Bump version

[MRG] Add support to read from url

HISTORY.md

@@ -4,6 +4,21 @@ Release History
 master
 ------
 
+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)
 ------------------
 

camelot/__version__.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-VERSION = (0, 5, 0)
+VERSION = (0, 6, 0)
 PRERELEASE = None # alpha, beta or rc
 REVISION = None
 

camelot/cli.py

@@ -43,6 +43,8 @@ pass_config = click.make_pass_decorator(Config)
               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
@@ -68,10 +70,10 @@ def cli(ctx, *args, **kwargs):
 @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_close_tol', default=2,
+@click.option('-l', '--line_tol', default=2,
               help='Tolerance parameter used to merge close vertical'
               ' and horizontal lines.')
-@click.option('-j', '--joint_close_tol', default=2,
+@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,
@@ -84,6 +86,8 @@ def cli(ctx, *args, **kwargs):
               ' 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.')
@@ -133,9 +137,11 @@ def lattice(c, *args, **kwargs):
               ' 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('-r', '--row_close_tol', default=2, help='Tolerance parameter'
+@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', '--col_close_tol', default=0, help='Tolerance parameter'
+@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']),

camelot/core.py

@@ -13,8 +13,6 @@ import pandas as pd
 # minimum number of vertical textline intersections for a textedge
 # to be considered valid
 TEXTEDGE_REQUIRED_ELEMENTS = 4
-# y coordinate tolerance for extending textedge
-TEXTEDGE_EXTEND_TOLERANCE = 50
 # padding added to table area on the left, right and bottom
 TABLE_AREA_PADDING = 10
 
@@ -55,11 +53,11 @@ class TextEdge(object):
         return '<TextEdge x={} y0={} y1={} align={} valid={}>'.format(
             round(self.x, 2), round(self.y0, 2), round(self.y1, 2), self.align, self.is_valid)
 
-    def update_coords(self, x, y0):
+    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=TEXTEDGE_EXTEND_TOLERANCE):
+        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
@@ -74,7 +72,8 @@ class TextEdges(object):
     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):
+    def __init__(self, edge_tol=50):
+        self.edge_tol = edge_tol
         self._textedges = {'left': [], 'right': [], 'middle': []}
 
     @staticmethod
@@ -115,7 +114,8 @@ class TextEdges(object):
             if idx is None:
                 self.add(textline, align)
             else:
-                self._textedges[align][idx].update_coords(x_coord, textline.y0)
+                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
@@ -359,7 +359,7 @@ class Table(object):
                 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.
@@ -376,11 +376,11 @@ class Table(object):
             # find closest x coord
             # iterate over y coords and find closest start and end points
             i = [i for i, t in enumerate(self.cols)
-                 if np.isclose(v[0], t[0], atol=joint_close_tol)]
+                 if np.isclose(v[0], t[0], atol=joint_tol)]
             j = [j for j, t in enumerate(self.rows)
-                 if np.isclose(v[3], t[0], atol=joint_close_tol)]
+                 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_close_tol)]
+                 if np.isclose(v[1], t[0], atol=joint_tol)]
             if not j:
                 continue
             J = j[0]
@@ -427,11 +427,11 @@ class Table(object):
             # find closest y coord
             # iterate over x coords and find closest start and end points
             i = [i for i, t in enumerate(self.rows)
-                 if np.isclose(h[1], t[0], atol=joint_close_tol)]
+                 if np.isclose(h[1], t[0], atol=joint_tol)]
             j = [j for j, t in enumerate(self.cols)
-                 if np.isclose(h[0], t[0], atol=joint_close_tol)]
+                 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_close_tol)]
+                 if np.isclose(h[2], t[0], atol=joint_tol)]
             if not j:
                 continue
             J = j[0]

camelot/handlers.py

@@ -8,7 +8,7 @@ from PyPDF2 import PdfFileReader, PdfFileWriter
 from .core import TableList
 from .parsers import Stream, Lattice
 from .utils import (TemporaryDirectory, get_page_layout, get_text_objects,
-                    get_rotation)
+                    get_rotation, is_url, download_url)
 
 
 class PDFHandler(object):
@@ -18,8 +18,8 @@ class PDFHandler(object):
 
     Parameters
     ----------
-    filename : str
-        Path to PDF file.
+    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'.
@@ -27,11 +27,13 @@ class PDFHandler(object):
         Password for decryption.
 
