], t[3]])
+            ax.add_patch(patches.Rectangle((t[0], t[1]), t[2] - t[0], t[3] - t[1]))
+        ax.set_xlim(min(xs) - 10, max(xs) + 10)
+        ax.set_ylim(min(ys) - 10, max(ys) + 10)
+        return fig
+
+    def grid(self, table):
+        """Generates a plot for the detected table grids
+        on the PDF page.
+
+        Parameters
+        ----------
+        table : camelot.core.Table
+
+        Returns
+        -------
+        fig : matplotlib.fig.Figure
+
+        """
+        fig = plt.figure()
+        ax = fig.add_subplot(111, aspect="equal")
+        for row in table.cells:
+            for cell in row:
+                if cell.left:
+                    ax.plot([cell.lb[0], cell.lt[0]], [cell.lb[1], cell.lt[1]])
+                if cell.right:
+                    ax.plot([cell.rb[0], cell.rt[0]], [cell.rb[1], cell.rt[1]])
+                if cell.top:
+                    ax.plot([cell.lt[0], cell.rt[0]], [cell.lt[1], cell.rt[1]])
+                if cell.bottom:
+                    ax.plot([cell.lb[0], cell.rb[0]], [cell.lb[1], cell.rb[1]])
+        return fig
+
+    def contour(self, table):
+        """Generates a plot for all table boundaries present
+        on the PDF page.
+
+        Parameters
+        ----------
+        table : camelot.core.Table
+
+        Returns
+        -------
+        fig : matplotlib.fig.Figure
+
+        """
+        try:
+            img, table_bbox = table._image
+            _FOR_LATTICE = True
+        except TypeError:
+            img, table_bbox = (None, {table._bbox: None})
+            _FOR_LATTICE = False
+        fig = plt.figure()
+        ax = fig.add_subplot(111, aspect="equal")
+
+        xs, ys = [], []
+        if not _FOR_LATTICE:
+            for t in table._text:
+                xs.extend([t[0], t[2]])
+                ys.extend([t[1], t[3]])
+                ax.add_patch(
+                    patches.Rectangle(
+                        (t[0], t[1]), t[2] - t[0], t[3] - t[1], color="blue"
+                    )
+                )
+
+        for t in table_bbox.keys():
+            ax.add_patch(
+                patches.Rectangle(
+                    (t[0], t[1]), t[2] - t[0], t[3] - t[1], fill=False, color="red"
+                )
             )
-        )
+            if not _FOR_LATTICE:
-    ax.set_xlim(min(xs) - 10, max(xs) + 10)
+                xs.extend([t[0], t[2]])
-    ax.set_ylim(min(ys) - 10, max(ys) + 10)
+                ys.extend([t[1], t[3]])
-    plt.show()
+                ax.set_xlim(min(xs) - 10, max(xs) + 10)
+                ax.set_ylim(min(ys) - 10, max(ys) + 10)
 
+        if _FOR_LATTICE:
+            ax.imshow(img)
+        return fig
 
-def plot_table(table):
+    def textedge(self, table):
-    """Generates a plot for the table.
+        """Generates a plot for relevant textedges.
 
-    Parameters
+        Parameters
-    ----------
+        ----------
-    table : camelot.core.Table
+        table : camelot.core.Table
 
-    """
+        Returns
-    for row in table.cells:
+        -------
-        for cell in row:
+        fig : matplotlib.fig.Figure
-            if cell.left:
-                plt.plot([cell.lb[0], cell.lt[0]],
-                         [cell.lb[1], cell.lt[1]])
-            if cell.right:
-                plt.plot([cell.rb[0], cell.rt[0]],
-                         [cell.rb[1], cell.rt[1]])
-            if cell.top:
-                plt.plot([cell.lt[0], cell.rt[0]],
-                         [cell.lt[1], cell.rt[1]])
-            if cell.bottom:
-                plt.plot([cell.lb[0], cell.rb[0]],
-                         [cell.lb[1], cell.rb[1]])
-    plt.show()
 
+        """
+        fig = plt.figure()
+        ax = fig.add_subplot(111, aspect="equal")
+        xs, ys = [], []
+        for t in table._text:
+            xs.extend([t[0], t[2]])
+            ys.extend([t[1], t[3]])
+            ax.add_patch(
+                patches.Rectangle((t[0], t[1]), t[2] - t[0], t[3] - t[1], color="blue")
+            )
+        ax.set_xlim(min(xs) - 10, max(xs) + 10)
+        ax.set_ylim(min(ys) - 10, max(ys) + 10)
 
-def plot_contour(image):
+        for te in table._textedges:
-    """Generates a plot for all table boundaries present on the
+            ax.plot([te.x, te.x], [te.y0, te.y1])
-    PDF page.
 
-    Parameters
+        return fig
-    ----------
-    image : tuple
 
-    """
+    def joint(self, table):
-    img, table_bbox = image
+        """Generates a plot for all line intersections present
-    for t in table_bbox.keys():
+        on the PDF page.
-        cv2.rectangle(img, (t[0], t[1]),
-                      (t[2], t[3]), (255, 0, 0), 20)
-    plt.imshow(img)
-    plt.show()
 
+        Parameters
+        ----------
+        table : camelot.core.Table
 
-def plot_joint(image):
+        Returns
-    """Generates a plot for all line intersections present on the
+        -------
-    PDF page.
+        fig : matplotlib.fig.Figure
 
-    Parameters
+        """
-    ----------
+        img, table_bbox = table._image
-    image : tuple
+        fig = plt.figure()
+        ax = fig.add_subplot(111, aspect="equal")
+        x_coord = []
+        y_coord = []
+        for k in table_bbox.keys():
+            for coord in table_bbox[k]:
+                x_coord.append(coord[0])
+                y_coord.append(coord[1])
+        ax.plot(x_coord, y_coord, "ro")
+        ax.imshow(img)
+        return fig
 
-    """
+    def line(self, table):
-    img, table_bbox = image
+        """Generates a plot for all line segments present
-    x_coord = []
+        on the PDF page.
-    y_coord = []
-    for k in table_bbox.keys():
-        for coord in table_bbox[k]:
-            x_coord.append(coord[0])
-            y_coord.append(coord[1])
-    plt.plot(x_coord, y_coord, 'ro')
-    plt.imshow(img)
-    plt.show()
 
+        Parameters
+        ----------
+        table : camelot.core.Table
 
-def plot_line(segments):
+        Returns
-    """Generates a plot for all line segments present on the PDF page.
+        -------
+        fig : matplotlib.fig.Figure
 
-    Parameters
+        """
-    ----------
+        fig = plt.figure()
-    segments : tuple
+        ax = fig.add_subplot(111, aspect="equal")
-
+        vertical, horizontal = table._segments
-    """
+        for v in vertical:
-    vertical, horizontal = segments
+            ax.plot([v[0], v[2]], [v[1], v[3]])
-    for v in vertical:
+        for h in horizontal:
-        plt.plot([v[0], v[2]], [v[1], v[3]])
+            ax.plot([h[0], h[2]], [h[1], h[3]])
-    for h in horizontal:
+        return fig
-        plt.plot([h[0], h[2]], [h[1], h[3]])
-    plt.show()

camelot/utils.py

@@ -1,12 +1,16 @@
-from __future__ import division
+# -*- coding: utf-8 -*-
+
+import os
+import re
+import random
 import shutil
+import string
 import tempfile
 import warnings
 from itertools import groupby
 from operator import itemgetter
 
 import numpy as np
-
 from pdfminer.pdfparser import PDFParser
 from pdfminer.pdfdocument import PDFDocument
 from pdfminer.pdfpage import PDFPage
@@ -14,43 +18,112 @@ from pdfminer.pdfpage import PDFTextExtractionNotAllowed
 from pdfminer.pdfinterp import PDFResourceManager
 from pdfminer.pdfinterp import PDFPageInterpreter
 from pdfminer.converter import PDFPageAggregator
-from pdfminer.layout import (LAParams, LTAnno, LTChar, LTTextLineHorizontal,
+from pdfminer.layout import (
-                             LTTextLineVertical)
+    LAParams,
+    LTAnno,
+    LTChar,
+    LTTextLineHorizontal,
+    LTTextLineVertical,
+    LTImage,
+)
+
+from urllib.request import Request, urlopen
+from urllib.parse import urlparse as parse_url
+from urllib.parse import uses_relative, uses_netloc, uses_params
 
 
-stream_kwargs = [
+_VALID_URLS = set(uses_relative + uses_netloc + uses_params)
-    'columns',
+_VALID_URLS.discard("")
-    'row_close_tol',
+
-    'col_close_tol'
+
-]
+# https://github.com/pandas-dev/pandas/blob/master/pandas/io/common.py
+def is_url(url):
+    """Check to see if a URL has a valid protocol.
+
+    Parameters
+    ----------
+    url : str or unicode
+
+    Returns
+    -------
+    isurl : bool
+        If url has a valid protocol return True otherwise False.
+
+    """
+    try:
+        return parse_url(url).scheme in _VALID_URLS
+    except Exception:
+        return False
+
+
+def random_string(length):
+    ret = ""
+    while length:
+        ret += random.choice(
+            string.digits + string.ascii_lowercase + string.ascii_uppercase
+        )
+        length -= 1
+    return ret
+
+
+def download_url(url):
+    """Download file from specified URL.
+
+    Parameters
+    ----------
+    url : str or unicode
+
+    Returns
+    -------
+    filepath : str or unicode
+        Temporary filepath.
+
+    """
+    filename = f"{random_string(6)}.pdf"
+    with tempfile.NamedTemporaryFile("wb", delete=False) as f:
+        headers = {"User-Agent": "Mozilla/5.0"}
+        request = Request(url, None, headers)
+        obj = urlopen(request)
+        content_type = obj.info().get_content_type()
+        if content_type != "application/pdf":
+            raise NotImplementedError("File format not supported")
+        f.write(obj.read())
+    filepath = os.path.join(os.path.dirname(f.name), filename)
+    shutil.move(f.name, filepath)
+    return filepath
+
+
+stream_kwargs = ["columns", "edge_tol", "row_tol", "column_tol"]
 lattice_kwargs = [
-    'process_background',
+    "process_background",
-    'line_size_scaling',
+    "line_scale",
-    'copy_text',
+    "copy_text",
-    'shift_text',
+    "shift_text",
-    'line_close_tol',
+    "line_tol",
-    'joint_close_tol',
+    "joint_tol",
-    'threshold_blocksize',
+    "threshold_blocksize",
-    'threshold_constant',
+    "threshold_constant",
-    'iterations'
+    "iterations",
+    "resolution",
 ]
 
 
-def validate_input(kwargs, flavor='lattice'):
+def validate_input(kwargs, flavor="lattice"):
     def check_intersection(parser_kwargs, input_kwargs):
         isec = set(parser_kwargs).intersection(set(input_kwargs.keys()))
         if isec:
-            raise ValueError("{} cannot be used with flavor='{}'".format(
+            raise ValueError(
-                             ",".join(sorted(isec)), flavor))
+                f"{','.join(sorted(isec))} cannot be used with flavor='{flavor}'"
+            )
 
-    if flavor == 'lattice':
+    if flavor == "lattice":
         check_intersection(stream_kwargs, kwargs)
     else:
         check_intersection(lattice_kwargs, kwargs)
 
 
-def remove_extra(kwargs, flavor='lattice'):
+def remove_extra(kwargs, flavor="lattice"):
-    if flavor == 'lattice':
+    if flavor == "lattice":
         for key in kwargs.keys():
             if key in stream_kwargs:
                 kwargs.pop(key)
@@ -180,29 +253,33 @@ def scale_image(tables, v_segments, h_segments, factors):
     v_segments_new = []
     for v in v_segments:
         x1, x2 = scale(v[0], scaling_factor_x), scale(v[2], scaling_factor_x)
-        y1, y2 = scale(abs(translate(-img_y, v[1])), scaling_factor_y), scale(
+        y1, y2 = (
-            abs(translate(-img_y, v[3])), scaling_factor_y)
+            scale(abs(translate(-img_y, v[1])), scaling_factor_y),
+            scale(abs(translate(-img_y, v[3])), scaling_factor_y),
+        )
         v_segments_new.append((x1, y1, x2, y2))
 
     h_segments_new = []
     for h in h_segments:
         x1, x2 = scale(h[0], scaling_factor_x), scale(h[2], scaling_factor_x)
-        y1, y2 = scale(abs(translate(-img_y, h[1])), scaling_factor_y), scale(
+        y1, y2 = (
-            abs(translate(-img_y, h[3])), scaling_factor_y)
+            scale(abs(translate(-img_y, h[1])), scaling_factor_y),
+            scale(abs(translate(-img_y, h[3])), scaling_factor_y),
+        )
         h_segments_new.append((x1, y1, x2, y2))
 
     return tables_new, v_segments_new, h_segments_new
 
 
-def get_rotation(lttextlh, lttextlv, ltchar):
+def get_rotation(chars, horizontal_text, vertical_text):
     """Detects if text in table is rotated or not using the current
     transformation matrix (CTM) and returns its orientation.
 
     Parameters
     ----------
-    lttextlh : list
+    horizontal_text : list
         List of PDFMiner LTTextLineHorizontal objects.
-    lttextlv : list
+    vertical_text : list
         List of PDFMiner LTTextLineVertical objects.
     ltchar : list
         List of PDFMiner LTChar objects.
@@ -215,13 +292,13 @@ def get_rotation(lttextlh, lttextlv, ltchar):
         rotated 90 degree clockwise.
 
     """
-    rotation = ''
+    rotation = ""
-    hlen = len([t for t in lttextlh if t.get_text().strip()])
+    hlen = len([t for t in horizontal_text if t.get_text().strip()])
-    vlen = len([t for t in lttextlv if t.get_text().strip()])
+    vlen = len([t for t in vertical_text if t.get_text().strip()])
     if hlen < vlen:
-        clockwise = sum(t.matrix[1] < 0 and t.matrix[2] > 0 for t in ltchar)
+        clockwise = sum(t.matrix[1] < 0 and t.matrix[2] > 0 for t in chars)
-        anticlockwise = sum(t.matrix[1] > 0 and t.matrix[2] < 0 for t in ltchar)
+        anticlockwise = sum(t.matrix[1] > 0 and t.matrix[2] < 0 for t in chars)
-        rotation = 'anticlockwise' if clockwise < anticlockwise else 'clockwise'
+        rotation = "anticlockwise" if clockwise < anticlockwise else "clockwise"
     return rotation
 
 
@@ -249,10 +326,16 @@ def segments_in_bbox(bbox, v_segments, h_segments):
     """
     lb = (bbox[0], bbox[1])
     rt = (bbox[2], bbox[3])
-    v_s = [v for v in v_segments if v[1] > lb[1] - 2 and
+    v_s = [
-           v[3] < rt[1] + 2 and lb[0] - 2 <= v[0] <= rt[0] + 2]
+        v
-    h_s = [h for h in h_segments if h[0] > lb[0] - 2 and
+        for v in v_segments
-           h[2] < rt[0] + 2 and lb[1] - 2 <= h[1] <= rt[1] + 2]
+        if v[1] > lb[1] - 2 and v[3] < rt[1] + 2 and lb[0] - 2 <= v[0] <= rt[0] + 2
+    ]
+    h_s = [
+        h
+        for h in h_segments
+        if h[0] > lb[0] - 2 and h[2] < rt[0] + 2 and lb[1] - 2 <= h[1] <= rt[1] + 2
+    ]
     return v_s, h_s
 
 
@@ -263,32 +346,125 @@ def text_in_bbox(bbox, text):
     ----------
     bbox : tuple
         Tuple (x1, y1, x2, y2) representing a bounding box where
-        (x1, y1) -> lb and (x2, y2) -> rt in PDFMiner coordinate
+        (x1, y1) -> lb and (x2, y2) -> rt in the PDF coordinate
         space.
     text : List of PDFMiner text objects.
 
     Returns
     -------
     t_bbox : list
-        List of PDFMiner text objects that lie inside table.
+        List of PDFMiner text objects that lie inside table, discarding the overlapping ones
 
     """
     lb = (bbox[0], bbox[1])
     rt = (bbox[2], bbox[3])
-    t_bbox = [t for t in text if lb[0] - 2 <= (t.x0 + t.x1) / 2.0
+    t_bbox = [
-              <= rt[0] + 2 and lb[1] - 2 <= (t.y0 + t.y1) / 2.0
+        t
-              <= rt[1] + 2]
+        for t in text
-    return t_bbox
+        if lb[0] - 2 <= (t.x0 + t.x1) / 2.0 <= rt[0] + 2
+        and lb[1] - 2 <= (t.y0 + t.y1) / 2.0 <= rt[1] + 2
+    ]
+
+    # Avoid duplicate text by discarding overlapping boxes
+    rest = {t for t in t_bbox}
+    for ba in t_bbox:
+        for bb in rest.copy():
+            if ba == bb:
+                continue
+            if bbox_intersect(ba, bb):
+                # if the intersection is larger than 80% of ba's size, we keep the longest
+                if (bbox_intersection_area(ba, bb) / bbox_area(ba)) > 0.8:
+                    if bbox_longer(bb, ba):
+                        rest.discard(ba)
+    unique_boxes = list(rest)
+
+    return unique_boxes
 
 
-def merge_close_lines(ar, line_close_tol=2):
+def bbox_intersection_area(ba, bb) -> float:
+    """Returns area of the intersection of the bounding boxes of two PDFMiner objects.
+
+    Parameters
+    ----------
+    ba : PDFMiner text object
+    bb : PDFMiner text object
+
+    Returns
+    -------
+    intersection_area : float
+        Area of the intersection of the bounding boxes of both objects
+
+    """
+    x_left = max(ba.x0, bb.x0)
+    y_top = min(ba.y1, bb.y1)
+    x_right = min(ba.x1, bb.x1)
+    y_bottom = max(ba.y0, bb.y0)
+
+    if x_right < x_left or y_bottom > y_top:
+        return 0.0
+
+    intersection_area = (x_right - x_left) * (y_top - y_bottom)
+    return intersection_area
+
+
+def bbox_area(bb) -> float:
+    """Returns area of the bounding box of a PDFMiner object.
+
+    Parameters
+    ----------
+    bb : PDFMiner text object
+
+    Returns
+    -------
+    area : float
+        Area of the bounding box of the object
+
+    """
+    return (bb.x1 - bb.x0) * (bb.y1 - bb.y0)
+
+
+def bbox_intersect(ba, bb) -> bool:
+    """Returns True if the bounding boxes of two PDFMiner objects intersect.
+
+    Parameters
+    ----------
+    ba : PDFMiner text object
+    bb : PDFMiner text object
+
+    Returns
+    -------
+    overlaps : bool
+        True if the bounding boxes intersect
+
+    """
+    return ba.x1 >= bb.x0 and bb.x1 >= ba.x0 and ba.y1 >= bb.y0 and bb.y1 >= ba.y0
+
+
+def bbox_longer(ba, bb) -> bool:
+    """Returns True if the bounding box of the first PDFMiner object is longer or equal to the second.
+
+    Parameters
+    ----------
+    ba : PDFMiner text object
+    bb : PDFMiner text object
+
+    Returns
+    -------
+    longer : bool
+        True if the bounding box of the first object is longer or equal
+
+    """
+    return (ba.x1 - ba.x0) >= (bb.x1 - bb.x0)
+
+
+def merge_close_lines(ar, line_tol=2):
     """Merges lines which are within a tolerance by calculating a
     moving mean, based on their x or y axis projections.
 
     Parameters
     ----------
     ar : list
-    line_close_tol : int, optional (default: 2)
+    line_tol : int, optional (default: 2)
 
     Returns
     -------
@@ -301,7 +477,7 @@ def merge_close_lines(ar, line_close_tol=2):
             ret.append(a)
         else:
             temp = ret[-1]
-            if np.isclose(temp, a, atol=line_close_tol):
+            if np.isclose(temp, a, atol=line_tol):
                 temp = (temp + a) / 2.0
                 ret[-1] = temp
             else:
@@ -309,7 +485,33 @@ def merge_close_lines(ar, line_close_tol=2):
     return ret
 
 
-def flag_font_size(textline, direction):
+def text_strip(text, strip=""):
+    """Strips any characters in `strip` that are present in `text`.
+    Parameters
+    ----------
+    text : str
+        Text to process and strip.
+    strip : str, optional (default: '')
+        Characters that should be stripped from `text`.
+    Returns
+    -------
+    stripped : str
+    """
+    if not strip:
+        return text
+
+    stripped = re.sub(
+        fr"[{''.join(map(re.escape, strip))}]", "", text, flags=re.UNICODE
+    )
+    return stripped
+
+
+# TODO: combine the following functions into a TextProcessor class which
+# applies corresponding transformations sequentially
+# (inspired from sklearn.pipeline.Pipeline)
+
+
+def flag_font_size(textline, direction, strip_text=""):
     """Flags super/subscripts in text by enclosing them with <s></s>.
     May give false positives.
 
@@ -319,16 +521,27 @@ def flag_font_size(textline, direction):
         List of PDFMiner LTChar objects.
     direction : string
         Direction of the PDFMiner LTTextLine object.
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
 
     Returns
     -------
     fstring : string
 
     """
-    if direction == 'horizontal':
+    if direction == "horizontal":
-        d = [(t.get_text(), np.round(t.height, decimals=6)) for t in textline if not isinstance(t, LTAnno)]
+        d = [
-    elif direction == 'vertical':
+            (t.get_text(), np.round(t.height, decimals=6))
-        d = [(t.get_text(), np.round(t.width, decimals=6)) for t in textline if not isinstance(t, LTAnno)]
+            for t in textline
+            if not isinstance(t, LTAnno)
+        ]
+    elif direction == "vertical":
+        d = [
+            (t.get_text(), np.round(t.width, decimals=6))
+            for t in textline
+            if not isinstance(t, LTAnno)
+        ]
     l = [np.round(size, decimals=6) for text, size in d]
     if len(set(l)) > 1:
         flist = []
@@ -336,21 +549,21 @@ def flag_font_size(textline, direction):
         for key, chars in groupby(d, itemgetter(1)):
             if key == min_size:
                 fchars = [t[0] for t in chars]
-                if ''.join(fchars).strip():
+                if "".join(fchars).strip():
-                    fchars.insert(0, '<s>')
+                    fchars.insert(0, "<s>")
-                    fchars.append('</s>')
+                    fchars.append("</s>")
-                    flist.append(''.join(fchars))
+                    flist.append("".join(fchars))
             else:
                 fchars = [t[0] for t in chars]
-                if ''.join(fchars).strip():
+                if "".join(fchars).strip():
-                    flist.append(''.join(fchars))
+                    flist.append("".join(fchars))
-        fstring = ''.join(flist).strip('\n')
+        fstring = "".join(flist)
     else:
-        fstring = ''.join([t.get_text() for t in textline]).strip('\n')
+        fstring = "".join([t.get_text() for t in textline])
-    return fstring
+    return text_strip(fstring, strip_text)
 
 
-def split_textline(table, textline, direction, flag_size=False):
+def split_textline(table, textline, direction, flag_size=False, strip_text=""):
     """Splits PDFMiner LTTextLine into substrings if it spans across
     multiple rows/columns.
 