     """
-    def __init__(self, filename, pages='1', password=None):
-        self.filename = filename
-        if not filename.lower().endswith('.pdf'):
+    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")
-        self.pages = self._get_pages(self.filename, pages)
+        self.pages = self._get_pages(self.filepath, pages)
         if password is None:
             self.password = ''
         else:
@@ -39,13 +41,13 @@ class PDFHandler(object):
             if sys.version_info[0] < 3:
                 self.password = self.password.encode('ascii')
 
-    def _get_pages(self, filename, pages):
+    def _get_pages(self, filepath, pages):
         """Converts pages string to list of ints.
 
         Parameters
         ----------
-        filename : str
-            Path to PDF file.
+        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.
@@ -60,7 +62,7 @@ class PDFHandler(object):
         if pages == '1':
             page_numbers.append({'start': 1, 'end': 1})
         else:
-            infile = PdfFileReader(open(filename, 'rb'), strict=False)
+            infile = PdfFileReader(open(filepath, 'rb'), strict=False)
             if infile.isEncrypted:
                 infile.decrypt(self.password)
             if pages == 'all':
@@ -79,20 +81,20 @@ class PDFHandler(object):
             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
-            Path to PDF file.
+        filepath : str
+            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(self.password)
@@ -125,7 +127,7 @@ class PDFHandler(object):
                 with open(fpath, 'wb') as f:
                     outfile.write(f)
 
-    def parse(self, flavor='lattice', suppress_stdout=False, **kwargs):
+    def parse(self, flavor='lattice', suppress_stdout=False, layout_kwargs={}, **kwargs):
         """Extracts tables by calling parser.get_tables on all single
         page PDFs.
 
@@ -136,6 +138,8 @@ class PDFHandler(object):
             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.
 
@@ -148,11 +152,12 @@ class PDFHandler(object):
         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))
                      for p in self.pages]
             parser = Lattice(**kwargs) if flavor == 'lattice' else Stream(**kwargs)
             for p in pages:
-                t = parser.extract_tables(p, suppress_stdout=suppress_stdout)
+                t = parser.extract_tables(p, suppress_stdout=suppress_stdout,
+                                          layout_kwargs=layout_kwargs)
                 tables.extend(t)
         return TableList(tables)

camelot/io.py

@@ -6,7 +6,7 @@ from .utils import validate_input, remove_extra
 
 
 def read_pdf(filepath, pages='1', password=None, flavor='lattice',
-             suppress_stdout=False, **kwargs):
+             suppress_stdout=False, layout_kwargs={}, **kwargs):
     """Read PDF and return extracted tables.
 
     Note: kwargs annotated with ^ can only be used with flavor='stream'
@@ -15,7 +15,7 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
     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'.
@@ -26,6 +26,8 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
         Lattice is used by default.
     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
@@ -38,10 +40,13 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
     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)
@@ -57,10 +62,10 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
     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)
@@ -77,10 +82,8 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
         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
-        PDFMiner char_margin, line_margin and word_margin.
-
-        For more information, refer `PDFMiner docs <https://euske.github.io/pdfminer/>`_.
+    resolution* : int, optional (default: 300)
+        Resolution used for PDF to PNG conversion.
 