@@ -365,6 +578,9 @@ def split_textline(table, textline, direction, flag_size=False):
         Whether or not to highlight a substring using <s></s>
         if its size is different from rest of the string. (Useful for
         super and subscripts.)
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
 
     Returns
     -------
@@ -377,38 +593,70 @@ def split_textline(table, textline, direction, flag_size=False):
     cut_text = []
     bbox = textline.bbox
     try:
-        if direction == 'horizontal' and not textline.is_empty():
+        if direction == "horizontal" and not textline.is_empty():
-            x_overlap = [i for i, x in enumerate(table.cols) if x[0] <= bbox[2] and bbox[0] <= x[1]]
+            x_overlap = [
-            r_idx = [j for j, r in enumerate(table.rows) if r[1] <= (bbox[1] + bbox[3]) / 2 <= r[0]]
+                i
+                for i, x in enumerate(table.cols)
+                if x[0] <= bbox[2] and bbox[0] <= x[1]
+            ]
+            r_idx = [
+                j
+                for j, r in enumerate(table.rows)
+                if r[1] <= (bbox[1] + bbox[3]) / 2 <= r[0]
+            ]
             r = r_idx[0]
-            x_cuts = [(c, table.cells[r][c].x2) for c in x_overlap if table.cells[r][c].right]
+            x_cuts = [
+                (c, table.cells[r][c].x2) for c in x_overlap if table.cells[r][c].right
+            ]
             if not x_cuts:
                 x_cuts = [(x_overlap[0], table.cells[r][-1].x2)]
             for obj in textline._objs:
                 row = table.rows[r]
                 for cut in x_cuts:
                     if isinstance(obj, LTChar):
-                        if (row[1] <= (obj.y0 + obj.y1) / 2 <= row[0] and
+                        if (
-                                (obj.x0 + obj.x1) / 2 <= cut[1]):
+                            row[1] <= (obj.y0 + obj.y1) / 2 <= row[0]
+                            and (obj.x0 + obj.x1) / 2 <= cut[1]
+                        ):
                             cut_text.append((r, cut[0], obj))
                             break
+                        else:
+                            # TODO: add test
+                            if cut == x_cuts[-1]:
+                                cut_text.append((r, cut[0] + 1, obj))
                     elif isinstance(obj, LTAnno):
                         cut_text.append((r, cut[0], obj))
-        elif direction == 'vertical' and not textline.is_empty():
+        elif direction == "vertical" and not textline.is_empty():
-            y_overlap = [j for j, y in enumerate(table.rows) if y[1] <= bbox[3] and bbox[1] <= y[0]]
+            y_overlap = [
-            c_idx = [i for i, c in enumerate(table.cols) if c[0] <= (bbox[0] + bbox[2]) / 2 <= c[1]]
+                j
+                for j, y in enumerate(table.rows)
+                if y[1] <= bbox[3] and bbox[1] <= y[0]
+            ]
+            c_idx = [
+                i
+                for i, c in enumerate(table.cols)
+                if c[0] <= (bbox[0] + bbox[2]) / 2 <= c[1]
+            ]
             c = c_idx[0]
-            y_cuts = [(r, table.cells[r][c].y1) for r in y_overlap if table.cells[r][c].bottom]
+            y_cuts = [
+                (r, table.cells[r][c].y1) for r in y_overlap if table.cells[r][c].bottom
+            ]
             if not y_cuts:
                 y_cuts = [(y_overlap[0], table.cells[-1][c].y1)]
             for obj in textline._objs:
                 col = table.cols[c]
                 for cut in y_cuts:
                     if isinstance(obj, LTChar):
-                        if (col[0] <= (obj.x0 + obj.x1) / 2 <= col[1] and
+                        if (
-                                (obj.y0 + obj.y1) / 2 >= cut[1]):
+                            col[0] <= (obj.x0 + obj.x1) / 2 <= col[1]
+                            and (obj.y0 + obj.y1) / 2 >= cut[1]
+                        ):
                             cut_text.append((cut[0], c, obj))
                             break
+                        else:
+                            # TODO: add test
+                            if cut == y_cuts[-1]:
+                                cut_text.append((cut[0] - 1, c, obj))
                     elif isinstance(obj, LTAnno):
                         cut_text.append((cut[0], c, obj))
     except IndexError:
@@ -416,14 +664,26 @@ def split_textline(table, textline, direction, flag_size=False):
     grouped_chars = []
     for key, chars in groupby(cut_text, itemgetter(0, 1)):
         if flag_size:
-            grouped_chars.append((key[0], key[1], flag_font_size([t[2] for t in chars], direction)))
+            grouped_chars.append(
+                (
+                    key[0],
+                    key[1],
+                    flag_font_size(
+                        [t[2] for t in chars], direction, strip_text=strip_text
+                    ),
+                )
+            )
         else:
             gchars = [t[2].get_text() for t in chars]
-            grouped_chars.append((key[0], key[1], ''.join(gchars).strip('\n')))
+            grouped_chars.append(
+                (key[0], key[1], text_strip("".join(gchars), strip_text))
+            )
     return grouped_chars
 
 
-def get_table_index(table, t, direction, split_text=False, flag_size=False):
+def get_table_index(
+    table, t, direction, split_text=False, flag_size=False, strip_text=""
+):
     """Gets indices of the table cell where given text object lies by
     comparing their y and x-coordinates.
 
@@ -441,6 +701,9 @@ def get_table_index(table, t, direction, split_text=False, flag_size=False):
         Whether or not to highlight a substring using <s></s>
         if its size is different from rest of the string. (Useful for
         super and subscripts)
+    strip_text : str, optional (default: '')
+        Characters that should be stripped from a string before
+        assigning it to a cell.
 
     Returns
     -------
@@ -459,8 +722,9 @@ def get_table_index(table, t, direction, split_text=False, flag_size=False):
     """
     r_idx, c_idx = [-1] * 2
     for r in range(len(table.rows)):
-        if ((t.y0 + t.y1) / 2.0 < table.rows[r][0] and
+        if (t.y0 + t.y1) / 2.0 < table.rows[r][0] and (t.y0 + t.y1) / 2.0 > table.rows[
-                (t.y0 + t.y1) / 2.0 > table.rows[r][1]):
+            r
+        ][1]:
             lt_col_overlap = []
             for c in table.cols:
                 if c[0] <= t.x1 and c[1] >= t.x0:
@@ -470,11 +734,12 @@ def get_table_index(table, t, direction, split_text=False, flag_size=False):
                 else:
                     lt_col_overlap.append(-1)
             if len(list(filter(lambda x: x != -1, lt_col_overlap))) == 0:
-                text = t.get_text().strip('\n')
+                text = t.get_text().strip("\n")
                 text_range = (t.x0, t.x1)
                 col_range = (table.cols[0][0], table.cols[-1][1])
-                warnings.warn("{} {} does not lie in column range {}".format(
+                warnings.warn(
-                    text, text_range, col_range))
+                    f"{text} {text_range} does not lie in column range {col_range}"
+                )
             r_idx = r
             c_idx = lt_col_overlap.index(max(lt_col_overlap))
             break
@@ -495,12 +760,26 @@ def get_table_index(table, t, direction, split_text=False, flag_size=False):
     error = ((X * (y0_offset + y1_offset)) + (Y * (x0_offset + x1_offset))) / charea
 
     if split_text:
-        return split_textline(table, t, direction, flag_size=flag_size), error
+        return (
+            split_textline(
+                table, t, direction, flag_size=flag_size, strip_text=strip_text
+            ),
+            error,
+        )
     else:
         if flag_size:
-            return [(r_idx, c_idx, flag_font_size(t._objs, direction))], error
+            return (
+                [
+                    (
+                        r_idx,
+                        c_idx,
+                        flag_font_size(t._objs, direction, strip_text=strip_text),
+                    )
+                ],
+                error,
+            )
         else:
-            return [(r_idx, c_idx, t.get_text().strip('\n'))], error
+            return [(r_idx, c_idx, text_strip(t.get_text(), strip_text))], error
 
 
 def compute_accuracy(error_weights):
@@ -551,25 +830,35 @@ def compute_whitespace(d):
     r_nempty_cells, c_nempty_cells = [], []
     for i in d:
         for j in i:
-            if j.strip() == '':
+            if j.strip() == "":
                 whitespace += 1
     whitespace = 100 * (whitespace / float(len(d) * len(d[0])))
     return whitespace
 
 
-def get_page_layout(filename, char_margin=1.0, line_margin=0.5, word_margin=0.1,
+def get_page_layout(
-               detect_vertical=True, all_texts=True):
+    filename,
+    line_overlap=0.5,
+    char_margin=1.0,
+    line_margin=0.5,
+    word_margin=0.1,
+    boxes_flow=0.5,
+    detect_vertical=True,
+    all_texts=True,
+):
     """Returns a PDFMiner LTPage object and page dimension of a single
-    page pdf. See https://euske.github.io/pdfminer/ to get definitions
+    page pdf. To get the definitions of kwargs, see
-    of kwargs.
+    https://pdfminersix.rtfd.io/en/latest/reference/composable.html.
 
     Parameters
     ----------
     filename : string
         Path to pdf file.
+    line_overlap : float
     char_margin : float
     line_margin : float
     word_margin : float
+    boxes_flow : float
     detect_vertical : bool
     all_texts : bool
 
@@ -581,16 +870,22 @@ def get_page_layout(filename, char_margin=1.0, line_margin=0.5, word_margin=0.1,
         Dimension of pdf page in the form (width, height).
 
     """
-    with open(filename, 'rb') as f:
+    with open(filename, "rb") as f:
         parser = PDFParser(f)
         document = PDFDocument(parser)
         if not document.is_extractable:
-            raise PDFTextExtractionNotAllowed
+            raise PDFTextExtractionNotAllowed(
-        laparams = LAParams(char_margin=char_margin,
+                f"Text extraction is not allowed: {filename}"
-                            line_margin=line_margin,
+            )
-                            word_margin=word_margin,
+        laparams = LAParams(
-                            detect_vertical=detect_vertical,
+            line_overlap=line_overlap,
-                            all_texts=all_texts)
+            char_margin=char_margin,
+            line_margin=line_margin,
+            word_margin=word_margin,
+            boxes_flow=boxes_flow,
+            detect_vertical=detect_vertical,
+            all_texts=all_texts,
+        )
         rsrcmgr = PDFResourceManager()
         device = PDFPageAggregator(rsrcmgr, laparams=laparams)
         interpreter = PDFPageInterpreter(rsrcmgr, device)
@@ -624,9 +919,11 @@ def get_text_objects(layout, ltype="char", t=None):
     """
     if ltype == "char":
         LTObject = LTChar
-    elif ltype == "lh":
+    elif ltype == "image":
+        LTObject = LTImage
+    elif ltype == "horizontal_text":
         LTObject = LTTextLineHorizontal
-    elif ltype == "lv":
+    elif ltype == "vertical_text":
         LTObject = LTTextLineVertical
     if t is None:
         t = []

docs/_static/csv/table_regions.csv

@@ -0,0 +1,4 @@
+"Età dell’Assicuratoall’epoca del decesso","Misura % dimaggiorazione"
+"18-75","1,00%"
+"76-80","0,50%"
+"81 in poi","0,10%"

docs/_templates/sidebarintro.html

@@ -4,13 +4,13 @@
   </a>
 </p>
 <p>
-<iframe src="https://ghbtns.com/github-btn.html?user=socialcopsdev&repo=camelot&type=watch&count=true&size=large"
+<iframe src="https://ghbtns.com/github-btn.html?user=camelot-dev&repo=camelot&type=watch&count=true&size=large"
   allowtransparency="true" frameborder="0" scrolling="0" width="200px" height="35px"></iframe>
 </p>
 
 <h3>Useful Links</h3>
 <ul>
-  <li><a href="https://github.com/socialcopsdev/camelot">Camelot @ GitHub</a></li>
+  <li><a href="https://github.com/camelot-dev/camelot">Camelot @ GitHub</a></li>
   <li><a href="https://pypi.org/project/camelot-py/">Camelot @ PyPI</a></li>
-  <li><a href="https://github.com/socialcopsdev/camelot/issues">Issue Tracker</a></li>
+  <li><a href="https://github.com/camelot-dev/camelot/issues">Issue Tracker</a></li>
 </ul>

docs/_templates/sidebarlogo.html

@@ -4,6 +4,6 @@
   </a>
 </p>
 <p>
-<iframe src="https://ghbtns.com/github-btn.html?user=socialcopsdev&repo=camelot&type=watch&count=true&size=large"
+<iframe src="https://ghbtns.com/github-btn.html?user=camelot-dev&repo=camelot&type=watch&count=true&size=large"
   allowtransparency="true" frameborder="0" scrolling="0" width="200px" height="35px"></iframe>
 </p>

docs/_themes/flask_theme_support.py

@@ -1,7 +1,19 @@
 # flasky pygments style based on tango style
 from pygments.style import Style
-from pygments.token import Keyword, Name, Comment, String, Error, \
+from pygments.token import (
-     Number, Operator, Generic, Whitespace, Punctuation, Other, Literal
+    Keyword,
+    Name,
+    Comment,
+    String,
+    Error,
+    Number,
+    Operator,
+    Generic,
+    Whitespace,
+    Punctuation,
+    Other,
+    Literal,
+)
 
 
 class FlaskyStyle(Style):
@@ -11,76 +23,67 @@ class FlaskyStyle(Style):
     styles = {
         # No corresponding class for the following:
         # Text:                    "", # class:  ''
-        Whitespace:                "underline #f8f8f8",       # class: 'w'
+        Whitespace: "underline #f8f8f8",  # class: 'w'
-        Error:                     "#a40000 border:#ef2929",  # class: 'err'
+        Error: "#a40000 border:#ef2929",  # class: 'err'
-        Other:                     "#000000",                 # class 'x'
+        Other: "#000000",  # class 'x'
-
+        Comment: "italic #8f5902",  # class: 'c'
-        Comment:                   "italic #8f5902",  # class: 'c'
+        Comment.Preproc: "noitalic",  # class: 'cp'
-        Comment.Preproc:           "noitalic",        # class: 'cp'
+        Keyword: "bold #004461",  # class: 'k'
-
+        Keyword.Constant: "bold #004461",  # class: 'kc'
-        Keyword:                   "bold #004461",    # class: 'k'
+        Keyword.Declaration: "bold #004461",  # class: 'kd'
-        Keyword.Constant:          "bold #004461",    # class: 'kc'
+        Keyword.Namespace: "bold #004461",  # class: 'kn'
-        Keyword.Declaration:       "bold #004461",    # class: 'kd'
+        Keyword.Pseudo: "bold #004461",  # class: 'kp'
-        Keyword.Namespace:         "bold #004461",    # class: 'kn'
+        Keyword.Reserved: "bold #004461",  # class: 'kr'
-        Keyword.Pseudo:            "bold #004461",    # class: 'kp'
+        Keyword.Type: "bold #004461",  # class: 'kt'
-        Keyword.Reserved:          "bold #004461",    # class: 'kr'
+        Operator: "#582800",  # class: 'o'
-        Keyword.Type:              "bold #004461",    # class: 'kt'
+        Operator.Word: "bold #004461",  # class: 'ow' - like keywords
-
+        Punctuation: "bold #000000",  # class: 'p'
-        Operator:                  "#582800",   # class: 'o'
-        Operator.Word:             "bold #004461",   # class: 'ow' - like keywords
-
-        Punctuation:               "bold #000000",   # class: 'p'
-
         # because special names such as Name.Class, Name.Function, etc.
         # are not recognized as such later in the parsing, we choose them
         # to look the same as ordinary variables.
-        Name:                      "#000000",         # class: 'n'
+        Name: "#000000",  # class: 'n'
-        Name.Attribute:            "#c4a000",         # class: 'na' - to be revised
+        Name.Attribute: "#c4a000",  # class: 'na' - to be revised
-        Name.Builtin:              "#004461",         # class: 'nb'
+        Name.Builtin: "#004461",  # class: 'nb'
-        Name.Builtin.Pseudo:       "#3465a4",         # class: 'bp'
+        Name.Builtin.Pseudo: "#3465a4",  # class: 'bp'
-        Name.Class:                "#000000",         # class: 'nc' - to be revised
+        Name.Class: "#000000",  # class: 'nc' - to be revised
-        Name.Constant:             "#000000",         # class: 'no' - to be revised
+        Name.Constant: "#000000",  # class: 'no' - to be revised
-        Name.Decorator:            "#888",            # class: 'nd' - to be revised
+        Name.Decorator: "#888",  # class: 'nd' - to be revised
-        Name.Entity:               "#ce5c00",         # class: 'ni'
+        Name.Entity: "#ce5c00",  # class: 'ni'
-        Name.Exception:            "bold #cc0000",    # class: 'ne'
+        Name.Exception: "bold #cc0000",  # class: 'ne'
-        Name.Function:             "#000000",         # class: 'nf'
+        Name.Function: "#000000",  # class: 'nf'
-        Name.Property:             "#000000",         # class: 'py'
+        Name.Property: "#000000",  # class: 'py'
-        Name.Label:                "#f57900",         # class: 'nl'
+        Name.Label: "#f57900",  # class: 'nl'
-        Name.Namespace:            "#000000",         # class: 'nn' - to be revised
+        Name.Namespace: "#000000",  # class: 'nn' - to be revised
-        Name.Other:                "#000000",         # class: 'nx'
+        Name.Other: "#000000",  # class: 'nx'
-        Name.Tag:                  "bold #004461",    # class: 'nt' - like a keyword
+        Name.Tag: "bold #004461",  # class: 'nt' - like a keyword
-        Name.Variable:             "#000000",         # class: 'nv' - to be revised
+        Name.Variable: "#000000",  # class: 'nv' - to be revised
-        Name.Variable.Class:       "#000000",         # class: 'vc' - to be revised
+        Name.Variable.Class: "#000000",  # class: 'vc' - to be revised
-        Name.Variable.Global:      "#000000",         # class: 'vg' - to be revised
+        Name.Variable.Global: "#000000",  # class: 'vg' - to be revised
-        Name.Variable.Instance:    "#000000",         # class: 'vi' - to be revised
+        Name.Variable.Instance: "#000000",  # class: 'vi' - to be revised
-
+        Number: "#990000",  # class: 'm'
-        Number:                    "#990000",         # class: 'm'
+        Literal: "#000000",  # class: 'l'
-
+        Literal.Date: "#000000",  # class: 'ld'
-        Literal:                   "#000000",         # class: 'l'
+        String: "#4e9a06",  # class: 's'
-        Literal.Date:              "#000000",         # class: 'ld'
+        String.Backtick: "#4e9a06",  # class: 'sb'
-
+        String.Char: "#4e9a06",  # class: 'sc'
-        String:                    "#4e9a06",         # class: 's'
+        String.Doc: "italic #8f5902",  # class: 'sd' - like a comment
-        String.Backtick:           "#4e9a06",         # class: 'sb'
+        String.Double: "#4e9a06",  # class: 's2'
-        String.Char:               "#4e9a06",         # class: 'sc'
+        String.Escape: "#4e9a06",  # class: 'se'
-        String.Doc:                "italic #8f5902",  # class: 'sd' - like a comment
+        String.Heredoc: "#4e9a06",  # class: 'sh'
-        String.Double:             "#4e9a06",         # class: 's2'
+        String.Interpol: "#4e9a06",  # class: 'si'
-        String.Escape:             "#4e9a06",         # class: 'se'
+        String.Other: "#4e9a06",  # class: 'sx'
-        String.Heredoc:            "#4e9a06",         # class: 'sh'
+        String.Regex: "#4e9a06",  # class: 'sr'
-        String.Interpol:           "#4e9a06",         # class: 'si'
+        String.Single: "#4e9a06",  # class: 's1'
-        String.Other:              "#4e9a06",         # class: 'sx'
+        String.Symbol: "#4e9a06",  # class: 'ss'
-        String.Regex:              "#4e9a06",         # class: 'sr'
+        Generic: "#000000",  # class: 'g'
-        String.Single:             "#4e9a06",         # class: 's1'
+        Generic.Deleted: "#a40000",  # class: 'gd'
-        String.Symbol:             "#4e9a06",         # class: 'ss'
+        Generic.Emph: "italic #000000",  # class: 'ge'
-
+        Generic.Error: "#ef2929",  # class: 'gr'
-        Generic:                   "#000000",         # class: 'g'
+        Generic.Heading: "bold #000080",  # class: 'gh'
-        Generic.Deleted:           "#a40000",         # class: 'gd'
+        Generic.Inserted: "#00A000",  # class: 'gi'
-        Generic.Emph:              "italic #000000",  # class: 'ge'
+        Generic.Output: "#888",  # class: 'go'
-        Generic.Error:             "#ef2929",         # class: 'gr'
+        Generic.Prompt: "#745334",  # class: 'gp'
-        Generic.Heading:           "bold #000080",    # class: 'gh'
+        Generic.Strong: "bold #000000",  # class: 'gs'
-        Generic.Inserted:          "#00A000",         # class: 'gi'
+        Generic.Subheading: "bold #800080",  # class: 'gu'
-        Generic.Output:            "#888",            # class: 'go'
+        Generic.Traceback: "bold #a40000",  # class: 'gt'
-        Generic.Prompt:            "#745334",         # class: 'gp'
-        Generic.Strong:            "bold #000000",    # class: 'gs'
-        Generic.Subheading:        "bold #800080",    # class: 'gu'
-        Generic.Traceback:         "bold #a40000",    # class: 'gt'
     }

docs/benchmark/lattice/agstat/agstat-data-pdfplumber.csv

@@ -0,0 +1,96 @@
+"0","1","2","3","4","5","6","7","8","9","10"
+"Sl.
+No.","District","n
+o
+i
+t
+a
+l3
+opu2-1hs)
+P1k
+d 20 la
+er n
+cto(I
+ef
+j
+o
+r
+P","%
+8
+8
+o )
+s
+ult t  tkh
+dna
+Aalen l
+v(I
+i
+u
+q
+E",")
+y
+n a
+umptiomentadult/donnes)
+nsres/h t
+ouimk
+Cqga
+al re00n L
+ot  4(I
+T @
+(","menteds, age)nes)
+uireg sewastton
+qn h 
+Reudis &ak
+al cld L
+tnen
+To(Ife(I","","","","",""
+"","","","","","","f
+i
+r
+a
+h
+K","i
+b
+a
+R","l
+a
+t
+o
+T","e
+c
+i
+R","y
+d
+d
+a
+P"
+"1","Balasore","23.65","20.81","3.04","3.47","2.78","0.86","3.64","0.17","0.25"
+"2","Bhadrak","15.34","13.50","1.97","2.25","3.50","0.05","3.55","1.30","1.94"
+"3","Balangir","17.01","14.97","2.19","2.50","6.23","0.10","6.33","3.83","5.72"
+"4","Subarnapur","6.70","5.90","0.86","0.98","4.48","1.13","5.61","4.63","6.91"
+"5","Cuttack","26.63","23.43","3.42","3.91","3.75","0.06","3.81","-0.10","-0.15"
+"6","Jagatsingpur","11.49","10.11","1.48","1.69","2.10","0.02","2.12","0.43","0.64"
+"7","Jajpur","18.59","16.36","2.39","2.73","2.13","0.04","2.17","-0.56","-0.84"
+"8","Kendrapara","14.62","12.87","1.88","2.15","2.60","0.07","2.67","0.52","0.78"
+"9","Dhenkanal","12.13","10.67","1.56","1.78","2.26","0.02","2.28","0.50","0.75"
+"10","Angul","12.93","11.38","1.66","1.90","1.73","0.02","1.75","-0.15","-0.22"
+"11","Ganjam","35.77","31.48","4.60","5.26","4.57","0.00","4.57","-0.69","-1.03"
+"12","Gajapati","5.85","5.15","0.75","0.86","0.68","0.01","0.69","-0.17","-0.25"
+"13","Kalahandi","16.12","14.19","2.07","2.37","5.42","1.13","6.55","4.18","6.24"
+"14","Nuapada","6.18","5.44","0.79","0.90","1.98","0.08","2.06","1.16","1.73"
+"15","Keonjhar","18.42","16.21","2.37","2.71","2.76","0.08","2.84","0.13","0.19"
+"16","Koraput","14.09","12.40","1.81","2.07","2.08","0.34","2.42","0.35","0.52"
+"17","Malkangiri","6.31","5.55","0.81","0.93","1.78","0.04","1.82","0.89","1.33"
+"18","Nabarangpur","12.50","11.00","1.61","1.84","3.26","0.02","3.28","1.44","2.15"
+"19","Rayagada","9.83","8.65","1.26","1.44","1.15","0.03","1.18","-0.26","-0.39"
+"20","Mayurbhanj","25.61","22.54","3.29","3.76","4.90","0.06","4.96","1.20","1.79"
+"21","Kandhamal","7.45","6.56","0.96","1.10","0.70","0.01","0.71","-0.39","-0.58"
+"22","Boudh","4.51","3.97","0.58","0.66","1.73","0.03","1.76","1.10","1.64"
+"23","Puri","17.29","15.22","2.22","2.54","2.45","0.99","3.44","0.90","1.34"
+"24","Khordha","23.08","20.31","2.97","3.39","2.02","0.03","2.05","-1.34","-2.00"
+"25","Nayagarh","9.78","8.61","1.26","1.44","2.10","0.00","2.10","0.66","0.99"
+"26","Sambalpur","10.62","9.35","1.37","1.57","3.45","0.71","4.16","2.59","3.87"
+"27","Bargarh","15.00","13.20","1.93","2.21","6.87","2.65","9.52","7.31","10.91"
+"28","Deogarh","3.18","2.80","0.41","0.47","1.12","0.07","1.19","0.72","1.07"
+"29","Jharsuguda","5.91","5.20","0.76","0.87","0.99","0.01","1.00","0.13","0.19"
+"30","","","18.66","2.72","3.11","4.72","0.02","4.74","1.63","2.43"

docs/benchmark/lattice/column_span_1/column_span_1-data-pdfplumber.csv

@@ -0,0 +1,56 @@
+"0","1","2","3","4","5","6","7"
+"Rate of Accidental Deaths & Suicides and Population Growth During 1967 to 2013","","","","","","",""
+"Sl. 
+No.","Year","Population 
+(in Lakh)","Accidental Deaths","","Suicides","","Percentage 
+Population 
+growth"
+"","","","Incidence","Rate","Incidence","Rate",""
+"(1)","(2)","(3)","(4)","(5)","(6)","(7)","(8)"
+"1.","1967","4999","126762","25.4","38829","7.8","2.2"
+"2.","1968","5111","126232","24.7","40688","8.0","2.2"
+"3.","1969","5225","130755","25.0","43633","8.4","2.2"
+"4.","1970","5343","139752","26.2","48428","9.1","2.3"
+"5.","1971","5512","105601","19.2","43675","7.9","3.2"
+"6.","1972","5635","106184","18.8","43601","7.7","2.2"
+"7.","1973","5759","130654","22.7","40807","7.1","2.2"
+"8.","1974","5883","110624","18.8","46008","7.8","2.2"
+"9.","1975","6008","113016","18.8","42890","7.1","2.1"
+"10.","1976","6136","111611","18.2","41415","6.7","2.1"
+"11.","1977","6258","117338","18.8","39718","6.3","2.0"
+"12.","1978","6384","118594","18.6","40207","6.3","2.0"
+"13.","1979","6510","108987","16.7","38217","5.9","2.0"
+"14.","1980","6636","116912","17.6","41663","6.3","1.9"
+"15.","1981","6840","122221","17.9","40245","5.9","3.1"
+"16.","1982","7052","125993","17.9","44732","6.3","3.1"
+"17.","1983","7204","128576","17.8","46579","6.5","2.2"
+"18.","1984","7356","134628","18.3","50571","6.9","2.1"
+"19.","1985","7509","139657","18.6","52811","7.0","2.1"
+"20.","1986","7661","147023","19.2","54357","7.1","2.0"
+"21.","1987","7814","152314","19.5","58568","7.5","2.0"
+"22.","1988","7966","163522","20.5","64270","8.1","1.9"
+"23.","1989","8118","169066","20.8","68744","8.5","1.9"
+"24.","1990","8270","174401","21.1","73911","8.9","1.9"
+"25.","1991","8496","188003","22.1","78450","9.2","2.7"
+"26.","1992","8677","194910","22.5","80149","9.2","2.1"
+"27.","1993","8838","192357","21.8","84244","9.5","1.9"
+"28.","1994","8997","190435","21.2","89195","9.9","1.8"
+"29.","1995","9160","222487","24.3","89178","9.7","1.8"
+"30.","1996","9319","220094","23.6","88241","9.5","1.7"
+"31.","1997","9552","233903","24.5","95829","10.0","2.5"
+"32.","1998","9709","258409","26.6","104713","10.8","1.6"
+"33.","1999","9866","271918","27.6","110587","11.2","1.6"
+"34.","2000","10021","255883","25.5","108593","10.8","1.6"
+"35.","2001","10270","271019","26.4","108506","10.6","2.5"
+"36.","2002","10506","260122","24.8","110417","10.5","2.3"
+"37.","2003","10682","259625","24.3","110851","10.4","1.7"
+"38.","2004","10856","277263","25.5","113697","10.5","1.6"
+"39.","2005","11028","294175","26.7","113914","10.3","1.6"
+"40.","2006","11198","314704","28.1","118112","10.5","1.5"
+"41.","2007","11366","340794","30.0","122637","10.8","1.5"
+"42.","2008","11531","342309","29.7","125017","10.8","1.4"
+"43.","2009","11694","357021","30.5","127151","10.9","1.4"
+"44.","2010","11858","384649","32.4","134599","11.4","1.4"
+"45.","2011","12102","390884","32.3","135585","11.2","2.1"
+"46.","2012","12134","394982","32.6","135445","11.2","1.0"
+"47.","2013","12288","400517","32.6","134799","11.0","1.0"

docs/benchmark/lattice/rotated/rotated-data-pdfplumber.csv

@@ -0,0 +1,18 @@
+"0","1","2"
+"","e
+bl
+a
+ail
+v
+a
+ 
+t
+o
+n
+ 
+a
+t
+a
+D
+ 
+*",""

docs/benchmark/lattice/twotables_1/twotables_1-data-pdfplumber.csv

@@ -0,0 +1,3 @@
+"0"
+"Sl."
+"No."

docs/benchmark/lattice/twotables_2/twotables_2-data-pdfplumber.csv

@@ -0,0 +1,3 @@
+"0"
+"Table 6 : DISTRIBUTION (%) OF HOUSEHOLDS BY LITERACY STATUS OF"
+"MALE HEAD OF THE HOUSEHOLD"

docs/benchmark/stream/12s0324/12s0324-data-camelot-page-1-table-1.csv

@@ -1,3 +1,7 @@
+"[In thousands (11,062.6 represents 11,062,600) For year ending December 31. Based on Uniform Crime Reporting (UCR)","","","","","","","","",""
+"Program. Represents arrests reported (not charged) by 12,910 agencies with a total population of 247,526,916 as estimated","","","","","","","","",""
+"by the FBI. Some persons may be arrested more than once during a year, therefore, the data in this table, in some cases,","","","","","","","","",""
+"could represent multiple arrests of the same person. See text, this section and source]","","","","","","","","",""
 "","","Total","","","Male","","","Female",""
 "Offense charged","","Under 18","18 years","","Under 18","18 years","","Under 18","18 years"
 "","Total","years","and over","Total","years","and over","Total","years","and over"
@@ -36,3 +40,4 @@
 "Curfew and loitering law violations  ..","91.0","91.0","(X)","63.1","63.1","(X)","28.0","28.0","(X)"
 "Runaways  . . . . . . . .. .. .. .. .. ....","75.8","75.8","(X)","34.0","34.0","(X)","41.8","41.8","(X)"
 "","– Represents zero. X Not applicable. 1 Buying, receiving, possessing stolen property. 2 Except forcible rape and prostitution.","","","","","","","",""
+"","Source: U.S. Department of Justice, Federal Bureau of Investigation, Uniform Crime Reports, Arrests Master Files.","","","","","","","",""

docs/benchmark/stream/12s0324/12s0324-data-camelot-page-1-table-2.csv

@@ -1,3 +1,7 @@
+"","Source: U.S. Department of Justice, Federal Bureau of Investigation, Uniform Crime Reports, Arrests Master Files.","","","",""
+"Table 325. Arrests by Race: 2009","","","","",""
+"[Based on Uniform Crime Reporting (UCR) Program. Represents arrests reported (not charged) by 12,371 agencies","","","","",""
+"with a total population of 239,839,971 as estimated by the FBI. See headnote, Table 324]","","","","",""
 "","","","","American",""
 "Offense charged","","","","Indian/Alaskan","Asian Pacific"
 "","Total","White","Black","Native","Islander"
@@ -34,3 +38,4 @@
 "Curfew and loitering law violations  . .. ... .. ....","89,578","54,439","33,207","872","1,060"
 "Runaways  . . . . . . . .. .. .. .. .. .. .... .. ..... .","73,616","48,343","19,670","1,653","3,950"
 "1 Except forcible rape and prostitution.","","","","",""
+"","Source: U.S. Department of Justice, Federal Bureau of Investigation, “Crime in the United States, Arrests,” September 2010,","","","",""

docs/benchmark/stream/birdisland/birdisland-data-camelot-page-1-table-1.csv

@@ -1,35 +1,43 @@
-"","","","","","SCN","Seed","Yield","Moisture","Lodgingg","g","Stand","","Gross"
+"","2012 BETTER VARIETIES Harvest Report for Minnesota Central  [ MNCE ]2012 BETTER VARIETIES Harvest Report for Minnesota Central  [ MNCE ]","","","","","","","","","","","","ALL SEASON TESTALL SEASON TEST",""
-"Company/Brandpy","","Product/Brand†","Technol.†","Mat.","Resist.","Trmt.†","Bu/A","%","%","","(x 1000)(",")","Income"
+"","Doug Toreen, Renville County, MN 55310          [ BIRD ISLAND ]Doug Toreen, Renville County, MN 55310","","","","","[ BIRD ISLAND ]","","","","","","","1.3 - 2.0 MAT. GROUP1.3 - 2.0 MAT. GROUP",""
-"KrugerKruger","","K2-1901K2 1901","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","56.456.4","7.67.6","00","","126.3126.3","","$846$846"
+"PREVPREV. CROP/HERB:","CROP/HERB","C/ S","Corn / Surpass, RoundupR","d","","","","","","","","","","S2MNCE01S2MNCE01"
-"StineStine","","19RA02 §19RA02 §","RR2YRR2Y","1 91.9","RR","CMBCMB","55.355.3","7 67.6","00","","120 0120.0","","$830$830"
+"SOIL DESCRIPTION:","","C","Canisteo clay loam, mod. well drained, non-irrigated","","","","","","","","","","",""
-"WensmanWensman","","W 3190NR2W 3190NR2","RR2YRR2Y","1 91.9","RR","AcAc","54 554.5","7 67.6","00","","119 5119.5","","$818$818"
+"SOIL CONDITIONS:","","","High P, high K, 6.7 pH, 3.9% OM, Low SCN","","","","","","","","","","","30"" ROW SPACING"
-"H ftHefty","","H17Y12H17Y12","RR2YRR2Y","1 71.7","MRMR","II","53 753.7","7 77.7","00","","124 4124.4","","$806$806"
+"TILLAGE/CULTIVATION:TILLAGE/CULTIVATION:","","","conventional w/ fall tillconventional w/ fall till","","","","","","","","","","",""
-"Dyna-Gro","","S15RY53","RR2Y","1.5","R","Ac","53.6","7.7","0","","126.8","","$804"
+"PEST MANAGEMENT:PEST MANAGEMENT:","","Roundup twiceRoundup twice","","","","","","","","","","","",""
-"LG SeedsLG Seeds","","C2050R2C2050R2","RR2YRR2Y","2.12.1","RR","AcAc","53.653.6","7.77.7","00","","123.9123.9","","$804$804"
+"SEEDED - RATE:","","May 15M15","140,000 /A140 000 /A","","","","","","","","TOP 30 foTOP 30 for YIELD of 63 TESTED","","YIELD of 63 TESTED",""
-"Titan ProTitan Pro","","19M4219M42","RR2YRR2Y","1.91.9","RR","CMBCMB","53.653.6","7.77.7","00","","121.0121.0","","$804$804"
+"HARVESTEDHARVESTED - STAND:","STAND","O t 3Oct 3","122 921 /A122,921 /A","","","","","","","","","AVERAGE of (3) REPLICATIONSAVERAGE of (3) REPLICATIONS","",""
-"StineStine","","19RA02 (2) §19RA02 (2) §","RR2YRR2Y","1 91.9","RR","CMBCMB","53 453.4","7 77.7","00","","123 9123.9","","$801$801"
+"","","","","","","SCN","Seed","Yield","Moisture","Lodgingg","g","Stand","","Gross"
-"AsgrowAsgrow","","AG1832 §AG1832 §","RR2YRR2Y","1 81.8","MRMR","Ac PVAc,PV","52 952.9","7 77.7","00","","122 0122.0","","$794$794"
+"","Company/Brandpy","Product/Brand†","","Technol.†","Mat.","Resist.","Trmt.†","Bu/A","%","%","","(x 1000)(",")","Income"
-"Prairie Brandiid","","PB-1566R2662","RR2Y2","1.5","R","CMB","52.8","7.7","0","","122.9","","$792$"
+"","KrugerKruger","K2-1901K2 1901","","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","56.456.4","7.67.6","00","","126.3126.3","","$846$846"
-"Channel","","1901R2","RR2Y","1.9","R","Ac,PV,","52.8","7.6","0","","123.4","","$791$"
+"","StineStine","19RA02 §19RA02 §","","RR2YRR2Y","1 91.9","RR","CMBCMB","55.355.3","7 67.6","00","","120 0120.0","","$830$830"
-"Titan ProTitan Pro","","20M120M1","RR2YRR2Y","2.02.0","RR","AmAm","52.552.5","7.57.5","00","","124.4124.4","","$788$788"
+"","WensmanWensman","W 3190NR2W 3190NR2","","RR2YRR2Y","1 91.9","RR","AcAc","54 554.5","7 67.6","00","","119 5119.5","","$818$818"
-"KrugerKruger","","K2-2002K2-2002","RR2YRR2Y","2 02.0","RR","Ac PVAc,PV","52 452.4","7 97.9","00","","125 4125.4","","$786$786"
+"","H ftHefty","H17Y12H17Y12","","RR2YRR2Y","1 71.7","MRMR","II","53 753.7","7 77.7","00","","124 4124.4","","$806$806"
-"ChannelChannel","","1700R21700R2","RR2YRR2Y","1 71.7","RR","Ac PVAc,PV","52 352.3","7 97.9","00","","123 9123.9","","$784$784"
+"","Dyna-Gro","S15RY53","","RR2Y","1.5","R","Ac","53.6","7.7","0","","126.8","","$804"
-"H ftHefty","","H16Y11H16Y11","RR2YRR2Y","1 61.6","MRMR","II","51 451.4","7 67.6","00","","123 9123.9","","$771$771"
+"","LG SeedsLG Seeds","C2050R2C2050R2","","RR2YRR2Y","2.12.1","RR","AcAc","53.653.6","7.77.7","00","","123.9123.9","","$804$804"
-"Anderson","","162R2Y","RR2Y","1.6","R","None","51.3","7.5","0","","119.5","","$770"
+"","Titan ProTitan Pro","19M4219M42","","RR2YRR2Y","1.91.9","RR","CMBCMB","53.653.6","7.77.7","00","","121.0121.0","","$804$804"
-"Titan ProTitan Pro","","15M2215M22","RR2YRR2Y","1.51.5","RR","CMBCMB","51.351.3","7.87.8","00","","125.4125.4","","$769$769"
+"","StineStine","19RA02 (2) §19RA02 (2) §","","RR2YRR2Y","1 91.9","RR","CMBCMB","53 453.4","7 77.7","00","","123 9123.9","","$801$801"
-"DairylandDairyland","","DSR-1710R2YDSR-1710R2Y","RR2YRR2Y","1 71.7","RR","CMBCMB","51 351.3","7 77.7","00","","122 0122.0","","$769$769"
+"","AsgrowAsgrow","AG1832 §AG1832 §","","RR2YRR2Y","1 81.8","MRMR","Ac PVAc,PV","52 952.9","7 77.7","00","","122 0122.0","","$794$794"
-"HeftyHefty","","H20R3H20R3","RR2YRR2Y","2 02.0","MRMR","II","50 550.5","8 28.2","00","","121 0121.0","","$757$757"
+"","Prairie Brandiid","PB-1566R2662","","RR2Y2","1.5","R","CMB","52.8","7.7","0","","122.9","","$792$"
-"PPrairie BrandiiBd","","PB 1743R2PB-1743R2","RR2YRR2Y","1 71.7","RR","CMBCMB","50 250.2","7 77.7","00","","125 8125.8","","$752$752"
+"","Channel","1901R2","","RR2Y","1.9","R","Ac,PV,","52.8","7.6","0","","123.4","","$791$"
-"Gold Country","","1741","RR2Y","1.7","R","Ac","50.1","7.8","0","","123.9","","$751"
+"","Titan ProTitan Pro","20M120M1","","RR2YRR2Y","2.02.0","RR","AmAm","52.552.5","7.57.5","00","","124.4124.4","","$788$788"
-"Trelaye ay","","20RR4303","RR2Y","2.00","R","Ac,Exc,","49.99 9","7.66","00","","127.88","","$749$9"
+"","KrugerKruger","K2-2002K2-2002","","RR2YRR2Y","2 02.0","RR","Ac PVAc,PV","52 452.4","7 97.9","00","","125 4125.4","","$786$786"
-"HeftyHefty","","H14R3H14R3","RR2YRR2Y","1.41.4","MRMR","II","49.749.7","7.77.7","00","","122.9122.9","","$746$746"
+"","ChannelChannel","1700R21700R2","","RR2YRR2Y","1 71.7","RR","Ac PVAc,PV","52 352.3","7 97.9","00","","123 9123.9","","$784$784"
-"Prairie BrandPrairie Brand","","PB-2099NRR2PB-2099NRR2","RR2YRR2Y","2 02.0","RR","CMBCMB","49 649.6","7 87.8","00","","126 3126.3","","$743$743"
+"","H ftHefty","H16Y11H16Y11","","RR2YRR2Y","1 61.6","MRMR","II","51 451.4","7 67.6","00","","123 9123.9","","$771$771"
-"WensmanWensman","","W 3174NR2W 3174NR2","RR2YRR2Y","1 71.7","RR","AcAc","49 349.3","7 67.6","00","","122 5122.5","","$740$740"
+"","Anderson","162R2Y","","RR2Y","1.6","R","None","51.3","7.5","0","","119.5","","$770"
-"KKruger","","K2 1602K2-1602","RR2YRR2Y","1 61.6","R","Ac,PV","48.78","7.66","00","","125.412","","$731$31"
+"","Titan ProTitan Pro","15M2215M22","","RR2YRR2Y","1.51.5","RR","CMBCMB","51.351.3","7.87.8","00","","125.4125.4","","$769$769"
-"NK Brand","","S18-C2 §§","RR2Y","1.8","R","CMB","48.7","7.7","0","","126.8","","$731$"
+"","DairylandDairyland","DSR-1710R2YDSR-1710R2Y","","RR2YRR2Y","1 71.7","RR","CMBCMB","51 351.3","7 77.7","00","","122 0122.0","","$769$769"
-"KrugerKruger","","K2-1902K2 1902","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","48.748.7","7.57.5","00","","124.4124.4","","$730$730"
+"","HeftyHefty","H20R3H20R3","","RR2YRR2Y","2 02.0","MRMR","II","50 550.5","8 28.2","00","","121 0121.0","","$757$757"
-"Prairie BrandPrairie Brand","","PB-1823R2PB-1823R2","RR2YRR2Y","1 81.8","RR","NoneNone","48 548.5","7 67.6","00","","121 0121.0","","$727$727"
+"","PPrairie BrandiiBd","PB 1743R2PB-1743R2","","RR2YRR2Y","1 71.7","RR","CMBCMB","50 250.2","7 77.7","00","","125 8125.8","","$752$752"
-"Gold CountryGold Country","","15411541","RR2YRR2Y","1 51.5","RR","AcAc","48 448.4","7 67.6","00","","110 4110.4","","$726$726"
+"","Gold Country","1741","","RR2Y","1.7","R","Ac","50.1","7.8","0","","123.9","","$751"
-"","","","","","","Test Average  =","47 647.6","7 77.7","00","","122 9122.9","","$713$713"
+"","Trelaye ay","20RR4303","","RR2Y","2.00","R","Ac,Exc,","49.99 9","7.66","00","","127.88","","$749$9"
-"","","","","","","LSD (0.10)  =","5.7","0.3","ns","","37.8","","566.4"
+"","HeftyHefty","H14R3H14R3","","RR2YRR2Y","1.41.4","MRMR","II","49.749.7","7.77.7","00","","122.9122.9","","$746$746"
-"","F.I.R.S.T. Managerg","","","","","C.V.  =","8.8","2.9","","","56.4","","846.2"
+"","Prairie BrandPrairie Brand","PB-2099NRR2PB-2099NRR2","","RR2YRR2Y","2 02.0","RR","CMBCMB","49 649.6","7 87.8","00","","126 3126.3","","$743$743"
+"","WensmanWensman","W 3174NR2W 3174NR2","","RR2YRR2Y","1 71.7","RR","AcAc","49 349.3","7 67.6","00","","122 5122.5","","$740$740"
+"","KKruger","K2 1602K2-1602","","RR2YRR2Y","1 61.6","R","Ac,PV","48.78","7.66","00","","125.412","","$731$31"
+"","NK Brand","S18-C2 §§","","RR2Y","1.8","R","CMB","48.7","7.7","0","","126.8","","$731$"
+"","KrugerKruger","K2-1902K2 1902","","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","48.748.7","7.57.5","00","","124.4124.4","","$730$730"
+"","Prairie BrandPrairie Brand","PB-1823R2PB-1823R2","","RR2YRR2Y","1 81.8","RR","NoneNone","48 548.5","7 67.6","00","","121 0121.0","","$727$727"
+"","Gold CountryGold Country","15411541","","RR2YRR2Y","1 51.5","RR","AcAc","48 448.4","7 67.6","00","","110 4110.4","","$726$726"
+"","","","","","","","Test Average  =","47 647.6","7 77.7","00","","122 9122.9","","$713$713"
+"","","","","","","","LSD (0.10)  =","5.7","0.3","ns","","37.8","","566.4"