     Returns
     -------
@@ -98,5 +101,6 @@ def read_pdf(filepath, pages='1', password=None, flavor='lattice',
         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, **kwargs)
+        tables = p.parse(flavor=flavor, suppress_stdout=suppress_stdout,
+                         layout_kwargs=layout_kwargs, **kwargs)
         return tables

camelot/parsers/base.py

@@ -8,13 +8,11 @@ from ..utils import get_page_layout, get_text_objects
 class BaseParser(object):
     """Defines a base parser.
     """
-    def _generate_layout(self, filename):
+    def _generate_layout(self, filename, layout_kwargs):
         self.filename = filename
+        self.layout_kwargs = layout_kwargs
         self.layout, self.dimensions = get_page_layout(
-            self.filename,
-            char_margin=self.char_margin,
-            line_margin=self.line_margin,
-            word_margin=self.word_margin)
+            filename, **layout_kwargs)
         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

camelot/parsers/lattice.py

@@ -50,10 +50,13 @@ 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)
@@ -70,17 +73,15 @@ 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
-        PDFMiner char_margin, line_margin and word_margin.
-
-        For more information, refer `PDFMiner docs <https://euske.github.io/pdfminer/>`_.
+    resolution : int, optional (default: 300)
+        Resolution used for PDF to PNG conversion.
 
     """
     def __init__(self, table_areas=None, process_background=False,
                  line_size_scaling=15, copy_text=None, shift_text=['l', 't'],
-                 split_text=False, flag_size=False, line_close_tol=2,
-                 joint_close_tol=2, threshold_blocksize=15, threshold_constant=-2,
-                 iterations=0, margins=(1.0, 0.5, 0.1), **kwargs):
+                 split_text=False, flag_size=False, strip_text='', line_tol=2,
+                 joint_tol=2, threshold_blocksize=15, threshold_constant=-2,
+                 iterations=0, resolution=300, **kwargs):
         self.table_areas = table_areas
         self.process_background = process_background
         self.line_size_scaling = line_size_scaling
@@ -88,12 +89,13 @@ class Lattice(BaseParser):
         self.shift_text = shift_text
         self.split_text = split_text
         self.flag_size = flag_size
-        self.line_close_tol = line_close_tol
-        self.joint_close_tol = joint_close_tol
+        self.strip_text = strip_text
+        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
 
     @staticmethod
     def _reduce_index(t, idx, shift_text):
@@ -214,7 +216,7 @@ class Lattice(BaseParser):
             '-sDEVICE=png16m',
             '-o',
             self.imagename,
-            '-r600',
+            '-r{}'.format(self.resolution),
             self.filename
         ]
         gs = get_executable()
@@ -283,9 +285,9 @@ class Lattice(BaseParser):
         rows.extend([tk[1], tk[3]])
         # sort horizontal and vertical segments
         cols = merge_close_lines(
-            sorted(cols), line_close_tol=self.line_close_tol)
+            sorted(cols), line_tol=self.line_tol)
         rows = merge_close_lines(
-            sorted(rows, reverse=True), line_close_tol=self.line_close_tol)
+            sorted(rows, reverse=True), line_tol=self.line_tol)
         # make grid using x and y coord of shortlisted rows and cols
         cols = [(cols[i], cols[i + 1])
                 for i in range(0, len(cols) - 1)]
@@ -302,7 +304,7 @@ class Lattice(BaseParser):
 
         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
@@ -315,7 +317,7 @@ class Lattice(BaseParser):
             for t in self.t_bbox[direction]:
                 indices, error = get_table_index(
                     table, t, direction, split_text=self.split_text,
-                    flag_size=self.flag_size)
+                    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)
@@ -348,8 +350,8 @@ class Lattice(BaseParser):
 
         return table
 
-    def extract_tables(self, filename, suppress_stdout=False):
-        self._generate_layout(filename)
+    def extract_tables(self, filename, suppress_stdout=False, layout_kwargs={}):
+        self._generate_layout(filename, layout_kwargs)
         if not suppress_stdout:
             logger.info('Processing {}'.format(os.path.basename(self.rootname)))
 

camelot/parsers/stream.py

@@ -38,29 +38,31 @@ 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_areas=None, columns=None, split_text=False,
-                 flag_size=False, row_close_tol=2, col_close_tol=0,
-                 margins=(1.0, 0.5, 0.1), **kwargs):
+                 flag_size=False, strip_text='', edge_tol=50, row_tol=2,
+                 column_tol=0, **kwargs):
         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.col_close_tol = col_close_tol
-        self.char_margin, self.line_margin, self.word_margin = margins
+        self.strip_text = strip_text
+        self.edge_tol = edge_tol
+        self.row_tol = row_tol
+        self.column_tol = column_tol
 
     @staticmethod
     def _text_bbox(t_bbox):
@@ -86,7 +88,7 @@ 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 +96,7 @@ class Stream(BaseParser):
         ----------
         text : list
             List of PDFMiner text objects.
-        row_close_tol : int, optional (default: 2)
+        row_tol : int, optional (default: 2)
 
         Returns
         -------
@@ -110,7 +112,7 @@ class Stream(BaseParser):
             # 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
@@ -120,7 +122,7 @@ class Stream(BaseParser):
         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 +130,7 @@ 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 +144,17 @@ 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
-                            np.isclose(higher[0], lower[1], atol=col_close_tol)):
+                            np.isclose(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])
@@ -189,7 +191,7 @@ class Stream(BaseParser):
         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.
 
@@ -208,7 +210,7 @@ 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)
                         for r in text if len(r) == max(elements) for t in r]
@@ -254,11 +256,10 @@ class Stream(BaseParser):
         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()
+        textedges = TextEdges(edge_tol=self.edge_tol)
         # generate left, middle and right textedges
         textedges.generate(textlines)
         # select relevant edges
@@ -300,7 +301,7 @@ class Stream(BaseParser):
         self.t_bbox = t_bbox
 
         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]
 
@@ -331,7 +332,7 @@ class Stream(BaseParser):
                     warnings.warn("No tables found in table area {}".format(
                         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), col_close_tol=self.col_close_tol)
+            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]
@@ -343,7 +344,7 @@ class Stream(BaseParser):
                             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_close_tol)
+            cols = self._add_columns(cols, inner_text, self.row_tol)
             cols = self._join_columns(cols, text_x_min, text_x_max)
 
         return cols, rows
@@ -359,7 +360,7 @@ class Stream(BaseParser):
             for t in self.t_bbox[direction]:
                 indices, error = get_table_index(
                     table, t, direction, split_text=self.split_text,
-                    flag_size=self.flag_size)
+                    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:
@@ -388,8 +389,8 @@ class Stream(BaseParser):
 
         return table
 
-    def extract_tables(self, filename, suppress_stdout=False):
-        self._generate_layout(filename)
+    def extract_tables(self, filename, suppress_stdout=False, layout_kwargs={}):
+        self._generate_layout(filename, layout_kwargs)
         if not suppress_stdout:
             logger.info('Processing {}'.format(os.path.basename(self.rootname)))
 

camelot/utils.py

@@ -1,12 +1,17 @@
+# -*- coding: utf-8 -*-
 from __future__ import division
+
+import os
+import sys
+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
@@ -18,18 +23,89 @@ from pdfminer.layout import (LAParams, LTAnno, LTChar, LTTextLineHorizontal,
                              LTTextLineVertical)
 
 
+PY3 = sys.version_info[0] >= 3
+if PY3:
+    from urllib.request import urlopen
+    from urllib.parse import urlparse as parse_url
+    from urllib.parse import uses_relative, uses_netloc, uses_params
+else:
+    from urllib2 import urlopen
+    from urlparse import urlparse as parse_url
+    from urlparse import uses_relative, uses_netloc, uses_params
+
+
+_VALID_URLS = set(uses_relative + uses_netloc + uses_params)
+_VALID_URLS.discard('')
+
+
+# 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 = '{}.pdf'.format(random_string(6))
+    with tempfile.NamedTemporaryFile('wb', delete=False) as f:
+        obj = urlopen(url)
+        if PY3:
+            content_type = obj.info().get_content_type()
+        else:
+            content_type = obj.info().getheader('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',
-    'row_close_tol',
-    'col_close_tol'
+    'row_tol',
+    'column_tol'
 ]
 lattice_kwargs = [
     'process_background',
     'line_size_scaling',
     'copy_text',
     'shift_text',
-    'line_close_tol',
-    'joint_close_tol',
+    'line_tol',
+    'joint_tol',
     'threshold_blocksize',
     'threshold_constant',
     'iterations'
@@ -281,14 +357,14 @@ def text_in_bbox(bbox, text):
     return t_bbox
 