docs/benchmark/stream/birdisland/birdisland-data-camelot-page-1-table-2.csv

@@ -0,0 +1,39 @@
+"TILLAGE/CULTIVATION:TILLAGE/CULTIVATION:","","conventional w/ fall tillconventional w/ fall till","","","","","","","","","","",""
+"PEST MANAGEMENT:PEST MANAGEMENT:","","Roundup twiceRoundup twice","","","","","","","","","","",""
+"SEEDED - RATE:","","May 15M15","140,000 /A140 000 /A","","","","","","","TOP 30 foTOP 30 for YIELD of 63 TESTED","","YIELD of 63 TESTED",""
+"HARVESTEDHARVESTED - STAND:STAND","","O t 3Oct 3","122 921 /A122,921 /A","","","","","","","","AVERAGE of (3) REPLICATIONSAVERAGE of (3) REPLICATIONS","",""
+"","","","","","SCN","Seed","Yield","Moisture","Lodgingg","g","Stand","","Gross"
+"Company/Brandpy","","Product/Brand†","Technol.†","Mat.","Resist.","Trmt.†","Bu/A","%","%","","(x 1000)(",")","Income"
+"KrugerKruger","","K2-1901K2 1901","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","56.456.4","7.67.6","00","","126.3126.3","","$846$846"
+"StineStine","","19RA02 §19RA02 §","RR2YRR2Y","1 91.9","RR","CMBCMB","55.355.3","7 67.6","00","","120 0120.0","","$830$830"
+"WensmanWensman","","W 3190NR2W 3190NR2","RR2YRR2Y","1 91.9","RR","AcAc","54 554.5","7 67.6","00","","119 5119.5","","$818$818"
+"H ftHefty","","H17Y12H17Y12","RR2YRR2Y","1 71.7","MRMR","II","53 753.7","7 77.7","00","","124 4124.4","","$806$806"
+"Dyna-Gro","","S15RY53","RR2Y","1.5","R","Ac","53.6","7.7","0","","126.8","","$804"
+"LG SeedsLG Seeds","","C2050R2C2050R2","RR2YRR2Y","2.12.1","RR","AcAc","53.653.6","7.77.7","00","","123.9123.9","","$804$804"
+"Titan ProTitan Pro","","19M4219M42","RR2YRR2Y","1.91.9","RR","CMBCMB","53.653.6","7.77.7","00","","121.0121.0","","$804$804"
+"StineStine","","19RA02 (2) §19RA02 (2) §","RR2YRR2Y","1 91.9","RR","CMBCMB","53 453.4","7 77.7","00","","123 9123.9","","$801$801"
+"AsgrowAsgrow","","AG1832 §AG1832 §","RR2YRR2Y","1 81.8","MRMR","Ac PVAc,PV","52 952.9","7 77.7","00","","122 0122.0","","$794$794"
+"Prairie Brandiid","","PB-1566R2662","RR2Y2","1.5","R","CMB","52.8","7.7","0","","122.9","","$792$"
+"Channel","","1901R2","RR2Y","1.9","R","Ac,PV,","52.8","7.6","0","","123.4","","$791$"
+"Titan ProTitan Pro","","20M120M1","RR2YRR2Y","2.02.0","RR","AmAm","52.552.5","7.57.5","00","","124.4124.4","","$788$788"
+"KrugerKruger","","K2-2002K2-2002","RR2YRR2Y","2 02.0","RR","Ac PVAc,PV","52 452.4","7 97.9","00","","125 4125.4","","$786$786"
+"ChannelChannel","","1700R21700R2","RR2YRR2Y","1 71.7","RR","Ac PVAc,PV","52 352.3","7 97.9","00","","123 9123.9","","$784$784"
+"H ftHefty","","H16Y11H16Y11","RR2YRR2Y","1 61.6","MRMR","II","51 451.4","7 67.6","00","","123 9123.9","","$771$771"
+"Anderson","","162R2Y","RR2Y","1.6","R","None","51.3","7.5","0","","119.5","","$770"
+"Titan ProTitan Pro","","15M2215M22","RR2YRR2Y","1.51.5","RR","CMBCMB","51.351.3","7.87.8","00","","125.4125.4","","$769$769"
+"DairylandDairyland","","DSR-1710R2YDSR-1710R2Y","RR2YRR2Y","1 71.7","RR","CMBCMB","51 351.3","7 77.7","00","","122 0122.0","","$769$769"
+"HeftyHefty","","H20R3H20R3","RR2YRR2Y","2 02.0","MRMR","II","50 550.5","8 28.2","00","","121 0121.0","","$757$757"
+"PPrairie BrandiiBd","","PB 1743R2PB-1743R2","RR2YRR2Y","1 71.7","RR","CMBCMB","50 250.2","7 77.7","00","","125 8125.8","","$752$752"
+"Gold Country","","1741","RR2Y","1.7","R","Ac","50.1","7.8","0","","123.9","","$751"
+"Trelaye ay","","20RR4303","RR2Y","2.00","R","Ac,Exc,","49.99 9","7.66","00","","127.88","","$749$9"
+"HeftyHefty","","H14R3H14R3","RR2YRR2Y","1.41.4","MRMR","II","49.749.7","7.77.7","00","","122.9122.9","","$746$746"
+"Prairie BrandPrairie Brand","","PB-2099NRR2PB-2099NRR2","RR2YRR2Y","2 02.0","RR","CMBCMB","49 649.6","7 87.8","00","","126 3126.3","","$743$743"
+"WensmanWensman","","W 3174NR2W 3174NR2","RR2YRR2Y","1 71.7","RR","AcAc","49 349.3","7 67.6","00","","122 5122.5","","$740$740"
+"KKruger","","K2 1602K2-1602","RR2YRR2Y","1 61.6","R","Ac,PV","48.78","7.66","00","","125.412","","$731$31"
+"NK Brand","","S18-C2 §§","RR2Y","1.8","R","CMB","48.7","7.7","0","","126.8","","$731$"
+"KrugerKruger","","K2-1902K2 1902","RR2YRR2Y","1.91.9","RR","Ac,PVAc,PV","48.748.7","7.57.5","00","","124.4124.4","","$730$730"
+"Prairie BrandPrairie Brand","","PB-1823R2PB-1823R2","RR2YRR2Y","1 81.8","RR","NoneNone","48 548.5","7 67.6","00","","121 0121.0","","$727$727"
+"Gold CountryGold Country","","15411541","RR2YRR2Y","1 51.5","RR","AcAc","48 448.4","7 67.6","00","","110 4110.4","","$726$726"
+"","","","","","","Test Average  =","47 647.6","7 77.7","00","","122 9122.9","","$713$713"
+"","","","","","","LSD (0.10)  =","5.7","0.3","ns","","37.8","","566.4"
+"","F.I.R.S.T. Managerg","","","","","C.V.  =","8.8","2.9","","","56.4","","846.2"

docs/benchmark/stream/district_health/district_health-data-pdfplumber.csv

@@ -0,0 +1,66 @@
+"0","1","2","3","4"
+"","DLHS-4 (2012-13)","","DLHS-3 (2007-08)",""
+"Indicators","TOTAL","RURAL","TOTAL","RURAL"
+"Child feeding practices (based on last-born child in the reference period) (%)","","","",""
+"Children age 0-5 months exclusively breastfed9 ..........................................................................  76.9  80.0 
+Children age 6-9 months receiving solid/semi-solid food and breast milk ....................................  78.6  75.0 
+Children age 12-23 months receiving breast feeding along with complementary feeding ...........  31.8  24.2   
+Children age 6-35 months exclusively breastfed for at least 6 months ........................................  4.7  3.4 
+Children under 3 years breastfed within one hour of birth ............................................................  42.9  46.5","","","NA","NA"
+"","","","85.9","89.3"
+"","","","NA","NA"
+"","","","30.0","27.7"
+"","","","50.6","52.9"
+"Birth Weight (%) (age below 36 months)","","","",""
+"Percentage of Children weighed at birth ......................................................................................  38.8  41.0  NA  NA 
+Percentage of Children with low birth weight (out of those who weighted) ( below 2.5 kg) .........  12.8  14.6  NA  NA","","","",""
+"Awareness about Diarrhoea (%)","","","",""
+"Women know about  what to do when a child gets  diarrhoea .....................................................  96.3  96.2","","","94.4","94.2"
+"Awareness about ARI (%)","","","",""
+"Women aware about danger signs of ARI10 .................................................................................  55.9  59.7","","","32.8","34.7"
+"Treatment of childhood diseases (based on last two surviving children born during the","","","",""
+"","","","",""
+"reference period) (%)","","","",""
+"","","","",""
+"Prevalence of diarrhoea in last 2 weeks for under 5 years old children .......................................  1.6  1.3  6.5  7.0 
+Children with diarrhoea in the last 2 weeks and received ORS11 .................................................  100.0  100.0  54.8  53.3 
+Children with diarrhoea in the last 2 weeks and sought advice/treatment ...................................  100.0  50.0  72.9  73.3 
+Prevalence of ARI in last 2 weeks for under 5 years old children ............................................   4.3  3.9  3.9  4.2 
+Children with acute respiratory infection or fever in last 2 weeks and sought advice/treatment  37.5  33.3  69.8  68.0 
+Children with diarrhoea in the last 2 weeks given Zinc along with ORS ......................................  66.6  50.0  NA  NA","","","6.5","7.0"
+"","","","54.8","53.3"
+"","","","72.9","73.3"
+"","","","3.9","4.2"
+"","","","69.8","68.0"
+"Awareness of RTI/STI and HIV/AIDS (%)","","","",""
+"Women who have heard of RTI/STI .............................................................................................  55.8  57.1 
+Women who have heard of HIV/AIDS ..........................................................................................  98.9  99.0 
+Women who have any symptoms of RTI/STI  ..............................................................................  13.9  13.5 
+Women who know the place to go for testing of HIV/AIDS12 .......................................................  59.9  57.1 
+Women underwent test for detecting HIV/AIDS12 ........................................................................  37.3  36.8","","","34.8","38.2"
+"","","","98.3","98.1"
+"","","","15.6","16.1"
+"","","","48.6","46.3"
+"","","","14.1","12.3"
+"Utilization of Government Health Services (%)","","","",""
+"Antenatal care ..............................................................................................................................  69.7  66.7  79.0  81.0 
+Treatment for pregnancy complications .......................................................................................  57.1  59.3  88.0  87.8 
+Treatment for post-delivery complications ...................................................................................  33.3  33.3  68.4  68.4 
+Treatment for vaginal discharge ...................................................................................................  20.0  25.0  73.9  71.4 
+Treatment for children with diarrhoea13 ........................................................................................  50.0  100.0  NA  NA 
+Treatment for children with ARI13 .................................................................................................  NA  NA  NA  NA","","","79.0","81.0"
+"","","","88.0","87.8"
+"","","","68.4","68.4"
+"","","","73.9","71.4"
+"Birth Registration (%)","","","",""
+"Children below age 5 years having birth registration done ..........................................................  40.6  44.3  NA  NA 
+Children below age 5 years who received birth certificate (out of those registered) ....................  65.9  63.6  NA  NA","","","",""
+"Personal Habits (age 15 years and above) (%)","","","",""
+"Men who use any kind of smokeless tobacco  .............................................................................  74.6  74.2  NA  NA 
+Women who use any kind of smokeless tobacco ........................................................................  59.5  58.9  NA  NA 
+Men who smoke ...........................................................................................................................  56.0  56.4  NA  NA 
+Women who smoke ......................................................................................................................  18.4  18.0  NA  NA 
+Men who consume alcohol ...........................................................................................................  58.4  58.2  NA  NA 
+Women who consume alcohol .....................................................................................................  10.9  9.3  NA  NA","","","",""
+"9 Children Who were given nothing but breast milk till the survey date  10Acute Respiratory Infections11Oral Rehydration Solutions/Salts.12Based on","","","",""
+"the women who have heard of HIV/AIDS.13 Last two weeks","","","",""

docs/benchmark/stream/health/health-data-pdfplumber.csv

@@ -0,0 +1,44 @@
+"0","1","2","3","4","5","6","7","8","9","10","11","12","13","14","15","16","17","18","19","20","21","22","23"
+"","Table: 5            Public Health Outlay 2012-13 (Budget Estimates)        (Rs. in 000)","","","","","","","","","","","","","","","","","","","","","",""
+"","States-A","","","Revenue","","","","","","Capital","","","","","","Total","","","Others(1)","","","Total",""
+"","","","","","","","","","","","","","","","","Revenue &","","","","","","",""
+"","","","Medical &  Family  Medical &  Family 
+Public  Welfare  Public  Welfare 
+Health  Health","","","","","","","","","","","","","","","","","","","",""
+"","","","","","","","","","","","","","","","","Capital","","","","","","",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"","Andhra Pradesh","","","47,824,589","","","9,967,837","","","1,275,000","","","15,000","","","59,082,426","","","14,898,243","","","73,980,669",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Arunachal Pradesh   2,241,609  107,549  23,000  0  2,372,158  86,336  2,458,494","","","","","","","","","","","","","","","","","","","","","","",""
+"","Assam","","","14,874,821","","","2,554,197","","","161,600","","","0","","","17,590,618","","","4,408,505","","","21,999,123",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Bihar   21,016,708  4,332,141  5,329,000  0  30,677,849  2,251,571  32,929,420","","","","","","","","","","","","","","","","","","","","","","",""
+"","Chhattisgarh","","","11,427,311","","","1,415,660","","","2,366,592","","","0","","","15,209,563","","","311,163","","","15,520,726",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Delhi  28,084,780  411,700  4,550,000  0  33,046,480  5,000  33,051,480","","","","","","","","","","","","","","","","","","","","","","",""
+"","Goa","","","4,055,567","","","110,000","","","330,053","","","0","","","4,495,620","","","12,560","","","4,508,180",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Gujarat  26,328,400  6,922,900  12,664,000  42,000  45,957,300  455,860  46,413,160","","","","","","","","","","","","","","","","","","","","","","",""
+"","Haryana","","","15,156,681","","","1,333,527","","","40,100","","","0","","","16,530,308","","","1,222,698","","","17,753,006",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Himachal Pradesh  8,647,229  1,331,529  580,800  0  10,559,558  725,315  11,284,873","","","","","","","","","","","","","","","","","","","","","","",""
+"","Jammu & Kashmir","","","14,411,984","","","270,840","","","3,188,550","","","0","","","17,871,374","","","166,229","","","18,037,603",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Jharkhand  8,185,079  3,008,077  3,525,558  0  14,718,714  745,139  15,463,853","","","","","","","","","","","","","","","","","","","","","","",""
+"","Karnataka","","","34,939,843","","","4,317,801","","","3,669,700","","","0","","","42,927,344","","","631,088","","","43,558,432",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Kerala  27,923,965  3,985,473  929,503  0  32,838,941  334,640  33,173,581","","","","","","","","","","","","","","","","","","","","","","",""
+"","Madhya Pradesh","","","28,459,540","","","4,072,016","","","3,432,711","","","0","","","35,964,267","","","472,139","","","36,436,406",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Maharashtra  55,011,100  6,680,721  5,038,576  0  66,730,397  313,762  67,044,159","","","","","","","","","","","","","","","","","","","","","","",""
+"","Manipur","","","2,494,600","","","187,700","","","897,400","","","0","","","3,579,700","","","0","","","3,579,700",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Meghalaya  2,894,093  342,893  705,500  5,000  3,947,486  24,128  3,971,614","","","","","","","","","","","","","","","","","","","","","","",""
+"","Mizoram","","","1,743,501","","","84,185","","","10,250","","","0","","","1,837,936","","","17,060","","","1,854,996",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Nagaland   2,368,724  204,329  226,400  0  2,799,453  783,054  3,582,507","","","","","","","","","","","","","","","","","","","","","","",""
+"","Odisha","","","14,317,179","","","2,552,292","","","1,107,250","","","0","","","17,976,721","","","451,438","","","18,428,159",""
+"","","","","","","","","","","","","","","","","","","","","","","",""
+"Puducherry  4,191,757  52,249  192,400  0  4,436,406  2,173  4,438,579","","","","","","","","","","","","","","","","","","","","","","",""
+"","Punjab","","","19,775,485","","","2,208,343","","","2,470,882","","","0","","","24,454,710","","","1,436,522","","","25,891,232",""
+"","","","","","","","","","","","","","","","","","","","","","","",""

docs/benchmark/stream/missing_values/missing_values-data-pdfplumber.csv

@@ -0,0 +1,71 @@
+"0","1","2","3","4"
+"","DLHS-4 (2012-13)","","DLHS-3 (2007-08)",""
+"Indicators","TOTAL","RURAL","TOTAL","RURAL"
+"Reported Prevalence of Morbidity","","","",""
+"Any Injury .....................................................................................................................................   1.9  2.1   
+Acute Illness .................................................................................................................................   4.5  5.6   
+Chronic Illness ..............................................................................................................................   5.1  4.1","","","",""
+"","","","",""
+"","","","",""
+"Reported Prevalence of Chronic Illness during last one year (%)","","","",""
+"Disease of respiratory system ......................................................................................................   11.7  15.0   
+Disease of cardiovascular system ................................................................................................   8.9  9.3   
+Persons suffering from tuberculosis .............................................................................................   2.2  1.5","","","",""
+"","","","",""
+"","","","",""
+"Anaemia Status by Haemoglobin Level14 (%)","","","",""
+"Children (6-59 months) having anaemia ......................................................................................   68.5  71.9   
+Children (6-59 months) having severe anaemia ..........................................................................   6.7  9.4   
+Children (6-9 Years) having anaemia - Male ................................................................................   67.1  71.4   
+Children (6-9 Years) having severe anaemia - Male ....................................................................   4.4  2.4   
+Children (6-9 Years) having anaemia - Female ...........................................................................   52.4  48.8   
+Children (6-9 Years) having severe anaemia - Female ................................................................   1.2  0.0   
+Children (6-14 years) having  anaemia - Male .............................................................................   50.8  62.5   
+Children (6-14 years) having severe anaemia - Male ..................................................................   3.7  3.6   
+Children (6-14 years) having  anaemia - Female .........................................................................   48.3  50.0   
+Children (6-14 years) having severe anaemia - Female ..............................................................   4.3  6.1   
+Children (10-19 Years15) having anaemia - Male .........................................................................   37.9  51.2   
+Children (10-19 Years15) having severe anaemia - Male .............................................................   3.5  4.0   
+Children (10-19 Years15) having anaemia - Female .....................................................................   46.6  52.1   
+Children (10-19 Years15) having severe anaemia - Female .........................................................   6.4  6.5   
+Adolescents (15-19 years) having  anaemia ................................................................................   39.4  46.5   
+Adolescents (15-19 years) having severe anaemia .....................................................................   5.4  5.1   
+Pregnant women (15-49 aged) having anaemia ..........................................................................   48.8  51.5   
+Pregnant women (15-49 aged) having severe anaemia ..............................................................   7.1  8.8   
+Women (15-49 aged) having anaemia .........................................................................................   45.2  51.7   
+Women (15-49 aged) having severe anaemia .............................................................................   4.8  5.9   
+Persons (20 years and above) having anaemia ...........................................................................   37.8  42.1   
+Persons (20 years and above) having Severe anaemia ..............................................................   4.6  4.8","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"","","","",""
+"Blood Sugar Level (age 18 years and above) (%)","","","",""
+"Blood Sugar Level >140 mg/dl (high) ...........................................................................................   12.9  11.1   
+Blood Sugar Level >160 mg/dl (very high) ...................................................................................   7.0  5.1","","","",""
+"","","","",""
+"Hypertension (age 18 years and above) (%)","","","",""
+"Above Normal Range (Systolic >140 mm of Hg & Diastolic >90 mm of Hg )  ..............................   23.8  22.8   
+Moderately High (Systolic >160 mm of Hg & Diastolic >100 mm of Hg ) .....................................   8.2  7.1   
+Very High (Systolic >180 mm of Hg & Diastolic >110 mm of Hg ) ...............................................   3.7  3.1","","","",""
+"","","","",""
+"","","","",""
+"14 Any anaemia below 11g/dl, severe anaemia below 7g/dl. 15 Excluding age group 19 years","","","",""
+"Chronic Illness :Any person with symptoms persisting for longer than one month is defined as suffering from chronic illness","","","",""

docs/conf.py

@@ -22,8 +22,8 @@ import sys
 # sys.path.insert(0, os.path.abspath('..'))
 
 # Insert Camelot's path into the system.
-sys.path.insert(0, os.path.abspath('..'))
+sys.path.insert(0, os.path.abspath(".."))
-sys.path.insert(0, os.path.abspath('_themes'))
+sys.path.insert(0, os.path.abspath("_themes"))
 
 import camelot
 
@@ -38,33 +38,33 @@ import camelot
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
 extensions = [
-    'sphinx.ext.autodoc',
+    "sphinx.ext.autodoc",
-    'sphinx.ext.napoleon',
+    "sphinx.ext.napoleon",
-    'sphinx.ext.intersphinx',
+    "sphinx.ext.intersphinx",
-    'sphinx.ext.todo',
+    "sphinx.ext.todo",
-    'sphinx.ext.viewcode',
+    "sphinx.ext.viewcode",
 ]
 
 # Add any paths that contain templates here, relative to this directory.
-templates_path = ['_templates']
+templates_path = ["_templates"]
 
 # The suffix(es) of source filenames.
 # You can specify multiple suffix as a list of string:
 #
 # source_suffix = ['.rst', '.md']
-source_suffix = '.rst'
+source_suffix = ".rst"
 
 # The encoding of source files.
 #
 # source_encoding = 'utf-8-sig'
 
 # The master toctree document.
-master_doc = 'index'
+master_doc = "index"
 
 # General information about the project.
-project = u'Camelot'
+project = u"Camelot"
-copyright = u'2018, Peeply Private Ltd (Singapore)'
+copyright = u"2021, Camelot Developers"
-author = u'Vinayak Mehta'
+author = u"Vinayak Mehta"
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
@@ -94,7 +94,7 @@ language = None
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This patterns also effect to html_static_path and html_extra_path
-exclude_patterns = ['_build']
+exclude_patterns = ["_build"]
 
 # The reST default role (used for this markup: `text`) to use for all
 # documents.
@@ -114,7 +114,7 @@ add_module_names = True
 # show_authors = False
 
 # The name of the Pygments (syntax highlighting) style to use.
-pygments_style = 'flask_theme_support.FlaskyStyle'
+pygments_style = "flask_theme_support.FlaskyStyle"
 
 # A list of ignored prefixes for module index sorting.
 # modindex_common_prefix = []
@@ -130,18 +130,18 @@ todo_include_todos = True
 
 # The theme to use for HTML and HTML Help pages.  See the documentation for
 # a list of builtin themes.
-html_theme = 'alabaster'
+html_theme = "alabaster"
 
 # Theme options are theme-specific and customize the look and feel of a theme
 # further.  For a list of options available for each theme, see the
 # documentation.
 html_theme_options = {
-    'show_powered_by': False,
+    "show_powered_by": False,
-    'github_user': 'socialcopsdev',
+    "github_user": "camelot-dev",
-    'github_repo': 'camelot',
+    "github_repo": "camelot",
-    'github_banner': True,
+    "github_banner": True,
-    'show_related': False,
+    "show_related": False,
-    'note_bg': '#FFF59C'
+    "note_bg": "#FFF59C",
 }
 
 # Add any paths that contain custom themes here, relative to this directory.
@@ -164,12 +164,12 @@ html_theme_options = {
 # The name of an image file (relative to this directory) to use as a favicon of
 # the docs.  This file should be a Windows icon file (.ico) being 16x16 or 32x32
 # pixels large.
-html_favicon = '_static/favicon.ico'
+html_favicon = "_static/favicon.ico"
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+html_static_path = ["_static"]
 
 # Add any extra paths that contain custom files (such as robots.txt or
 # .htaccess) here, relative to this directory. These files are copied
@@ -189,10 +189,21 @@ html_use_smartypants = True
 
 # Custom sidebar templates, maps document names to template names.
 html_sidebars = {
-    'index': ['sidebarintro.html', 'relations.html', 'sourcelink.html',
+    "index": [
-              'searchbox.html', 'hacks.html'],
+        "sidebarintro.html",
-    '**': ['sidebarlogo.html', 'localtoc.html', 'relations.html',
+        "relations.html",
-           'sourcelink.html', 'searchbox.html', 'hacks.html']
+        "sourcelink.html",
+        "searchbox.html",
+        "hacks.html",
+    ],
+    "**": [
+        "sidebarlogo.html",
+        "localtoc.html",
+        "relations.html",
+        "sourcelink.html",
+        "searchbox.html",
+        "hacks.html",
+    ],
 }
 
 # Additional templates that should be rendered to pages, maps page names to
@@ -249,34 +260,30 @@ html_show_copyright = True
 # html_search_scorer = 'scorer.js'
 
 # Output file base name for HTML help builder.
-htmlhelp_basename = 'Camelotdoc'
+htmlhelp_basename = "Camelotdoc"
 
 # -- Options for LaTeX output ---------------------------------------------
 
 latex_elements = {
-     # The paper size ('letterpaper' or 'a4paper').
+    # The paper size ('letterpaper' or 'a4paper').
-     #
+    #
-     # 'papersize': 'letterpaper',
+    # 'papersize': 'letterpaper',
-
+    # The font size ('10pt', '11pt' or '12pt').
-     # The font size ('10pt', '11pt' or '12pt').
+    #
-     #
+    # 'pointsize': '10pt',
-     # 'pointsize': '10pt',
+    # Additional stuff for the LaTeX preamble.
-
+    #
-     # Additional stuff for the LaTeX preamble.
+    # 'preamble': '',
-     #
+    # Latex figure (float) alignment
-     # 'preamble': '',
+    #
-
+    # 'figure_align': 'htbp',
-     # Latex figure (float) alignment
-     #
-     # 'figure_align': 'htbp',
 }
 
 # Grouping the document tree into LaTeX files. List of tuples
 # (source start file, target name, title,
 #  author, documentclass [howto, manual, or own class]).
 latex_documents = [
-    (master_doc, 'Camelot.tex', u'Camelot Documentation',
+    (master_doc, "Camelot.tex", u"Camelot Documentation", u"Vinayak Mehta", "manual"),
-     u'Vinayak Mehta', 'manual'),
 ]
 
 # The name of an image file (relative to this directory) to place at the top of
@@ -316,10 +323,7 @@ latex_documents = [
 
 # One entry per manual page. List of tuples
 # (source start file, name, description, authors, manual section).
-man_pages = [
+man_pages = [(master_doc, "Camelot", u"Camelot Documentation", [author], 1)]
-    (master_doc, 'Camelot', u'Camelot Documentation',
-     [author], 1)
-]
 
 # If true, show URL addresses after external links.
 #
@@ -332,9 +336,15 @@ man_pages = [
 # (source start file, target name, title, author,
 #  dir menu entry, description, category)
 texinfo_documents = [
-    (master_doc, 'Camelot', u'Camelot Documentation',
+    (
-     author, 'Camelot', 'One line description of project.',
+        master_doc,
-     'Miscellaneous'),
+        "Camelot",
+        u"Camelot Documentation",
+        author,
+        "Camelot",
+        "One line description of project.",
+        "Miscellaneous",
+    ),
 ]
 
 # Documents to append as an appendix to all manuals.
@@ -356,6 +366,6 @@ texinfo_documents = [
 
 # Example configuration for intersphinx: refer to the Python standard library.
 intersphinx_mapping = {
-    'https://docs.python.org/2': None,
+    "https://docs.python.org/2": None,
-    'http://pandas.pydata.org/pandas-docs/stable': None
+    "http://pandas.pydata.org/pandas-docs/stable": None,
 }

docs/dev/contributing.rst

@@ -7,7 +7,7 @@ If you're reading this, you're probably looking to contributing to Camelot. *Tim
 
 This document will help you get started with contributing documentation, code, testing and filing issues. If you have any questions, feel free to reach out to `Vinayak Mehta`_, the author and maintainer.
 
-.. _Vinayak Mehta: https://vinayak-mehta.github.io
+.. _Vinayak Mehta: https://www.vinayakmehta.com
 
 Code Of Conduct
 ---------------
@@ -29,15 +29,15 @@ Your first contribution
 
 A great way to start contributing to Camelot is to pick an issue tagged with the `help wanted`_ or the `good first issue`_ tags. If you're unable to find a good first issue, feel free to contact the maintainer.
 
-.. _help wanted: https://github.com/socialcopsdev/camelot/labels/help%20wanted
+.. _help wanted: https://github.com/camelot-dev/camelot/labels/help%20wanted
-.. _good first issue: https://github.com/socialcopsdev/camelot/labels/good%20first%20issue
+.. _good first issue: https://github.com/camelot-dev/camelot/labels/good%20first%20issue
 
 Setting up a development environment
 ------------------------------------
 
 To install the dependencies needed for development, you can use pip::
 
-    $ pip install camelot-py[dev]
+    $ pip install "camelot-py[dev]"
 
 Alternatively, you can clone the project repository, and install using pip::
 
@@ -51,7 +51,7 @@ Submit a pull request
 
 The preferred workflow for contributing to Camelot is to fork the `project repository`_ on GitHub, clone, develop on a branch and then finally submit a pull request. Here are the steps:
 
-.. _project repository: https://github.com/socialcopsdev/camelot
+.. _project repository: https://github.com/camelot-dev/camelot
 
 1. Fork the project repository. Click on the ‘Fork’ button near the top of the page. This creates a copy of the code under your account on the GitHub.
 
@@ -134,7 +134,7 @@ Filing Issues
 
 We use `GitHub issues`_ to keep track of all issues and pull requests. Before opening an issue (which asks a question or reports a bug), please use GitHub search to look for existing issues (both open and closed) that may be similar.
 
-.. _GitHub issues: https://github.com/socialcopsdev/camelot/issues
+.. _GitHub issues: https://github.com/camelot-dev/camelot/issues
 
 Questions
 ^^^^^^^^^

docs/index.rst

@@ -8,11 +8,15 @@ Camelot: PDF Table Extraction for Humans
 
 Release v\ |version|. (:ref:`Installation <install>`)
 
-.. image:: https://travis-ci.org/socialcopsdev/camelot.svg?branch=master
+.. image:: https://travis-ci.org/camelot-dev/camelot.svg?branch=master
-    :target: https://travis-ci.org/socialcopsdev/camelot
+    :target: https://travis-ci.org/camelot-dev/camelot
 
-.. image:: https://codecov.io/github/socialcopsdev/camelot/badge.svg?branch=master&service=github
+.. image:: https://readthedocs.org/projects/camelot-py/badge/?version=master
-    :target: https://codecov.io/github/socialcopsdev/camelot?branch=master
+    :target: https://camelot-py.readthedocs.io/en/master/
+    :alt: Documentation Status
+
+.. image:: https://codecov.io/github/camelot-dev/camelot/badge.svg?branch=master&service=github
+    :target: https://codecov.io/github/camelot-dev/camelot?branch=master
 
 .. image:: https://img.shields.io/pypi/v/camelot-py.svg
     :target: https://pypi.org/project/camelot-py/
@@ -23,11 +27,24 @@ Release v\ |version|. (:ref:`Installation <install>`)
 .. image:: https://img.shields.io/pypi/pyversions/camelot-py.svg
     :target: https://pypi.org/project/camelot-py/
 
-**Camelot** is a Python library that makes it easy for *anyone* to extract tables from PDF files!
+.. image:: https://badges.gitter.im/camelot-dev/Lobby.png
+    :target: https://gitter.im/camelot-dev/Lobby
+
+.. image:: https://img.shields.io/badge/code%20style-black-000000.svg
+    :target: https://github.com/ambv/black
+
+.. image:: https://img.shields.io/badge/continous%20quality-deepsource-lightgrey
+    :target: https://deepsource.io/gh/camelot-dev/camelot/?ref=repository-badge
+
+**Camelot** is a Python library that can help you extract tables from PDFs!
+
+.. note:: You can also check out `Excalibur`_, the web interface to Camelot!
+
+.. _Excalibur: https://github.com/camelot-dev/excalibur
 
 ----
 
-**Here's how you can extract tables from PDF files.** Check out the PDF used in this example `here`_.
+**Here's how you can extract tables from PDFs.** You can check out the PDF used in this example `here`_.
 
 .. _here: _static/pdf/foo.pdf
 
@@ -37,7 +54,7 @@ Release v\ |version|. (:ref:`Installation <install>`)
     >>> tables = camelot.read_pdf('foo.pdf')
     >>> tables
     <TableList n=1>
-    >>> tables.export('foo.csv', f='csv', compress=True) # json, excel, html
+    >>> tables.export('foo.csv', f='csv', compress=True) # json, excel, html, markdown, sqlite
     >>> tables[0]
     <Table shape=(7, 7)>
     >>> tables[0].parsing_report
@@ -47,51 +64,61 @@ Release v\ |version|. (:ref:`Installation <install>`)
         'order': 1,
         'page': 1
     }
-    >>> tables[0].to_csv('foo.csv') # to_json, to_excel, to_html
+    >>> tables[0].to_csv('foo.csv') # to_json, to_excel, to_html, to_markdown, to_sqlite
     >>> tables[0].df # get a pandas DataFrame!
 
 .. csv-table::
   :file: _static/csv/foo.csv
 
-There's a :ref:`command-line interface <cli>` too!
+Camelot also comes packaged with a :ref:`command-line interface <cli>`!
 
 .. note:: Camelot only works with text-based PDFs and not scanned documents. (As Tabula `explains`_, "If you can click and drag to select text in your table in a PDF viewer, then your PDF is text-based".)
 
+You can check out some frequently asked questions :ref:`here <faq>`.
+
 .. _explains: https://github.com/tabulapdf/tabula#why-tabula
 
 Why Camelot?
 ------------
 
-- **You are in control.** Unlike other libraries and tools which either give a nice output or fail miserably (with no in-between), Camelot gives you the power to tweak table extraction. (This is important since everything in the real world, including PDF table extraction, is fuzzy.)
+- **Configurability**: Camelot gives you control over the table extraction process with :ref:`tweakable settings <advanced>`.
-- *Bad* tables can be discarded based on **metrics** like accuracy and whitespace, without ever having to manually look at each table.
+- **Metrics**: You can discard bad tables based on metrics like accuracy and whitespace, without having to manually look at each table.
-- Each table is a **pandas DataFrame**, which seamlessly integrates into `ETL and data analysis workflows`_.
+- **Output**: Each table is extracted into a **pandas DataFrame**, which seamlessly integrates into `ETL and data analysis workflows`_. You can also export tables to multiple formats, which include CSV, JSON, Excel, HTML, Markdown, and Sqlite.
-- **Export** to multiple formats, including JSON, Excel and HTML.
-
-See `comparison with other PDF table extraction libraries and tools`_.
 
 .. _ETL and data analysis workflows: https://gist.github.com/vinayak-mehta/e5949f7c2410a0e12f25d3682dc9e873
-.. _comparison with other PDF table extraction libraries and tools: https://github.com/socialcopsdev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
+
+See `comparison with similar libraries and tools`_.
+
+.. _comparison with similar libraries and tools: https://github.com/camelot-dev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
+
+Support the development
+-----------------------
+
+If Camelot has helped you, please consider supporting its development with a one-time or monthly donation `on OpenCollective`_!
+
+.. _on OpenCollective: https://opencollective.com/camelot
 
 The User Guide
 --------------
 
-This part of the documentation begins with some background information about why Camelot was created, takes a small dip into the implementation details and then focuses on step-by-step instructions for getting the most out of Camelot.
+This part of the documentation begins with some background information about why Camelot was created, takes you through some implementation details, and then focuses on step-by-step instructions for getting the most out of Camelot.
 
 .. toctree::
    :maxdepth: 2
 
    user/intro
+   user/install-deps
    user/install
    user/how-it-works
    user/quickstart
    user/advanced
+   user/faq
    user/cli
 
 The API Documentation/Guide
 ---------------------------
 
-If you are looking for information on a specific function, class, or method,
+If you are looking for information on a specific function, class, or method, this part of the documentation is for you.
-this part of the documentation is for you.
 
 .. toctree::
    :maxdepth: 2
@@ -101,8 +128,7 @@ this part of the documentation is for you.
 The Contributor Guide
 ---------------------
 
-If you want to contribute to the project, this part of the documentation is for
+If you want to contribute to the project, this part of the documentation is for you.
-you.
 
 .. toctree::
    :maxdepth: 2

docs/user/advanced.rst

@@ -24,25 +24,34 @@ To process background lines, you can pass ``process_background=True``.
     >>> tables = camelot.read_pdf('background_lines.pdf', process_background=True)
     >>> tables[1].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -back background_lines.pdf
+
 .. csv-table::
   :file: ../_static/csv/background_lines.csv
 
-Plot geometry
+Visual debugging
--------------
+----------------
 
-You can use a :class:`table <camelot.core.Table>` object's :meth:`plot() <camelot.core.TableList.plot>` method to plot various geometries that were detected by Camelot while processing the PDF page. This can help you select table areas, column separators and debug bad table outputs, by tweaking different configuration parameters.
+.. note:: Visual debugging using ``plot()`` requires `matplotlib <https://matplotlib.org/>`_ which is an optional dependency. You can install it using ``$ pip install camelot-py[plot]``.
 
-The following geometries are available for plotting. You can pass them to the :meth:`plot() <camelot.core.TableList.plot>` method, which will then generate a `matplotlib <https://matplotlib.org/>`_ plot for the passed geometry.
+You can use the :class:`plot() <camelot.plotting.PlotMethods>` method to generate a `matplotlib <https://matplotlib.org/>`_ plot of various elements that were detected on the PDF page while processing it. This can help you select table areas, column separators and debug bad table outputs, by tweaking different configuration parameters.
+
+You can specify the type of element you want to plot using the ``kind`` keyword argument. The generated plot can be saved to a file by passing a ``filename`` keyword argument. The following plot types are supported:
 
 - 'text'
-- 'table'
+- 'grid'
 - 'contour'
 - 'line'
 - 'joint'
+- 'textedge'
 
-.. note:: The last three geometries can only be used with :ref:`Lattice <lattice>`, i.e. when ``flavor='lattice'``.
+.. note:: 'line' and 'joint' can only be used with :ref:`Lattice <lattice>` and 'textedge' can only be used with :ref:`Stream <stream>`.
 
-Let's generate a plot for each geometry using this `PDF <../_static/pdf/foo.pdf>`__ as an example. First, let's get all the tables out.
+Let's generate a plot for each type using this `PDF <../_static/pdf/foo.pdf>`__ as an example. First, let's get all the tables out.
 
 ::
 
@@ -50,8 +59,6 @@ Let's generate a plot for each geometry using this `PDF <../_static/pdf/foo.pdf>
     >>> tables
     <TableList n=1>
 
-.. _geometry_text:
-
 text
 ^^^^
 
@@ -59,9 +66,15 @@ Let's plot all the text present on the table's PDF page.
 
 ::
 
-    >>> tables[0].plot('text')
+    >>> camelot.plot(tables[0], kind='text').show()
 
-.. figure:: ../_static/png/geometry_text.png
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -plot text foo.pdf
+
+.. figure:: ../_static/png/plot_text.png
     :height: 674
     :width: 1366
     :scale: 50%
@@ -72,18 +85,22 @@ This, as we shall later see, is very helpful with :ref:`Stream <stream>` for not
 
 .. note:: The *x-y* coordinates shown above change as you move your mouse cursor on the image, which can help you note coordinates.
 
-.. _geometry_table:
-
 table
 ^^^^^
 
-Let's plot the table (to see if it was detected correctly or not). This geometry type, along with contour, line and joint is useful for debugging and improving the extraction output, in case the table wasn't detected correctly. (More on that later.)
+Let's plot the table (to see if it was detected correctly or not). This plot type, along with contour, line and joint is useful for debugging and improving the extraction output, in case the table wasn't detected correctly. (More on that later.)
 
 ::
 
-    >>> tables[0].plot('table')
+    >>> camelot.plot(tables[0], kind='grid').show()
 
-.. figure:: ../_static/png/geometry_table.png
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -plot grid foo.pdf
+
+.. figure:: ../_static/png/plot_table.png
     :height: 674
     :width: 1366
     :scale: 50%
@@ -92,8 +109,6 @@ Let's plot the table (to see if it was detected correctly or not). This geometry
 
 The table is perfect!
 
-.. _geometry_contour:
-
 contour
 ^^^^^^^
 
@@ -101,17 +116,21 @@ Now, let's plot all table boundaries present on the table's PDF page.
 
 ::
 
-    >>> tables[0].plot('contour')
+    >>> camelot.plot(tables[0], kind='contour').show()
 
-.. figure:: ../_static/png/geometry_contour.png
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -plot contour foo.pdf
+
+.. figure:: ../_static/png/plot_contour.png
     :height: 674
     :width: 1366
     :scale: 50%
     :alt: A plot of all contours on a PDF page
     :align: left
 
-.. _geometry_line:
-
 line
 ^^^^
 
@@ -119,17 +138,21 @@ Cool, let's plot all line segments present on the table's PDF page.
 
 ::
 
-    >>> tables[0].plot('line')
+    >>> camelot.plot(tables[0], kind='line').show()
 
-.. figure:: ../_static/png/geometry_line.png
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -plot line foo.pdf
+
+.. figure:: ../_static/png/plot_line.png
     :height: 674
     :width: 1366
     :scale: 50%
     :alt: A plot of all lines on a PDF page
     :align: left
 
-.. _geometry_joint:
-
 joint
 ^^^^^
 
@@ -137,50 +160,111 @@ Finally, let's plot all line intersections present on the table's PDF page.
 
 ::
 
-    >>> tables[0].plot('joint')
+    >>> camelot.plot(tables[0], kind='joint').show()
 
-.. figure:: ../_static/png/geometry_joint.png
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -plot joint foo.pdf
+
+.. figure:: ../_static/png/plot_joint.png
     :height: 674
     :width: 1366
     :scale: 50%
     :alt: A plot of all line intersections on a PDF page
     :align: left
 
+textedge
+^^^^^^^^
+
+You can also visualize the textedges found on a page by specifying ``kind='textedge'``. To know more about what a "textedge" is, you can see pages 20, 35 and 40 of `Anssi Nurminen's master's thesis <http://dspace.cc.tut.fi/dpub/bitstream/handle/123456789/21520/Nurminen.pdf?sequence=3>`_.
+
+::
+
+    >>> camelot.plot(tables[0], kind='textedge').show()
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -plot textedge foo.pdf
+
+.. figure:: ../_static/png/plot_textedge.png
+    :height: 674
+    :width: 1366
+    :scale: 50%
+    :alt: A plot of relevant textedges on a PDF page
+    :align: left
+
 Specify table areas
 -------------------
 
-Since :ref:`Stream <stream>` treats the whole page as a table, `for now`_, it's useful to specify table boundaries in cases such as `these <../_static/pdf/table_areas.pdf>`__. You can :ref:`plot the text <geometry_text>` on this page and note the left-top and right-bottom coordinates of the table.
+In cases such as `these <../_static/pdf/table_areas.pdf>`__, it can be useful to specify exact table boundaries. You can plot the text on this page and note the top left and bottom right coordinates of the table.
 
 Table areas that you want Camelot to analyze can be passed as a list of comma-separated strings to :meth:`read_pdf() <camelot.read_pdf>`, using the ``table_areas`` keyword argument.
 
-.. _for now: https://github.com/socialcopsdev/camelot/issues/102
-
 ::
 
     >>> tables = camelot.read_pdf('table_areas.pdf', flavor='stream', table_areas=['316,499,566,337'])
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -T 316,499,566,337 table_areas.pdf
+
 .. csv-table::
   :file: ../_static/csv/table_areas.csv
 
+.. note:: ``table_areas`` accepts strings of the form x1,y1,x2,y2 where (x1, y1) -> top-left and (x2, y2) -> bottom-right in PDF coordinate space. In PDF coordinate space, the bottom-left corner of the page is the origin, with coordinates (0, 0).
+
+Specify table regions
+---------------------
+
+However there may be cases like `[1] <../_static/pdf/table_regions.pdf>`__ and `[2] <https://github.com/camelot-dev/camelot/blob/master/tests/files/tableception.pdf>`__, where the table might not lie at the exact coordinates every time but in an approximate region.
+
+You can use the ``table_regions`` keyword argument to :meth:`read_pdf() <camelot.read_pdf>` to solve for such cases. When ``table_regions`` is specified, Camelot will only analyze the specified regions to look for tables.
+
+::
+
+    >>> tables = camelot.read_pdf('table_regions.pdf', table_regions=['170,370,560,270'])
+    >>> tables[0].df
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -R 170,370,560,270 table_regions.pdf
+
+.. csv-table::
+  :file: ../_static/csv/table_regions.csv
+
 Specify column separators
 -------------------------
 
-In cases like `these <../_static/pdf/column_separators.pdf>`__, where the text is very close to each other, it is possible that Camelot may guess the column separators' coordinates incorrectly. To correct this, you can explicitly specify the *x* coordinate for each column separator by :ref:`plotting the text <geometry_text>` on the page.
+In cases like `these <../_static/pdf/column_separators.pdf>`__, where the text is very close to each other, it is possible that Camelot may guess the column separators' coordinates incorrectly. To correct this, you can explicitly specify the *x* coordinate for each column separator by plotting the text on the page.
 
 You can pass the column separators as a list of comma-separated strings to :meth:`read_pdf() <camelot.read_pdf>`, using the ``columns`` keyword argument.
 
 In case you passed a single column separators string list, and no table area is specified, the separators will be applied to the whole page. When a list of table areas is specified and you need to specify column separators as well, **the length of both lists should be equal**. Each table area will be mapped to each column separators' string using their indices.
 
-For example, if you have specified two table areas, ``table_areas=['12,23,43,54', '20,33,55,67']``, and only want to specify column separators for the first table, you can pass an empty string for the second table in the column separators' list like this, ``columns=['10,120,200,400', '']``.
+For example, if you have specified two table areas, ``table_areas=['12,54,43,23', '20,67,55,33']``, and only want to specify column separators for the first table, you can pass an empty string for the second table in the column separators' list like this, ``columns=['10,120,200,400', '']``.
 
-Let's get back to the *x* coordinates we got from :ref:`plotting text <geometry_text>` that exists on this `PDF <../_static/pdf/column_separators.pdf>`__, and get the table out!
+Let's get back to the *x* coordinates we got from plotting the text that exists on this `PDF <../_static/pdf/column_separators.pdf>`__, and get the table out!
 
 ::
 
     >>> tables = camelot.read_pdf('column_separators.pdf', flavor='stream', columns=['72,95,209,327,442,529,566,606,683'])
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -C 72,95,209,327,442,529,566,606,683 column_separators.pdf
+
 .. csv-table::
 
     "...","...","...","...","...","...","...","...","...","..."
@@ -200,6 +284,12 @@ To deal with cases like the output from the previous section, you can pass ``spl
     >>> tables = camelot.read_pdf('column_separators.pdf', flavor='stream', columns=['72,95,209,327,442,529,566,606,683'], split_text=True)
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot -split stream -C 72,95,209,327,442,529,566,606,683 column_separators.pdf
+
 .. csv-table::
 
     "...","...","...","...","...","...","...","...","...","..."
@@ -220,13 +310,19 @@ In this case, the text that `other tools`_ return, will be ``24.912``. This is r
 
 You can solve this by passing ``flag_size=True``, which will enclose the superscripts and subscripts with ``<s></s>``, based on font size, as shown below.
 