 
-def merge_close_lines(ar, line_close_tol=2):
+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 +377,7 @@ 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 +385,12 @@ def merge_close_lines(ar, line_close_tol=2):
     return ret
 
 
-def flag_font_size(textline, direction):
+# 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.
 
@@ -319,6 +400,9 @@ 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
     -------
@@ -344,13 +428,13 @@ def flag_font_size(textline, direction):
                 fchars = [t[0] for t in chars]
                 if ''.join(fchars).strip():
                     flist.append(''.join(fchars))
-        fstring = ''.join(flist)
+        fstring = ''.join(flist).strip(strip_text)
     else:
-        fstring = ''.join([t.get_text() for t in textline])
+        fstring = ''.join([t.get_text() for t in textline]).strip(strip_text)
     return fstring
 
 
-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 +449,9 @@ 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
     -------
@@ -416,14 +503,15 @@ 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)))
+            grouped_chars.append((key[0], key[1], ''.join(gchars).strip(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 +529,9 @@ 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
     -------
@@ -495,12 +586,12 @@ 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())], error
+            return [(r_idx, c_idx, t.get_text().strip(strip_text))], error
 
 
 def compute_accuracy(error_weights):

docs/user/advanced.rst

@@ -24,6 +24,12 @@ 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
 
@@ -63,6 +69,12 @@ Let's plot all the text present on the table's PDF page.
     >>> camelot.plot(tables[0], kind='text')
     >>> plt.show()
 
+.. 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
@@ -84,6 +96,12 @@ Let's plot the table (to see if it was detected correctly or not). This plot typ
     >>> camelot.plot(tables[0], kind='grid')
     >>> plt.show()
 
+.. 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
@@ -103,6 +121,12 @@ Now, let's plot all table boundaries present on the table's PDF page.
     >>> camelot.plot(tables[0], kind='contour')
     >>> plt.show()
 
+.. 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
@@ -120,6 +144,12 @@ Cool, let's plot all line segments present on the table's PDF page.
     >>> camelot.plot(tables[0], kind='line')
     >>> plt.show()
 
+.. 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
@@ -137,6 +167,12 @@ Finally, let's plot all line intersections present on the table's PDF page.
     >>> camelot.plot(tables[0], kind='joint')
     >>> plt.show()
 
+.. 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
@@ -154,6 +190,12 @@ You can also visualize the textedges found on a page by specifying ``kind='texte
     >>> camelot.plot(tables[0], kind='textedge')
     >>> plt.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
@@ -175,6 +217,12 @@ Table areas that you want Camelot to analyze can be passed as a list of comma-se
     >>> 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
 
@@ -196,6 +244,12 @@ Let's get back to the *x* coordinates we got from plotting the text that exists
     >>> 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::
 
     "...","...","...","...","...","...","...","...","...","..."
@@ -215,6 +269,12 @@ 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::
 
     "...","...","...","...","...","...","...","...","...","..."
@@ -242,6 +302,12 @@ You can solve this by passing ``flag_size=True``, which will enclose the supersc
     >>> 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::
 
     "...","...","...","...","...","...","...","...","...","...","..."
@@ -250,10 +316,87 @@ 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/socialcopsdev/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/socialcopsdev/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')
+    >>> plt.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')
+    >>> plt.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.
 
 ::
 
@@ -271,9 +414,15 @@ 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"
@@ -317,6 +466,12 @@ Clearly, the smaller lines separating the headers, couldn't be detected. Let's t
     >>> camelot.plot(tables[0], kind='grid')
     >>> plt.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
     :align: left
@@ -380,6 +535,12 @@ 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[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"
@@ -425,6 +586,12 @@ 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","..."
@@ -434,3 +601,14 @@ 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/socialcopsdev/camelot/issues/170>`_ and `#215 <https://github.com/socialcopsdev/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://euske.github.io/pdfminer/>`_.
+
+::
+
+    >>> tables = camelot.read_pdf('foo.pdf', layout_kwargs={'detect_vertical': False})

docs/user/cli.rst

@@ -15,7 +15,7 @@ You can print the help for the interface by typing ``camelot --help`` in your fa
 
   Options:
     --version                       Show the version and exit.
-    -v, --verbose                   Verbose.
+    -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.

docs/user/quickstart.rst

@@ -70,6 +70,12 @@ You can also export all tables at once, using the :class:`tables <camelot.core.T
 
     >>> tables.export('foo.csv', f='csv')
 
+.. 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'`` or ``f='html'``.
 
 .. 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.
@@ -85,6 +91,12 @@ 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
@@ -98,6 +110,12 @@ To extract tables from encrypted PDF files you must provide a password when call
     >>> 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`_.

tests/data.py

@@ -312,6 +312,63 @@ data_stream_flag_size = [
     ["ALL STATES", "513.38", "436.02", "-", "25.57", "51.06", "14.18", "-", "8.21", "11.83", "11.08"]
 ]
 
+data_stream_strip_text = [
+    ["V i n s   a u   Ve r r e", ""],
+    ["Les Blancs", "12.5CL"],
+    ["A.O.P Côtes du Rhône", ""],
+    ["Domaine de la Guicharde «  Autour de la chapelle » 2016", "8 €"],
+    ["A.O.P Vacqueyras", ""],
+    ["Domaine de Montvac  « Melodine » 2016", "10 €"],
+    ["A.O.P Châteauneuf du Pape", ""],
+    ["Domaine de Beaurenard 2017", "13 €"],
+    ["A.O.P Côteaux du Languedoc", ""],
+    ["Villa Tempora « Un temps pour elle » 2014", "9 €"],
+    ["A.O.P Côtes de Provence", ""],
+    ["Château Grand Boise 2017", "9 €"],
+    ["Les Rosés", "12,5 CL"],
+    ["A.O.P Côtes du Rhône", ""],
+    ["Domaine de la Florane « A fleur de Pampre » 2016", "8 €"],
+    ["Famille Coulon (Domaine Beaurenard) Biotifulfox 2017", "8 €"],
+    ["A.O.P Vacqueyras", ""],
+    ["Domaine de Montvac 2017", "9 €"],
+    ["A.O.P Languedoc", ""],
+    ["Domaine de Joncas « Nébla » 2015", "8 €"],
+    ["Villa Tempora « L’arroseur arrosé » 2015", "9 €"],
+    ["A.O.P Côtes de Provence", ""],
+    ["Château Grand Boise « Sainte Victoire » 2017", "9 €"],
+    ["Château Léoube 2016", "10 €"]
+]
+
+data_stream_edge_tol = [
+    ["Key figures", ""],
+    ["", "2016"],
+    ["(all amounts in EUR)", ""],
+    ["C\nlass A", ""],
+    ["N\net Asset Value at 31 December", "5,111,372"],
+    ["N\number of outstanding units at 31 December", "49,136"],
+    ["N\net Asset Value per unit at 31 December", "104.03"],
+    ["C\nlass B", ""],
+    ["N\net Asset Value at 31 December", "49,144,825"],
+    ["N\number of outstanding units at 31 December", "471,555"],
+    ["N\net Asset Value per unit at 31 December", "104.22"],
+    ["T\notal for the Fund", ""],
+    ["N\net Asset Value at 31 December", "54,256,197"],
+    ["N\number of outstanding units at 31 December", "520,691"],
+    ["I\nnvestment result", ""],
+    ["Direct result", "-"],
+    ["Revaluation", "2,076,667"],
+    ["Costs", "(106,870)"],
+    ["T\notal investment result for the period1", "1,969,797"],
+    ["I\nnvestment result per unit2", ""],
+    ["Direct result", "-"],
+    ["Revaluation", "3.99"],
+    ["Costs", "(0.21)"],
+    ["T\notal investment result per unit", "3.78"],
+    ["1 The results cover the period from inception of the Fund at 8 April 2016 through 31 December 2016.", ""],
+    ["2 The result per unit is calculated using the total number of outstanding unit as per the end of the", ""],
+    ["period.", ""]
+]
+
 data_lattice = [
     ["Cycle \nName", "KI \n(1/km)", "Distance \n(mi)", "Percent Fuel Savings", "", "", ""],
     ["", "", "", "Improved \nSpeed", "Decreased \nAccel", "Eliminate \nStops", "Decreased \nIdle"],
@@ -485,9 +542,49 @@ data_lattice_shift_text_right_bottom = [
 ]
 