-.. _other tools: https://github.com/socialcopsdev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
+.. _other tools: https://github.com/camelot-dev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
 
 ::
 
     >>> tables = camelot.read_pdf('superscript.pdf', flavor='stream', flag_size=True)
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot -flag stream superscript.pdf
+
 .. csv-table::
 
     "...","...","...","...","...","...","...","...","...","...","..."
@@ -235,10 +331,85 @@ You can solve this by passing ``flag_size=True``, which will enclose the supersc
     "Madhya Pradesh","27.13","23.57","-","-","3.56","0.38","-","1.86","-","1.28"
     "...","...","...","...","...","...","...","...","...","...","..."
 
-Control how text is grouped into rows
+Strip characters from text
--------------------------------------
+--------------------------
 
-You can pass ``row_close_tol=<+int>`` to group the rows closer together, as shown below.
+You can strip unwanted characters like spaces, dots and newlines from a string using the ``strip_text`` keyword argument. Take a look at `this PDF <https://github.com/camelot-dev/camelot/blob/master/tests/files/tabula/12s0324.pdf>`_ as an example, the text at the start of each row contains a lot of unwanted spaces, dots and newlines.
+
+::
+
+    >>> tables = camelot.read_pdf('12s0324.pdf', flavor='stream', strip_text=' .\n')
+    >>> tables[0].df
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot -strip ' .\n' stream 12s0324.pdf
+
+.. csv-table::
+
+    "...","...","...","...","...","...","...","...","...","..."
+    "Forcible rape","17.5","2.6","14.9","17.2","2.5","14.7","–","–","–"
+    "Robbery","102.1","25.5","76.6","90.0","22.9","67.1","12.1","2.5","9.5"
+    "Aggravated assault","338.4","40.1","298.3","264.0","30.2","233.8","74.4","9.9","64.5"
+    "Property crime","1,396 .4","338 .7","1,057 .7","875 .9","210 .8","665 .1","608 .2","127 .9","392 .6"
+    "Burglary","240.9","60.3","180.6","205.0","53.4","151.7","35.9","6.9","29.0"
+    "...","...","...","...","...","...","...","...","...","..."
+
+Improve guessed table areas
+---------------------------
+
+While using :ref:`Stream <stream>`, automatic table detection can fail for PDFs like `this one <https://github.com/camelot-dev/camelot/blob/master/tests/files/edge_tol.pdf>`_. That's because the text is relatively far apart vertically, which can lead to shorter textedges being calculated.
+
+.. note:: To know more about how textedges are calculated to guess table areas, you can see pages 20, 35 and 40 of `Anssi Nurminen's master's thesis <http://dspace.cc.tut.fi/dpub/bitstream/handle/123456789/21520/Nurminen.pdf?sequence=3>`_.
+
+Let's see the table area that is detected by default.
+
+::
+
+    >>> tables = camelot.read_pdf('edge_tol.pdf', flavor='stream')
+    >>> camelot.plot(tables[0], kind='contour').show()
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -plot contour edge.pdf
+
+.. figure:: ../_static/png/edge_tol_1.png
+    :height: 674
+    :width: 1366
+    :scale: 50%
+    :alt: Table area with default edge_tol
+    :align: left
+
+To improve the detected area, you can increase the ``edge_tol`` (default: 50) value to counter the effect of text being placed relatively far apart vertically. Larger ``edge_tol`` will lead to longer textedges being detected, leading to an improved guess of the table area. Let's use a value of 500.
+
+::
+
+    >>> tables = camelot.read_pdf('edge_tol.pdf', flavor='stream', edge_tol=500)
+    >>> camelot.plot(tables[0], kind='contour').show()
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -e 500 -plot contour edge.pdf
+
+.. figure:: ../_static/png/edge_tol_2.png
+    :height: 674
+    :width: 1366
+    :scale: 50%
+    :alt: Table area with default edge_tol
+    :align: left
+
+As you can see, the guessed table area has improved!
+
+Improve guessed table rows
+--------------------------
+
+You can pass ``row_tol=<+int>`` to group the rows closer together, as shown below.
 
 ::
 
@@ -256,9 +427,15 @@ You can pass ``row_close_tol=<+int>`` to group the rows closer together, as show
 
 ::
 
-    >>> tables = camelot.read_pdf('group_rows.pdf', flavor='stream', row_close_tol=10)
+    >>> tables = camelot.read_pdf('group_rows.pdf', flavor='stream', row_tol=10)
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot stream -r 10 group_rows.pdf
+
 .. csv-table::
 
     "Clave","Nombre Entidad","Clave","","Nombre Municipio","Clave","Nombre Localidad"
@@ -270,11 +447,11 @@ You can pass ``row_close_tol=<+int>`` to group the rows closer together, as show
 Detect short lines
 ------------------
 
-There might be cases while using :ref:`Lattice <lattice>` when smaller lines don't get detected. The size of the smallest line that gets detected is calculated by dividing the PDF page's dimensions with a scaling factor called ``line_size_scaling``. By default, its value is 15.
+There might be cases while using :ref:`Lattice <lattice>` when smaller lines don't get detected. The size of the smallest line that gets detected is calculated by dividing the PDF page's dimensions with a scaling factor called ``line_scale``. By default, its value is 15.
 
-As you can guess, the larger the ``line_size_scaling``, the smaller the size of lines getting detected.
+As you can guess, the larger the ``line_scale``, the smaller the size of lines getting detected.
 
-.. warning:: Making ``line_size_scaling`` very large (>150) will lead to text getting detected as lines.
+.. warning:: Making ``line_scale`` very large (>150) will lead to text getting detected as lines.
 
 Here's a `PDF <../_static/pdf/short_lines.pdf>`__ where small lines separating the the headers don't get detected with the default value of 15.
 
@@ -282,23 +459,29 @@ Here's a `PDF <../_static/pdf/short_lines.pdf>`__ where small lines separating t
     :alt: A PDF table with short lines
     :align: left
 
-Let's :ref:`plot the table <geometry_table>` for this PDF.
+Let's plot the table for this PDF.
 
 ::
 
     >>> tables = camelot.read_pdf('short_lines.pdf')
-    >>> tables[0].plot('table')
+    >>> camelot.plot(tables[0], kind='grid').show()
 
 .. figure:: ../_static/png/short_lines_1.png
     :alt: A plot of the PDF table with short lines
     :align: left
 
-Clearly, the smaller lines separating the headers, couldn't be detected. Let's try with ``line_size_scaling=40``, and `plot the table <geometry_table>`_ again.
+Clearly, the smaller lines separating the headers, couldn't be detected. Let's try with ``line_scale=40``, and plot the table again.
 
 ::
 
-    >>> tables = camelot.read_pdf('short_lines.pdf', line_size_scaling=40)
+    >>> tables = camelot.read_pdf('short_lines.pdf', line_scale=40)
-    >>> tables[0].plot('table')
+    >>> camelot.plot(tables[0], kind='grid').show()
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -scale 40 -plot grid short_lines.pdf
 
 .. figure:: ../_static/png/short_lines_2.png
     :alt: An improved plot of the PDF table with short lines
@@ -339,7 +522,7 @@ We'll use the `PDF <../_static/pdf/short_lines.pdf>`__ from the previous example
 
 ::
 
-    >>> tables = camelot.read_pdf('short_lines.pdf', line_size_scaling=40, shift_text=[''])
+    >>> tables = camelot.read_pdf('short_lines.pdf', line_scale=40, shift_text=[''])
     >>> tables[0].df
 
 .. csv-table::
@@ -360,9 +543,15 @@ No surprises there — it did remain in place (observe the strings "2400" and "A
 
 ::
 
-    >>> tables = camelot.read_pdf('short_lines.pdf', line_size_scaling=40, shift_text=['r', 'b'])
+    >>> tables = camelot.read_pdf('short_lines.pdf', line_scale=40, shift_text=['r', 'b'])
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -scale 40 -shift r -shift b short_lines.pdf
+
 .. csv-table::
 
     "Investigations","No. ofHHs","Age/Sex/Physiological  Group","Preva-lence","C.I*","RelativePrecision","Sample sizeper State"
@@ -408,6 +597,12 @@ We don't need anything else. Now, let's pass ``copy_text=['v']`` to copy text in
     >>> tables = camelot.read_pdf('copy_text.pdf', copy_text=['v'])
     >>> tables[0].df
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot lattice -copy v copy_text.pdf
+
 .. csv-table::
 
     "Sl. No.","Name of State/UT","Name of District","Disease/ Illness","No. of Cases","No. of Deaths","Date of start of outbreak","Date of reporting","Current Status","..."
@@ -417,3 +612,37 @@ We don't need anything else. Now, let's pass ``copy_text=['v']`` to copy text in
     "4","West Bengal","West Medinipur","iv. Acute Diarrhoeal Disease","145","0","04/01/14","05/01/14","Under control","..."
     "4","West Bengal","Birbhum","v.  Food Poisoning","199","0","31/12/13","31/12/13","Under control","..."
     "4","West Bengal","Howrah","vi. Viral Hepatitis A &E","85","0","26/12/13","27/12/13","Under surveillance","..."
+
+Tweak layout generation
+-----------------------
+
+Camelot is built on top of PDFMiner's functionality of grouping characters on a page into words and sentences. In some cases (such as `#170 <https://github.com/camelot-dev/camelot/issues/170>`_ and `#215 <https://github.com/camelot-dev/camelot/issues/215>`_), PDFMiner can group characters that should belong to the same sentence into separate sentences.
+
+To deal with such cases, you can tweak PDFMiner's `LAParams kwargs <https://github.com/euske/pdfminer/blob/master/pdfminer/layout.py#L33>`_ to improve layout generation, by passing the keyword arguments as a dict using ``layout_kwargs`` in :meth:`read_pdf() <camelot.read_pdf>`. To know more about the parameters you can tweak, you can check out `PDFMiner docs <https://pdfminersix.rtfd.io/en/latest/reference/composable.html>`_.
+
+::
+
+    >>> tables = camelot.read_pdf('foo.pdf', layout_kwargs={'detect_vertical': False})
+
+.. _image-conversion-backend:
+
+Use alternate image conversion backends
+---------------------------------------
+
+When using the :ref:`Lattice <lattice>` flavor, Camelot uses ``ghostscript`` to convert PDF pages to images for line recognition. If you face installation issues with ``ghostscript``, you can use an alternate image conversion backend called ``poppler``. You can specify which image conversion backend you want to use with::
+
+    >>> tables = camelot.read_pdf(filename, backend="ghostscript")  # default
+    >>> tables = camelot.read_pdf(filename, backend="poppler")
+
+.. note:: ``ghostscript`` will be replaced by ``poppler`` as the default image conversion backend in ``v0.12.0``.
+
+If you face issues with both ``ghostscript`` and ``poppler``, you can supply your own image conversion backend::
+
+    >>> class ConversionBackend(object):
+    >>>     def convert(pdf_path, png_path):
+    >>>         # read pdf page from pdf_path
+    >>>         # convert pdf page to image
+    >>>         # write image to png_path
+    >>>         pass
+    >>>
+    >>> tables = camelot.read_pdf(filename, backend=ConversionBackend())

docs/user/cli.rst

@@ -11,12 +11,14 @@ You can print the help for the interface by typing ``camelot --help`` in your fa
 
   Usage: camelot [OPTIONS] COMMAND [ARGS]...
 
-  Camelot: PDF Table Extraction for Humans
+    Camelot: PDF Table Extraction for Humans
 
   Options:
     --version                       Show the version and exit.
+    -q, --quiet TEXT                Suppress logs and warnings.
     -p, --pages TEXT                Comma-separated page numbers. Example: 1,3,4
                                     or 1,4-end.
+    -pw, --password TEXT            Password for decryption.
     -o, --output TEXT               Output file path.
     -f, --format [csv|json|excel|html]
                                     Output file format.
@@ -24,6 +26,8 @@ You can print the help for the interface by typing ``camelot --help`` in your fa
     -split, --split_text            Split text that spans across multiple cells.
     -flag, --flag_size              Flag text based on font size. Useful to
                                     detect super/subscripts.
+    -strip, --strip_text            Characters that should be stripped from a
+                                    string before assigning it to a cell.
     -M, --margins <FLOAT FLOAT FLOAT>...
                                     PDFMiner char_margin, line_margin and
                                     word_margin.

docs/user/faq.rst

@@ -0,0 +1,70 @@
+.. _faq:
+
+Frequently Asked Questions
+==========================
+
+This part of the documentation answers some common questions. To add questions, please open an issue `here <https://github.com/camelot-dev/camelot/issues/new>`_.
+
+Does Camelot work with image-based PDFs?
+----------------------------------------
+
+**No**, Camelot only works with text-based PDFs and not scanned documents. (As Tabula `explains <https://github.com/tabulapdf/tabula#why-tabula>`_, "If you can click and drag to select text in your table in a PDF viewer, then your PDF is text-based".)
+
+How to reduce memory usage for long PDFs?
+-----------------------------------------
+
+During table extraction from long PDF documents, RAM usage can grow significantly.
+
+A simple workaround is to divide the extraction into chunks, and save extracted data to disk at the end of every chunk.
+
+For more details, check out this code snippet from `@anakin87 <https://github.com/anakin87>`_:
+
+::
+
+    import camelot
+
+
+    def chunks(l, n):
+        """Yield successive n-sized chunks from l."""
+        for i in range(0, len(l), n):
+            yield l[i : i + n]
+
+
+    def extract_tables(filepath, pages, chunks=50, export_path=".", params={}):
+        """
+        Divide the extraction work into n chunks. At the end of every chunk,
+        save data on disk and free RAM.
+
+        filepath : str
+            Filepath or URL of the PDF file.
+        pages : str, optional (default: '1')
+            Comma-separated page numbers.
+            Example: '1,3,4' or '1,4-end' or 'all'.
+        """
+
+        # get list of pages from camelot.handlers.PDFHandler
+        handler = camelot.handlers.PDFHandler(filepath)
+        page_list = handler._get_pages(filepath, pages=pages)
+
+        # chunk pages list
+        page_chunks = list(chunks(page_list, chunks))
+
+        # extraction and export
+        for chunk in page_chunks:
+            pages_string = str(chunk).replace("[", "").replace("]", "")
+            tables = camelot.read_pdf(filepath, pages=pages_string, **params)
+            tables.export(f"{export_path}/tables.csv")
+
+How can I supply my own image conversion backend to Lattice?
+------------------------------------------------------------
+
+When using the :ref:`Lattice <lattice>` flavor, you can supply your own :ref:`image conversion backend <image-conversion-backend>` by creating a class with a ``convert`` method as follows::
+
+    >>> class ConversionBackend(object):
+    >>>     def convert(pdf_path, png_path):
+    >>>         # read pdf page from pdf_path
+    >>>         # convert pdf page to image
+    >>>         # write image to png_path
+    >>>         pass
+    >>>
+    >>> tables = camelot.read_pdf(filename, backend=ConversionBackend())

docs/user/how-it-works.rst

@@ -5,24 +5,24 @@ How It Works
 
 This part of the documentation includes a high-level explanation of how Camelot extracts tables from PDF files.
 
-You can choose between two table parsing methods, *Stream* and *Lattice*. These names for parsing methods inside Camelot were inspired from `Tabula`_.
+You can choose between two table parsing methods, *Stream* and *Lattice*. These names for parsing methods inside Camelot were inspired from `Tabula <https://github.com/tabulapdf/tabula>`_.
-
-.. _Tabula: https://github.com/tabulapdf/tabula
 
 .. _stream:
 
 Stream
 ------
 
-Stream can be used to parse tables that have whitespaces between cells to simulate a table structure. It looks for these spaces between text to form a table representation.
+Stream can be used to parse tables that have whitespaces between cells to simulate a table structure. It is built on top of PDFMiner's functionality of grouping characters on a page into words and sentences, using `margins <https://euske.github.io/pdfminer/#tools>`_.
 
-It is built on top of PDFMiner's functionality of grouping characters on a page into words and sentences, using `margins`_. After getting the words on a page, it groups them into rows based on their *y* coordinates. It then tries to guess the number of columns the table might have by calculating the mode of the number of words in each row. This mode is used to calculate *x* ranges for the table's columns. It then adds columns to this column range list based on any words that may lie outside or inside the current column *x* ranges.
+1. Words on the PDF page are grouped into text rows based on their *y* axis overlaps.
 
-.. _margins: https://euske.github.io/pdfminer/#tools
+2. Textedges are calculated and then used to guess interesting table areas on the PDF page. You can read `Anssi Nurminen's master's thesis <https://pdfs.semanticscholar.org/a9b1/67a86fb189bfcd366c3839f33f0404db9c10.pdf>`_ to know more about this table detection technique. [See pages 20, 35 and 40]
 
-.. note:: By default, Stream treats the whole PDF page as a table, which isn't ideal when there are more than two tables on a page with different number of columns. Automatic table detection for Stream is `in the works`_.
+3. The number of columns inside each table area are then guessed. This is done by calculating the mode of number of words in each text row. Based on this mode, words in each text row are chosen to calculate a list of column *x* ranges.
 
-.. _in the works: https://github.com/socialcopsdev/camelot/issues/102
+4. Words that lie inside/outside the current column *x* ranges are then used to extend the current list of columns.
+
+5. Finally, a table is formed using the text rows' *y* ranges and column *x* ranges and words found on the page are assigned to the table's cells based on their *x* and *y* coordinates.
 
 .. _lattice:
 
@@ -39,7 +39,7 @@ Let's see how Lattice processes the second page of `this PDF`_, step-by-step.
 
 1. Line segments are detected.
 
-.. image:: ../_static/png/geometry_line.png
+.. image:: ../_static/png/plot_line.png
     :height: 674
     :width: 1366
     :scale: 50%
@@ -49,7 +49,7 @@ Let's see how Lattice processes the second page of `this PDF`_, step-by-step.
 
 .. _and: https://en.wikipedia.org/wiki/Logical_conjunction
 
-.. image:: ../_static/png/geometry_joint.png
+.. image:: ../_static/png/plot_joint.png
     :height: 674
     :width: 1366
     :scale: 50%
@@ -59,7 +59,7 @@ Let's see how Lattice processes the second page of `this PDF`_, step-by-step.
 
 .. _or: https://en.wikipedia.org/wiki/Logical_disjunction
 
-.. image:: ../_static/png/geometry_contour.png
+.. image:: ../_static/png/plot_contour.png
     :height: 674
     :width: 1366
     :scale: 50%
@@ -75,7 +75,7 @@ Let's see how Lattice processes the second page of `this PDF`_, step-by-step.
 
 5. Spanning cells are detected using the line segments and line intersections.
 
-.. image:: ../_static/png/geometry_table.png
+.. image:: ../_static/png/plot_table.png
     :height: 674
     :width: 1366
     :scale: 50%

docs/user/install-deps.rst

@@ -0,0 +1,62 @@
+.. _install_deps:
+
+Installation of dependencies
+============================
+
+The dependencies `Ghostscript <https://www.ghostscript.com>`_ and `Tkinter <https://wiki.python.org/moin/TkInter>`_ can be installed using your system's package manager or by running their installer.
+
+OS-specific instructions
+------------------------
+
+Ubuntu
+^^^^^^
+::
+
+    $ apt install ghostscript python3-tk
+
+MacOS
+^^^^^
+::
+
+    $ brew install ghostscript tcl-tk
+
+Windows
+^^^^^^^
+
+For Ghostscript, you can get the installer at their `downloads page <https://www.ghostscript.com/download/gsdnld.html>`_. And for Tkinter, you can download the `ActiveTcl Community Edition <https://www.activestate.com/activetcl/downloads>`_ from ActiveState.
+
+Checks to see if dependencies are installed correctly
+-----------------------------------------------------
+
+You can run the following checks to see if the dependencies were installed correctly.
+
+For Ghostscript
+^^^^^^^^^^^^^^^
+
+Open the Python REPL and run the following:
+
+For Ubuntu/MacOS::
+
+    >>> from ctypes.util import find_library
+    >>> find_library("gs")
+    "libgs.so.9"
+
+For Windows::
+
+    >>> import ctypes
+    >>> from ctypes.util import find_library
+    >>> find_library("".join(("gsdll", str(ctypes.sizeof(ctypes.c_voidp) * 8), ".dll")))
+    <name-of-ghostscript-library-on-windows>
+
+**Check:** The output of the ``find_library`` function should not be empty.
+
+If the output is empty, then it's possible that the Ghostscript library is not available one of the ``LD_LIBRARY_PATH``/``DYLD_LIBRARY_PATH``/``PATH`` variables depending on your operating system. In this case, you may have to modify one of those path variables.
+
+For Tkinter
+^^^^^^^^^^^
+
+Launch Python and then import Tkinter::
+
+    >>> import tkinter
+
+**Check:** Importing ``tkinter`` should not raise an import error.

docs/user/install.rst

@@ -3,98 +3,38 @@
 Installation of Camelot
 =======================
 
-This part of the documentation covers how to install Camelot. First, you'll need to install the dependencies, which include `Tkinter`_ and `ghostscript`_.
+This part of the documentation covers the steps to install Camelot.
 
-.. _Tkinter: https://wiki.python.org/moin/TkInter
+After :ref:`installing the dependencies <install_deps>`, which include `Ghostscript <https://www.ghostscript.com>`_ and `Tkinter <https://wiki.python.org/moin/TkInter>`_, you can use one of the following methods to install Camelot:
-.. _ghostscript: https://www.ghostscript.com
 
-Install the dependencies
+.. warning:: The ``lattice`` flavor will fail to run if Ghostscript is not installed. You may run into errors as shown in `issue #193 <https://github.com/camelot-dev/camelot/issues/193>`_.
-------------------------
 
-These can be installed using your system's package manager. You can run one of the following, based on your OS.
+pip
+---
 
-For Ubuntu
+To install Camelot from PyPI using ``pip``, please include the extra ``cv`` requirement as shown::
-^^^^^^^^^^
-::
 
-    $ apt install python-tk ghostscript
+    $ pip install "camelot-py[base]"
 
-Or for Python 3::
+conda
+-----
 
-    $ apt install python3-tk ghostscript
+`conda`_ is a package manager and environment management system for the `Anaconda <https://anaconda.org>`_ distribution. It can be used to install Camelot from the ``conda-forge`` channel::
 
-For macOS
+    $ conda install -c conda-forge camelot-py
-^^^^^^^^^
-::
 
-    $ brew install tcl-tk ghostscript
+From the source code
+--------------------
 
-For Windows
+After :ref:`installing the dependencies <install_deps>`, you can install Camelot from source by:
-^^^^^^^^^^^
-
-For Tkinter, you can download the `ActiveTcl Community Edition`_ from ActiveState. For ghostscript, you can get the installer at the `ghostscript downloads page`_.
-
-After installing ghostscript, you'll need to reboot your system to make sure that the ghostscript executable's path is in the windows PATH environment variable. In case you don't want to reboot, you can manually add the ghostscript executable's path to the PATH variable, `as shown here`_.
-
-.. _ActiveTcl Community Edition: https://www.activestate.com/activetcl/downloads
-.. _ghostscript downloads page: https://www.ghostscript.com/download/gsdnld.html
-.. _as shown here: https://java.com/en/download/help/path.xml
-
-----
-
-You can do the following checks to see if the dependencies were installed correctly.
-
-For Tkinter
-^^^^^^^^^^^
-
-Launch Python, and then at the prompt, type::
-
-    >>> import Tkinter
-
-Or in Python 3::
-
-    >>> import tkinter
-
-If you have Tkinter, Python will not print an error message, and if not, you will see an ``ImportError``.
-
-For ghostscript
-^^^^^^^^^^^^^^^
-
-Run the following to check the ghostscript version.
-
-For Ubuntu/macOS::
-
-    $ gs -version
-
-For Windows::
-
-    C:\> gswin64c.exe -version
-
-Or for Windows 32-bit::
-
-    C:\> gswin32c.exe -version
-
-If you have ghostscript, you should see the ghostscript version and copyright information.
-
-$ pip install camelot-py
-------------------------
-
-After installing the dependencies, you can simply use pip to install Camelot::
-
-    $ pip install camelot-py
-
-Get the source code
--------------------
-
-Alternatively, you can install from the source by:
 
 1. Cloning the GitHub repository.
 ::
 
-    $ git clone https://www.github.com/socialcopsdev/camelot
+    $ git clone https://www.github.com/camelot-dev/camelot
 
-2. Then simply using pip again.
+2. And then simply using pip again.
 ::
 
     $ cd camelot
-    $ pip install .
+    $ pip install ".[base]"

docs/user/intro.rst

@@ -27,7 +27,7 @@ Here is a `comparison`_ of Camelot's output with outputs from other open-source
 .. _pdf-table-extract: https://github.com/ashima/pdf-table-extract
 .. _PDFTables: https://pdftables.com/
 .. _Smallpdf: https://smallpdf.com
-.. _comparison: https://github.com/socialcopsdev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
+.. _comparison: https://github.com/camelot-dev/camelot/wiki/Comparison-with-other-PDF-Table-Extraction-libraries-and-tools
 
 What's in a name?
 -----------------

docs/user/quickstart.rst

@@ -14,7 +14,7 @@ Begin by importing the Camelot module::
 
     >>> import camelot
 
-Now, let's try to read a PDF. (You can check out the PDF used in this example `here`_.) Since the PDF has a table with clearly demarcated lines, we will use the :ref:`Lattice <lattice>` method here. To do that, we will set the ``mesh`` keyword argument to ``True``.
+Now, let's try to read a PDF. (You can check out the PDF used in this example `here`_.) Since the PDF has a table with clearly demarcated lines, we will use the :ref:`Lattice <lattice>` method here.
 
 .. note:: :ref:`Lattice <lattice>` is used by default. You can use :ref:`Stream <stream>` with ``flavor='stream'``.
 
@@ -56,7 +56,7 @@ Woah! The accuracy is top-notch and there is less whitespace, which means the ta
 .. csv-table::
   :file: ../_static/csv/foo.csv
 
-Looks good! You can now export the table as a CSV file using its :meth:`to_csv() <camelot.core.Table.to_csv>` method. Alternatively you can use :meth:`to_json() <camelot.core.Table.to_json>`, :meth:`to_excel() <camelot.core.Table.to_excel>` or :meth:`to_html() <camelot.core.Table.to_html>` methods to export the table as JSON, Excel and HTML files respectively.
+Looks good! You can now export the table as a CSV file using its :meth:`to_csv() <camelot.core.Table.to_csv>` method. Alternatively you can use :meth:`to_json() <camelot.core.Table.to_json>`, :meth:`to_excel() <camelot.core.Table.to_excel>` :meth:`to_html() <camelot.core.Table.to_html>` :meth:`to_markdown() <camelot.core.Table.to_markdown>` or :meth:`to_sqlite() <camelot.core.Table.to_sqlite>` methods to export the table as JSON, Excel, HTML files or a sqlite database respectively.
 
 ::
 
@@ -70,7 +70,13 @@ You can also export all tables at once, using the :class:`tables <camelot.core.T
 
     >>> tables.export('foo.csv', f='csv')
 
-This will export all tables as CSV files at the path specified. Alternatively, you can use ``f='json'``, ``f='excel'`` or ``f='html'``.
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot --format csv --output foo.csv lattice foo.pdf
+
+This will export all tables as CSV files at the path specified. Alternatively, you can use ``f='json'``, ``f='excel'``, ``f='html'``, ``f='markdown'`` or ``f='sqlite'``.
 
 .. note:: The :meth:`export() <camelot.core.TableList.export>` method exports files with a ``page-*-table-*`` suffix. In the example above, the single table in the list will be exported to ``foo-page-1-table-1.csv``. If the list contains multiple tables, multiple CSV files will be created. To avoid filling up your path with multiple files, you can use ``compress=True``, which will create a single ZIP file at your path with all the CSV files.
 
@@ -85,8 +91,42 @@ By default, Camelot only uses the first page of the PDF to extract tables. To sp
 
     >>> camelot.read_pdf('your.pdf', pages='1,2,3')
 
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot --pages 1,2,3 lattice your.pdf
+
 The ``pages`` keyword argument accepts pages as comma-separated string of page numbers. You can also specify page ranges — for example, ``pages=1,4-10,20-30`` or ``pages=1,4-10,20-end``.
 
-------------------------
+Reading encrypted PDFs
+----------------------
+
+To extract tables from encrypted PDF files you must provide a password when calling :meth:`read_pdf() <camelot.read_pdf>`.
+
+::
+
+    >>> tables = camelot.read_pdf('foo.pdf', password='userpass')
+    >>> tables
+    <TableList n=1>
+
+.. tip::
+    Here's how you can do the same with the :ref:`command-line interface <cli>`.
+    ::
+
+        $ camelot --password userpass lattice foo.pdf
+
+Currently Camelot only supports PDFs encrypted with ASCII passwords and algorithm `code 1 or 2`_. An exception is thrown if the PDF cannot be read. This may be due to no password being provided, an incorrect password, or an unsupported encryption algorithm.
+
+Further encryption support may be added in future, however in the meantime if your PDF files are using unsupported encryption algorithms you are advised to remove encryption before calling :meth:`read_pdf() <camelot.read_pdf>`. This can been successfully achieved with third-party tools such as `QPDF`_.
+
+::
+
+    $ qpdf --password=<PASSWORD> --decrypt input.pdf output.pdf
+
+.. _code 1 or 2: https://github.com/mstamy2/PyPDF2/issues/378
+.. _QPDF: https://www.github.com/qpdf/qpdf
+
+----
 
 Ready for more? Check out the :ref:`advanced <advanced>` section.

requirements-dev.txt

@@ -1,5 +0,0 @@
-codecov==2.0.15
-pytest==3.8.0
-pytest-cov==2.6.0
-pytest-runner==4.2
-Sphinx==1.7.9

requirements.txt

@@ -1,8 +0,0 @@
-click==6.7
-matplotlib==2.2.3
-numpy==1.15.2
-opencv-python==3.4.2.17
-openpyxl==2.5.8
-pandas==0.23.4
-pdfminer.six==20170720
-PyPDF2==1.26.0

setup.cfg

@@ -2,5 +2,5 @@
 test=pytest
 
 [tool:pytest]
-addopts = --verbose --cov-config .coveragerc --cov-report term --cov-report xml --cov=camelot tests
+addopts = --verbose --cov-config .coveragerc --cov-report term --cov-report xml --cov=camelot --mpl
 python_files = tests/test_*.py

setup.py

@@ -6,52 +6,78 @@ from setuptools import find_packages
 
 here = os.path.abspath(os.path.dirname(__file__))
 about = {}
-with open(os.path.join(here, 'camelot', '__version__.py'), 'r') as f:
+with open(os.path.join(here, "camelot", "__version__.py"), "r") as f:
     exec(f.read(), about)
 
-with open('README.md', 'r') as f:
+with open("README.md", "r") as f:
     readme = f.read()
 
 
+requires = [
+    "chardet>=3.0.4",
+    "click>=6.7",
+    "numpy>=1.13.3",
+    "openpyxl>=2.5.8",
+    "pandas>=0.23.4",
+    "pdfminer.six>=20200726",
+    "PyPDF2>=1.26.0",
+    "tabulate>=0.8.9",
+]
+
+base_requires = ["ghostscript>=0.7", "opencv-python>=3.4.2.17", "pdftopng>=0.2.3"]
+
+plot_requires = [
+    "matplotlib>=2.2.3",
+]
+
+dev_requires = [
+    "codecov>=2.0.15",
+    "pytest>=5.4.3",
+    "pytest-cov>=2.10.0",
+    "pytest-mpl>=0.11",
+    "pytest-runner>=5.2",
+    "Sphinx>=3.1.2",
+    "sphinx-autobuild>=2021.3.14",
+]
+
+all_requires = base_requires + plot_requires
+dev_requires = dev_requires + all_requires
+
+
 def setup_package():
-    reqs = []
+    metadata = dict(
-    with open('requirements.txt', 'r') as f:
+        name=about["__title__"],
-        for line in f:
+        version=about["__version__"],
-            reqs.append(line.strip())
+        description=about["__description__"],
-
+        long_description=readme,
-    dev_reqs = []
+        long_description_content_type="text/markdown",
-    with open('requirements-dev.txt', 'r') as f:
+        url=about["__url__"],
-        for line in f:
+        author=about["__author__"],
-            dev_reqs.append(line.strip())
+        author_email=about["__author_email__"],
-
+        license=about["__license__"],
-    metadata = dict(name=about['__title__'],
+        packages=find_packages(exclude=("tests",)),
-                    version=about['__version__'],
+        install_requires=requires,
-                    description=about['__description__'],
+        extras_require={
-                    long_description=readme,
+            "all": all_requires,
-                    long_description_content_type="text/markdown",
+            "base": base_requires,
-                    url=about['__url__'],
+            "cv": base_requires,  # deprecate
-                    author=about['__author__'],
+            "dev": dev_requires,
-                    author_email=about['__author_email__'],
+            "plot": plot_requires,
-                    license=about['__license__'],
+        },
-                    packages=find_packages(exclude=('tests',)),
+        entry_points={
-                    install_requires=reqs,
+            "console_scripts": [
-                    extras_require={
+                "camelot = camelot.cli:cli",
-                        'dev': dev_reqs
+            ],
-                    },
+        },
-                    entry_points={
+        classifiers=[
-                        'console_scripts': [
+            # Trove classifiers
-                            'camelot = camelot.cli:cli',
+            # Full list: https://pypi.python.org/pypi?%3Aaction=list_classifiers
-                        ],
+            "License :: OSI Approved :: MIT License",
-                    },
+            "Programming Language :: Python :: 3.6",
-                    classifiers=[
+            "Programming Language :: Python :: 3.7",
-                        # Trove classifiers
+            "Programming Language :: Python :: 3.8",
-                        # Full list: https://pypi.python.org/pypi?%3Aaction=list_classifiers
+        ],
-                        'License :: OSI Approved :: MIT License',
+    )
-                        'Programming Language :: Python :: 2.7',
-                        'Programming Language :: Python :: 3.5',
-                        'Programming Language :: Python :: 3.6',
-                        'Programming Language :: Python :: 3.7'
-                    ])
 
     try:
         from setuptools import setup
@@ -61,5 +87,5 @@ def setup_package():
     setup(**metadata)
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     setup_package()

tests/__init__.py

@@ -0,0 +1,3 @@
+import matplotlib
+
+matplotlib.use("agg")

tests/test_cli.py

@@ -1,7 +1,9 @@
 # -*- coding: utf-8 -*-
 
 import os
+import sys
 
+import pytest
 from click.testing import CliRunner
 
 from camelot.cli import cli
@@ -9,71 +11,181 @@ from camelot.utils import TemporaryDirectory
 
 
 testdir = os.path.dirname(os.path.abspath(__file__))
-testdir = os.path.join(testdir, 'files')
+testdir = os.path.join(testdir, "files")
+
+skip_on_windows = pytest.mark.skipif(
+    sys.platform.startswith("win"),
+    reason="Ghostscript not installed in Windows test environment",
+)
 
 
+def test_help_output():
+    runner = CliRunner()
+    prog_name = runner.get_default_prog_name(cli)
+    result = runner.invoke(cli, ["--help"])
+    output = result.output
+
+    assert prog_name == "camelot"
+    assert result.output.startswith("Usage: %(prog_name)s [OPTIONS] COMMAND" % locals())
+    assert all(
+        v in result.output
+        for v in ["Options:", "--version", "--help", "Commands:", "lattice", "stream"]
+    )
+
+
+@skip_on_windows
 def test_cli_lattice():
     with TemporaryDirectory() as tempdir:
-        infile = os.path.join(testdir, 'foo.pdf')
+        infile = os.path.join(testdir, "foo.pdf")
-        outfile = os.path.join(tempdir, 'foo.csv')
+        outfile = os.path.join(tempdir, "foo.csv")
         runner = CliRunner()
-        result = runner.invoke(cli, ['--format', 'csv', '--output', outfile,
+        result = runner.invoke(
-                                     'lattice', infile])
+            cli, ["--format", "csv", "--output", outfile, "lattice", infile]
+        )
         assert result.exit_code == 0
-        assert result.output == 'Found 1 tables\n'
+        assert "Found 1 tables" in result.output
 
-        result = runner.invoke(cli, ['--format', 'csv',
+        result = runner.invoke(cli, ["--format", "csv", "lattice", infile])
-                                     'lattice', infile])
+        output_error = "Error: Please specify output file path using --output"
-        output_error = 'Error: Please specify output file path using --output'
         assert output_error in result.output
 
-        result = runner.invoke(cli, ['--output', outfile,
+        result = runner.invoke(cli, ["--output", outfile, "lattice", infile])
-                                     'lattice', infile])
+        format_error = "Please specify output file format using --format"
-        format_error = 'Please specify output file format using --format'
         assert format_error in result.output
 
 
 def test_cli_stream():
     with TemporaryDirectory() as tempdir:
-        infile = os.path.join(testdir, 'budget.pdf')
+        infile = os.path.join(testdir, "budget.pdf")
-        outfile = os.path.join(tempdir, 'budget.csv')
+        outfile = os.path.join(tempdir, "budget.csv")
         runner = CliRunner()
-        result = runner.invoke(cli, ['--format', 'csv', '--output', outfile,
+        result = runner.invoke(
-                                     'stream', infile])
+            cli, ["--format", "csv", "--output", outfile, "stream", infile]
+        )
         assert result.exit_code == 0
-        assert result.output == 'Found 1 tables\n'
+        assert result.output == "Found 1 tables\n"
 
-        result = runner.invoke(cli, ['--format', 'csv', 'stream', infile])
+        result = runner.invoke(cli, ["--format", "csv", "stream", infile])
-        output_error = 'Error: Please specify output file path using --output'
+        output_error = "Error: Please specify output file path using --output"
         assert output_error in result.output
 
-        result = runner.invoke(cli, ['--output', outfile, 'stream', infile])
+        result = runner.invoke(cli, ["--output", outfile, "stream", infile])
-        format_error = 'Please specify output file format using --format'
+        format_error = "Please specify output file format using --format"
         assert format_error in result.output
 
 
+def test_cli_password():
+    with TemporaryDirectory() as tempdir:
+        infile = os.path.join(testdir, "health_protected.pdf")
+        outfile = os.path.join(tempdir, "health_protected.csv")
+        runner = CliRunner()
+        result = runner.invoke(
+            cli,
+            [
+                "--password",
+                "userpass",
+                "--format",
+                "csv",
+                "--output",
+                outfile,
+                "stream",
+                infile,
+            ],
+        )
+        assert result.exit_code == 0
+        assert result.output == "Found 1 tables\n"
+
+        output_error = "file has not been decrypted"
+        # no password
+        result = runner.invoke(
+            cli, ["--format", "csv", "--output", outfile, "stream", infile]
+        )
+        assert output_error in str(result.exception)
+
+        # bad password
+        result = runner.invoke(
+            cli,
+            [
+                "--password",
+                "wrongpass",
+                "--format",
+                "csv",
+                "--output",
+                outfile,
+                "stream",
+                infile,
+            ],
+        )
+        assert output_error in str(result.exception)
+
+
 def test_cli_output_format():
     with TemporaryDirectory() as tempdir:
-        infile = os.path.join(testdir, 'health.pdf')
+        infile = os.path.join(testdir, "health.pdf")
-        outfile = os.path.join(tempdir, 'health.{}')
+
         runner = CliRunner()
 
         # json
-        result = runner.invoke(cli, ['--format', 'json', '--output', outfile.format('json'),
+        outfile = os.path.join(tempdir, "health.json")
-                                     'stream', infile])
+        result = runner.invoke(
-        assert result.exit_code == 0
+            cli,
+            ["--format", "json", "--output", outfile, "stream", infile],
+        )
+        assert result.exit_code == 0, f"Output: {result.output}"
 
         # excel
-        result = runner.invoke(cli, ['--format', 'excel', '--output', outfile.format('xlsx'),
+        outfile = os.path.join(tempdir, "health.xlsx")
-                                     'stream', infile])
+        result = runner.invoke(
-        assert result.exit_code == 0
+            cli,
+            ["--format", "excel", "--output", outfile, "stream", infile],
+        )
+        assert result.exit_code == 0, f"Output: {result.output}"
 
         # html
-        result = runner.invoke(cli, ['--format', 'html', '--output', outfile.format('html'),
+        outfile = os.path.join(tempdir, "health.html")
-                                     'stream', infile])
+        result = runner.invoke(
-        assert result.exit_code == 0
+            cli,
+            ["--format", "html", "--output", outfile, "stream", infile],
+        )
+        assert result.exit_code == 0, f"Output: {result.output}"
+
+        # markdown
+        outfile = os.path.join(tempdir, "health.md")
+        result = runner.invoke(
+            cli,
+            ["--format", "markdown", "--output", outfile, "stream", infile],
+        )
+        assert result.exit_code == 0, f"Output: {result.output}"
 
         # zip
-        result = runner.invoke(cli, ['--zip', '--format', 'csv', '--output', outfile.format('csv'),
+        outfile = os.path.join(tempdir, "health.csv")
-                                     'stream', infile])
+        result = runner.invoke(
-        assert result.exit_code == 0
+            cli,
+            [
+                "--zip",
+                "--format",
+                "csv",
+                "--output",
+                outfile,
+                "stream",
+                infile,
+            ],
+        )
+        assert result.exit_code == 0, f"Output: {result.output}"
+
+
+def test_cli_quiet():
+    with TemporaryDirectory() as tempdir:
+        infile = os.path.join(testdir, "empty.pdf")
+        outfile = os.path.join(tempdir, "empty.csv")
+        runner = CliRunner()
+
+        result = runner.invoke(
+            cli, ["--format", "csv", "--output", outfile, "stream", infile]
+        )
+        assert "No tables found on page-1" in result.output
+
+        result = runner.invoke(
+            cli, ["--quiet", "--format", "csv", "--output", outfile, "stream", infile]
+        )
+        assert "No tables found on page-1" not in result.output

tests/test_common.py

@@ -1,148 +1,174 @@
 # -*- coding: utf-8 -*-
 
 import os
+import sys
 
+import pytest
 import pandas as pd
+from pandas.testing import assert_frame_equal
 
 import camelot
+from camelot.io import PDFHandler
+from camelot.core import Table, TableList
+from camelot.__version__ import generate_version
+from camelot.backends import ImageConversionBackend
 
 from .data import *
 
 testdir = os.path.dirname(os.path.abspath(__file__))
 testdir = os.path.join(testdir, "files")
 
+skip_on_windows = pytest.mark.skipif(
+    sys.platform.startswith("win"),
+    reason="Ghostscript not installed in Windows test environment",
+)
 
+
+def test_version_generation():
+    version = (0, 7, 3)
+    assert generate_version(version, prerelease=None, revision=None) == "0.7.3"
+
+
+def test_version_generation_with_prerelease_revision():
+    version = (0, 7, 3)
+    prerelease = "alpha"
+    revision = 2
+    assert (
+        generate_version(version, prerelease=prerelease, revision=revision)
+        == "0.7.3-alpha.2"
+    )
+
+
+@skip_on_windows
 def test_parsing_report():
-    parsing_report = {
+    parsing_report = {"accuracy": 99.02, "whitespace": 12.24, "order": 1, "page": 1}
-        'accuracy': 99.02,
-        'whitespace': 12.24,
-        'order': 1,
-        'page': 1
-    }
 
     filename = os.path.join(testdir, "foo.pdf")
     tables = camelot.read_pdf(filename)
     assert tables[0].parsing_report == parsing_report
 
 
-def test_stream():
+def test_password():
     df = pd.DataFrame(data_stream)
 
-    filename = os.path.join(testdir, "health.pdf")
+    filename = os.path.join(testdir, "health_protected.pdf")
-    tables = camelot.read_pdf(filename, flavor="stream")
+    tables = camelot.read_pdf(filename, password="ownerpass", flavor="stream")
-    assert df.equals(tables[0].df)
+    assert_frame_equal(df, tables[0].df)
+
+    tables = camelot.read_pdf(filename, password="userpass", flavor="stream")
+    assert_frame_equal(df, tables[0].df)
 
 
-def test_stream_table_rotated():
+def test_repr_poppler():
-    df = pd.DataFrame(data_stream_table_rotated)
-
-    filename = os.path.join(testdir, "clockwise_table_2.pdf")
-    tables = camelot.read_pdf(filename, flavor="stream")
-    assert df.equals(tables[0].df)
-
-    filename = os.path.join(testdir, "anticlockwise_table_2.pdf")
-    tables = camelot.read_pdf(filename, flavor="stream")
-    assert df.equals(tables[0].df)
-
-
-def test_stream_table_area():
-    df = pd.DataFrame(data_stream_table_area)
-
-    filename = os.path.join(testdir, "tabula/us-007.pdf")
-    tables = camelot.read_pdf(filename, flavor="stream", table_area=["320,500,573,335"])
-    assert df.equals(tables[0].df)
-
-
-def test_stream_columns():
-    df = pd.DataFrame(data_stream_columns)
-
-    filename = os.path.join(testdir, "mexican_towns.pdf")
-    tables = camelot.read_pdf(
-        filename, flavor="stream", columns=["67,180,230,425,475"], row_close_tol=10)
-    assert df.equals(tables[0].df)
-
-
-def test_stream_split_text():
-    df = pd.DataFrame(data_stream_split_text)
-
-    filename = os.path.join(testdir, "tabula/m27.pdf")
-    tables = camelot.read_pdf(
-        filename, flavor="stream", columns=["72,95,209,327,442,529,566,606,683"], split_text=True)
-    assert df.equals(tables[0].df)
-
-
-def test_stream_flag_size():
-    df = pd.DataFrame(data_stream_flag_size)
-
-    filename = os.path.join(testdir, "superscript.pdf")
-    tables = camelot.read_pdf(filename, flavor="stream", flag_size=True)
-    assert df.equals(tables[0].df)
-
-
-def test_lattice():
-    df = pd.DataFrame(data_lattice)
-
-    filename = os.path.join(
-        testdir, "tabula/icdar2013-dataset/competition-dataset-us/us-030.pdf")
-    tables = camelot.read_pdf(filename, pages="2")
-    assert df.equals(tables[0].df)
-
-
-def test_lattice_table_rotated():
-    df = pd.DataFrame(data_lattice_table_rotated)
-
-    filename = os.path.join(testdir, "clockwise_table_1.pdf")
-    tables = camelot.read_pdf(filename)
-    assert df.equals(tables[0].df)
-
-    filename = os.path.join(testdir, "anticlockwise_table_1.pdf")
-    tables = camelot.read_pdf(filename)
-    assert df.equals(tables[0].df)
-
-
-def test_lattice_table_area():
-    df = pd.DataFrame(data_lattice_table_area)
-
-    filename = os.path.join(testdir, "twotables_2.pdf")
-    tables = camelot.read_pdf(filename, table_area=["80,693,535,448"])
-    assert df.equals(tables[0].df)
-
-
-def test_lattice_process_background():
-    df = pd.DataFrame(data_lattice_process_background)
-
-    filename = os.path.join(testdir, "background_lines_1.pdf")
-    tables = camelot.read_pdf(filename, process_background=True)
-    assert df.equals(tables[1].df)
-
-
-def test_lattice_copy_text():
-    df = pd.DataFrame(data_lattice_copy_text)
-
-    filename = os.path.join(testdir, "row_span_1.pdf")
-    tables = camelot.read_pdf(filename, line_size_scaling=60, copy_text="v")
-    assert df.equals(tables[0].df)
-
-
-def test_lattice_shift_text():
-    df_lt = pd.DataFrame(data_lattice_shift_text_left_top)
-    df_disable = pd.DataFrame(data_lattice_shift_text_disable)
-    df_rb = pd.DataFrame(data_lattice_shift_text_right_bottom)
-
-    filename = os.path.join(testdir, "column_span_2.pdf")
-    tables = camelot.read_pdf(filename, line_size_scaling=40)
-    assert df_lt.equals(tables[0].df)
-
-    tables = camelot.read_pdf(filename, line_size_scaling=40, shift_text=[''])
-    assert df_disable.equals(tables[0].df)
-
-    tables = camelot.read_pdf(filename, line_size_scaling=40, shift_text=['r', 'b'])
-    assert df_rb.equals(tables[0].df)
-
-    
-def test_repr():
     filename = os.path.join(testdir, "foo.pdf")
-    tables = camelot.read_pdf(filename)
+    tables = camelot.read_pdf(filename, backend="poppler")
     assert repr(tables) == "<TableList n=1>"
     assert repr(tables[0]) == "<Table shape=(7, 7)>"
-    assert repr(tables[0].cells[0][0]) == "<Cell x1=120.48 y1=218.42 x2=164.64 y2=233.89>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=219 x2=165 y2=234>"
+
+
+@skip_on_windows
+def test_repr_ghostscript():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="ghostscript")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=218 x2=165 y2=234>"
+
+
+def test_url_poppler():
+    url = "https://camelot-py.readthedocs.io/en/master/_static/pdf/foo.pdf"
+    tables = camelot.read_pdf(url, backend="poppler")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=219 x2=165 y2=234>"
+
+
+@skip_on_windows
+def test_url_ghostscript():
+    url = "https://camelot-py.readthedocs.io/en/master/_static/pdf/foo.pdf"
+    tables = camelot.read_pdf(url, backend="ghostscript")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=218 x2=165 y2=234>"
+
+
+def test_pages_poppler():
+    url = "https://camelot-py.readthedocs.io/en/master/_static/pdf/foo.pdf"
+    tables = camelot.read_pdf(url, backend="poppler")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=219 x2=165 y2=234>"
+
+    tables = camelot.read_pdf(url, pages="1-end", backend="poppler")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=219 x2=165 y2=234>"
+
+    tables = camelot.read_pdf(url, pages="all", backend="poppler")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=219 x2=165 y2=234>"
+
+
+@skip_on_windows
+def test_pages_ghostscript():
+    url = "https://camelot-py.readthedocs.io/en/master/_static/pdf/foo.pdf"
+    tables = camelot.read_pdf(url, backend="ghostscript")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=218 x2=165 y2=234>"
+
+    tables = camelot.read_pdf(url, pages="1-end", backend="ghostscript")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=218 x2=165 y2=234>"
+
+    tables = camelot.read_pdf(url, pages="all", backend="ghostscript")
+    assert repr(tables) == "<TableList n=1>"
+    assert repr(tables[0]) == "<Table shape=(7, 7)>"
+    assert repr(tables[0].cells[0][0]) == "<Cell x1=120 y1=218 x2=165 y2=234>"
+
+
+def test_table_order():
+    def _make_table(page, order):
+        t = Table([], [])
+        t.page = page
+        t.order = order
+        return t
+
+    table_list = TableList(
+        [_make_table(2, 1), _make_table(1, 1), _make_table(3, 4), _make_table(1, 2)]
+    )
+
+    assert [(t.page, t.order) for t in sorted(table_list)] == [
+        (1, 1),
+        (1, 2),
+        (2, 1),
+        (3, 4),
+    ]
+    assert [(t.page, t.order) for t in sorted(table_list, reverse=True)] == [
+        (3, 4),
+        (2, 1),
+        (1, 2),
+        (1, 1),
+    ]
+
+
+def test_handler_pages_generator():
+    filename = os.path.join(testdir, "foo.pdf")
+
+    handler = PDFHandler(filename)
+    assert handler._get_pages("1") == [1]
+
+    handler = PDFHandler(filename)
+    assert handler._get_pages("all") == [1]
+
+    handler = PDFHandler(filename)
+    assert handler._get_pages("1-end") == [1]
+
+    handler = PDFHandler(filename)
+    assert handler._get_pages("1,2,3,4") == [1, 2, 3, 4]
+
+    handler = PDFHandler(filename)
+    assert handler._get_pages("1,2,5-10") == [1, 2, 5, 6, 7, 8, 9, 10]

tests/test_errors.py

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import os
+import sys
 import warnings
 
 import pytest
@@ -10,44 +11,145 @@ import camelot
 
 testdir = os.path.dirname(os.path.abspath(__file__))
 testdir = os.path.join(testdir, "files")
-filename = os.path.join(testdir, 'foo.pdf')
+filename = os.path.join(testdir, "foo.pdf")
+
+skip_on_windows = pytest.mark.skipif(
+    sys.platform.startswith("win"),
+    reason="Ghostscript not installed in Windows test environment",
+)
 
 
 def test_unknown_flavor():
-    message = ("Unknown flavor specified."
+    message = "Unknown flavor specified." " Use either 'lattice' or 'stream'"
-               " Use either 'lattice' or 'stream'")
+    with pytest.raises(NotImplementedError, match=message):
-    with pytest.raises(NotImplementedError, message=message):
+        tables = camelot.read_pdf(filename, flavor="chocolate")
-        tables = camelot.read_pdf(filename, flavor='chocolate')
 
 
 def test_input_kwargs():
     message = "columns cannot be used with flavor='lattice'"
-    with pytest.raises(ValueError, message=message):
+    with pytest.raises(ValueError, match=message):
-        tables = camelot.read_pdf(filename, columns=['10,20,30,40'])
+        tables = camelot.read_pdf(filename, columns=["10,20,30,40"])
 
 
 def test_unsupported_format():
-    message = 'File format not supported'
+    message = "File format not supported"
-    filename = os.path.join(testdir, 'foo.csv')
+    filename = os.path.join(testdir, "foo.csv")
-    with pytest.raises(NotImplementedError, message=message):
+    with pytest.raises(NotImplementedError, match=message):
         tables = camelot.read_pdf(filename)
 
 
-def test_stream_equal_length():
+@skip_on_windows
-    message = ("Length of table_area and columns"
+def test_no_tables_found_logs_suppressed():
-               " should be equal")
+    filename = os.path.join(testdir, "foo.pdf")
-    with pytest.raises(ValueError, message=message):
-        tables = camelot.read_pdf(filename, flavor='stream',
-            table_area=['10,20,30,40'], columns=['10,20,30,40', '10,20,30,40'])
-
-
-def test_no_tables_found():
-    filename = os.path.join(testdir, 'blank.pdf')
-    # TODO: use pytest.warns
     with warnings.catch_warnings():
-        warnings.simplefilter('error')
+        # the test should fail if any warning is thrown
+        warnings.simplefilter("error")
         try:
+            tables = camelot.read_pdf(filename, suppress_stdout=True)
+        except Warning as e:
+            warning_text = str(e)
+            pytest.fail(f"Unexpected warning: {warning_text}")
+
+
+def test_no_tables_found_warnings_suppressed():
+    filename = os.path.join(testdir, "empty.pdf")
+    with warnings.catch_warnings():
+        # the test should fail if any warning is thrown
+        warnings.simplefilter("error")
+        try:
+            tables = camelot.read_pdf(filename, suppress_stdout=True)
+        except Warning as e:
+            warning_text = str(e)
+            pytest.fail(f"Unexpected warning: {warning_text}")
+
+
+def test_no_password():
+    filename = os.path.join(testdir, "health_protected.pdf")
+    message = "file has not been decrypted"
+    with pytest.raises(Exception, match=message):
+        tables = camelot.read_pdf(filename)
+
+
+def test_bad_password():
+    filename = os.path.join(testdir, "health_protected.pdf")
+    message = "file has not been decrypted"
+    with pytest.raises(Exception, match=message):
+        tables = camelot.read_pdf(filename, password="wrongpass")
+
+
+def test_stream_equal_length():
+    message = "Length of table_areas and columns" " should be equal"
+    with pytest.raises(ValueError, match=message):
+        tables = camelot.read_pdf(
+            filename,
+            flavor="stream",
+            table_areas=["10,20,30,40"],
+            columns=["10,20,30,40", "10,20,30,40"],
+        )
+
+
+def test_image_warning():
+    filename = os.path.join(testdir, "image.pdf")
+    with warnings.catch_warnings():
+        warnings.simplefilter("error", category=UserWarning)
+        with pytest.raises(UserWarning) as e:
             tables = camelot.read_pdf(filename)
-        except Exception as e:
+            assert (
-            assert type(e).__name__ == 'UserWarning'
+                str(e.value)
-            assert str(e) == 'No tables found on page-1'
+                == "page-1 is image-based, camelot only works on text-based pages."
+            )
+
+
+def test_stream_no_tables_on_page():
+    filename = os.path.join(testdir, "empty.pdf")
+    with warnings.catch_warnings():
+        warnings.simplefilter("error")
+        with pytest.raises(UserWarning) as e:
+            tables = camelot.read_pdf(filename, flavor="stream")
+        assert str(e.value) == "No tables found on page-1"
+
+
+def test_stream_no_tables_in_area():
+    filename = os.path.join(testdir, "only_page_number.pdf")
+    with warnings.catch_warnings():
+        warnings.simplefilter("error")
+        with pytest.raises(UserWarning) as e:
+            tables = camelot.read_pdf(filename, flavor="stream")
+        assert str(e.value) == "No tables found in table area 1"
+
+
+def test_lattice_no_tables_on_page():
+    filename = os.path.join(testdir, "empty.pdf")
+    with warnings.catch_warnings():
+        warnings.simplefilter("error", category=UserWarning)
+        with pytest.raises(UserWarning) as e:
+            tables = camelot.read_pdf(filename, flavor="lattice")
+        assert str(e.value) == "No tables found on page-1"
+
+
+def test_lattice_unknown_backend():
+    message = "Unknown backend 'mupdf' specified. Please use either 'poppler' or 'ghostscript'."
+    with pytest.raises(NotImplementedError, match=message):
+        tables = camelot.read_pdf(filename, backend="mupdf")
+
+
+def test_lattice_no_convert_method():
+    class ConversionBackend(object):
+        pass
+
+    message = "must implement a 'convert' method"
+    with pytest.raises(NotImplementedError, match=message):
+        tables = camelot.read_pdf(filename, backend=ConversionBackend())
+
+
+def test_lattice_ghostscript_deprecation_warning():
+    ghostscript_deprecation_warning = (
+        "'ghostscript' will be replaced by 'poppler' as the default image conversion"
+        " backend in v0.12.0. You can try out 'poppler' with backend='poppler'."
+    )
+
+    with warnings.catch_warnings():
+        warnings.simplefilter("error")
+        with pytest.raises(DeprecationWarning) as e:
+            tables = camelot.read_pdf(filename)
+            assert str(e.value) == ghostscript_deprecation_warning

tests/test_image_conversion_backend.py

@@ -0,0 +1,60 @@
+# -*- coding: utf-8 -*-
+
+import pytest
+
+import camelot.backends.image_conversion
+from camelot.backends import ImageConversionBackend
+
+
+class PopplerBackendError(object):
+    def convert(self, pdf_path, png_path):
+        raise ValueError("Image conversion failed")
+
+
+class GhostscriptBackendError(object):
+    def convert(self, pdf_path, png_path):
+        raise ValueError("Image conversion failed")
+
+
+class GhostscriptBackendNoError(object):
+    def convert(self, pdf_path, png_path):
+        pass
+
+
+def test_poppler_backend_error_when_no_use_fallback(monkeypatch):
+    BACKENDS = {
+        "poppler": PopplerBackendError,
+        "ghostscript": GhostscriptBackendNoError,
+    }
+    monkeypatch.setattr(
+        "camelot.backends.image_conversion.BACKENDS", BACKENDS, raising=True
+    )
+    backend = ImageConversionBackend(use_fallback=False)
+
+    message = "Image conversion failed with image conversion backend 'poppler'"
+    with pytest.raises(ValueError, match=message):
+        backend.convert("foo", "bar")
+
+
+def test_ghostscript_backend_when_use_fallback(monkeypatch):
+    BACKENDS = {
+        "poppler": PopplerBackendError,
+        "ghostscript": GhostscriptBackendNoError,
+    }
+    monkeypatch.setattr(
+        "camelot.backends.image_conversion.BACKENDS", BACKENDS, raising=True
+    )
+    backend = ImageConversionBackend()
+    backend.convert("foo", "bar")
+
+
+def test_ghostscript_backend_error_when_use_fallback(monkeypatch):
+    BACKENDS = {"poppler": PopplerBackendError, "ghostscript": GhostscriptBackendError}
+    monkeypatch.setattr(
+        "camelot.backends.image_conversion.BACKENDS", BACKENDS, raising=True
+    )
+    backend = ImageConversionBackend()
+
+    message = "Image conversion failed with image conversion backend 'ghostscript'"
+    with pytest.raises(ValueError, match=message):
+        backend.convert("foo", "bar")

tests/test_lattice.py

@@ -0,0 +1,120 @@
+# -*- coding: utf-8 -*-
+
+import os
+import sys
+
+import pytest
+import pandas as pd
+from pandas.testing import assert_frame_equal
+
+import camelot
+from camelot.core import Table, TableList
+from camelot.__version__ import generate_version
+
+from .data import *
+
+testdir = os.path.dirname(os.path.abspath(__file__))
+testdir = os.path.join(testdir, "files")
+
+skip_on_windows = pytest.mark.skipif(
+    sys.platform.startswith("win"),
+    reason="Ghostscript not installed in Windows test environment",
+)
+
+
+@skip_on_windows
+def test_lattice():
+    df = pd.DataFrame(data_lattice)
+
+    filename = os.path.join(
+        testdir, "tabula/icdar2013-dataset/competition-dataset-us/us-030.pdf"
+    )
+    tables = camelot.read_pdf(filename, pages="2")
+    assert_frame_equal(df, tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_table_rotated():
+    df = pd.DataFrame(data_lattice_table_rotated)
+
+    filename = os.path.join(testdir, "clockwise_table_1.pdf")
+    tables = camelot.read_pdf(filename)
+    assert_frame_equal(df, tables[0].df)
+
+    filename = os.path.join(testdir, "anticlockwise_table_1.pdf")
+    tables = camelot.read_pdf(filename)
+    assert_frame_equal(df, tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_two_tables():
+    df1 = pd.DataFrame(data_lattice_two_tables_1)
+    df2 = pd.DataFrame(data_lattice_two_tables_2)
+
+    filename = os.path.join(testdir, "twotables_2.pdf")
+    tables = camelot.read_pdf(filename)
+    assert len(tables) == 2
+    assert df1.equals(tables[0].df)
+    assert df2.equals(tables[1].df)
+
+
+@skip_on_windows
+def test_lattice_table_regions():
+    df = pd.DataFrame(data_lattice_table_regions)
+
+    filename = os.path.join(testdir, "table_region.pdf")
+    tables = camelot.read_pdf(filename, table_regions=["170,370,560,270"])
+    assert_frame_equal(df, tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_table_areas():
+    df = pd.DataFrame(data_lattice_table_areas)
+
+    filename = os.path.join(testdir, "twotables_2.pdf")
+    tables = camelot.read_pdf(filename, table_areas=["80,693,535,448"])
+    assert_frame_equal(df, tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_process_background():
+    df = pd.DataFrame(data_lattice_process_background)
+
+    filename = os.path.join(testdir, "background_lines_1.pdf")
+    tables = camelot.read_pdf(filename, process_background=True)
+    assert_frame_equal(df, tables[1].df)
+
+
+@skip_on_windows
+def test_lattice_copy_text():
+    df = pd.DataFrame(data_lattice_copy_text)
+
+    filename = os.path.join(testdir, "row_span_1.pdf")
+    tables = camelot.read_pdf(filename, line_scale=60, copy_text="v")
+    assert_frame_equal(df, tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_shift_text():
+    df_lt = pd.DataFrame(data_lattice_shift_text_left_top)
+    df_disable = pd.DataFrame(data_lattice_shift_text_disable)
+    df_rb = pd.DataFrame(data_lattice_shift_text_right_bottom)
+
+    filename = os.path.join(testdir, "column_span_2.pdf")
+    tables = camelot.read_pdf(filename, line_scale=40)
+    assert df_lt.equals(tables[0].df)
+
+    tables = camelot.read_pdf(filename, line_scale=40, shift_text=[""])
+    assert df_disable.equals(tables[0].df)
+
+    tables = camelot.read_pdf(filename, line_scale=40, shift_text=["r", "b"])
+    assert df_rb.equals(tables[0].df)
+
+
+@skip_on_windows
+def test_lattice_arabic():
+    df = pd.DataFrame(data_arabic)
+
+    filename = os.path.join(testdir, "tabula/arabic.pdf")
+    tables = camelot.read_pdf(filename)
+    assert_frame_equal(df, tables[0].df)

tests/test_plotting.py

@@ -0,0 +1,98 @@
+# -*- coding: utf-8 -*-
+
+import os
+import sys
+
+import pytest
+
+import camelot
+
+testdir = os.path.dirname(os.path.abspath(__file__))
+testdir = os.path.join(testdir, "files")
+
+skip_on_windows = pytest.mark.skipif(
+    sys.platform.startswith("win"),
+    reason="Ghostscript not installed in Windows test environment",
+)
+
+
+@skip_on_windows
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_text_plot():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename)
+    return camelot.plot(tables[0], kind="text")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_textedge_plot():
+    filename = os.path.join(testdir, "tabula/12s0324.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    return camelot.plot(tables[0], kind="textedge")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_lattice_contour_plot_poppler():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="poppler")
+    return camelot.plot(tables[0], kind="contour")
+
+
+@skip_on_windows
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_lattice_contour_plot_ghostscript():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="ghostscript")
+    return camelot.plot(tables[0], kind="contour")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_stream_contour_plot():
+    filename = os.path.join(testdir, "tabula/12s0324.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    return camelot.plot(tables[0], kind="contour")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_line_plot_poppler():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="poppler")
+    return camelot.plot(tables[0], kind="line")
+
+
+@skip_on_windows
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_line_plot_ghostscript():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="ghostscript")
+    return camelot.plot(tables[0], kind="line")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_joint_plot_poppler():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="poppler")
+    return camelot.plot(tables[0], kind="joint")
+
+
+@skip_on_windows
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_joint_plot_ghostscript():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="ghostscript")
+    return camelot.plot(tables[0], kind="joint")
+
+
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_grid_plot_poppler():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="poppler")
+    return camelot.plot(tables[0], kind="grid")
+
+
+@skip_on_windows
+@pytest.mark.mpl_image_compare(baseline_dir="files/baseline_plots", remove_text=True)
+def test_grid_plot_ghostscript():
+    filename = os.path.join(testdir, "foo.pdf")
+    tables = camelot.read_pdf(filename, backend="ghostscript")
+    return camelot.plot(tables[0], kind="grid")

tests/test_stream.py

@@ -0,0 +1,133 @@
+# -*- coding: utf-8 -*-
+
+import os
+
+import pytest
+import pandas as pd
+from pandas.testing import assert_frame_equal
+
+import camelot
+from camelot.core import Table, TableList
+from camelot.__version__ import generate_version
+
+from .data import *
+
+testdir = os.path.dirname(os.path.abspath(__file__))
+testdir = os.path.join(testdir, "files")
+
+
+def test_stream():
+    df = pd.DataFrame(data_stream)
+
+    filename = os.path.join(testdir, "health.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_table_rotated():
+    df = pd.DataFrame(data_stream_table_rotated)
+
+    filename = os.path.join(testdir, "clockwise_table_2.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    assert_frame_equal(df, tables[0].df)
+
+    filename = os.path.join(testdir, "anticlockwise_table_2.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_two_tables():
+    df1 = pd.DataFrame(data_stream_two_tables_1)
+    df2 = pd.DataFrame(data_stream_two_tables_2)
+
+    filename = os.path.join(testdir, "tabula/12s0324.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+
+    assert len(tables) == 2
+    assert df1.equals(tables[0].df)
+    assert df2.equals(tables[1].df)
+
+
+def test_stream_table_regions():
+    df = pd.DataFrame(data_stream_table_areas)
+
+    filename = os.path.join(testdir, "tabula/us-007.pdf")
+    tables = camelot.read_pdf(
+        filename, flavor="stream", table_regions=["320,460,573,335"]
+    )
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_table_areas():
+    df = pd.DataFrame(data_stream_table_areas)
+
+    filename = os.path.join(testdir, "tabula/us-007.pdf")
+    tables = camelot.read_pdf(
+        filename, flavor="stream", table_areas=["320,500,573,335"]
+    )
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_columns():
+    df = pd.DataFrame(data_stream_columns)
+
+    filename = os.path.join(testdir, "mexican_towns.pdf")
+    tables = camelot.read_pdf(
+        filename, flavor="stream", columns=["67,180,230,425,475"], row_tol=10
+    )
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_split_text():
+    df = pd.DataFrame(data_stream_split_text)
+
+    filename = os.path.join(testdir, "tabula/m27.pdf")
+    tables = camelot.read_pdf(
+        filename,
+        flavor="stream",
+        columns=["72,95,209,327,442,529,566,606,683"],
+        split_text=True,
+    )
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_flag_size():
+    df = pd.DataFrame(data_stream_flag_size)
+
+    filename = os.path.join(testdir, "superscript.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream", flag_size=True)
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_strip_text():
+    df = pd.DataFrame(data_stream_strip_text)
+
+    filename = os.path.join(testdir, "detect_vertical_false.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream", strip_text=" ,\n")
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_edge_tol():
+    df = pd.DataFrame(data_stream_edge_tol)
+
+    filename = os.path.join(testdir, "edge_tol.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream", edge_tol=500)
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_layout_kwargs():
+    df = pd.DataFrame(data_stream_layout_kwargs)
+
+    filename = os.path.join(testdir, "detect_vertical_false.pdf")
+    tables = camelot.read_pdf(
+        filename, flavor="stream", layout_kwargs={"detect_vertical": False}
+    )
+    assert_frame_equal(df, tables[0].df)
+
+
+def test_stream_duplicated_text():
+    df = pd.DataFrame(data_stream_duplicated_text)
+
+    filename = os.path.join(testdir, "birdisland.pdf")
+    tables = camelot.read_pdf(filename, flavor="stream")
+    assert_frame_equal(df, tables[0].df)