 data_arabic = [
-    ['ً\n\xa0\nﺎﺒﺣﺮﻣ', 'ﻥﺎﻄﻠﺳ\xa0ﻲﻤﺳﺍ'],
-    ['ﻝﺎﻤﺸﻟﺍ\xa0ﺎﻨﻴﻟﻭﺭﺎﻛ\xa0ﺔﻳﻻﻭ\xa0ﻦﻣ\xa0ﺎﻧﺍ', '؟ﺖﻧﺍ\xa0ﻦﻳﺍ\xa0ﻦﻣ'],
-    ['1234', 'ﻂﻄﻗ\xa047\xa0ﻱﺪﻨﻋ'],
-    ['؟ﻙﺎﺒﺷ\xa0ﺖﻧﺍ\xa0ﻞﻫ', 'ﺔﻳﺰﻴﻠﺠﻧﻻﺍ\xa0ﻲﻓ\xa0Jeremy\xa0ﻲﻤﺳﺍ'],
-    ['Jeremy\xa0is\xa0ﻲﻣﺮﺟ\xa0in\xa0Arabic', '']
+    ["ً\n\xa0\nﺎﺒﺣﺮﻣ", "ﻥﺎﻄﻠﺳ\xa0ﻲﻤﺳﺍ"],
+    ["ﻝﺎﻤﺸﻟﺍ\xa0ﺎﻨﻴﻟﻭﺭﺎﻛ\xa0ﺔﻳﻻﻭ\xa0ﻦﻣ\xa0ﺎﻧﺍ", "؟ﺖﻧﺍ\xa0ﻦﻳﺍ\xa0ﻦﻣ"],
+    ["1234", "ﻂﻄﻗ\xa047\xa0ﻱﺪﻨﻋ"],
+    ["؟ﻙﺎﺒﺷ\xa0ﺖﻧﺍ\xa0ﻞﻫ", "ﺔﻳﺰﻴﻠﺠﻧﻻﺍ\xa0ﻲﻓ\xa0Jeremy\xa0ﻲﻤﺳﺍ"],
+    ["Jeremy\xa0is\xa0ﻲﻣﺮﺟ\xa0in\xa0Arabic", ""]
+]
+
+data_stream_layout_kwargs = [
+    ["V i n s   a u   Ve r r e", ""],
+    ["Les Blancs", "12.5CL"],
+    ["A.O.P Côtes du Rhône", ""],
+    ["Domaine de la Guicharde «  Autour de la chapelle » 2016", "8 €"],
+    ["A.O.P Vacqueyras", ""],
+    ["Domaine de Montvac  « Melodine » 2016", "10 €"],
+    ["A.O.P Châteauneuf du Pape", ""],
+    ["Domaine de Beaurenard 2017", "13 €"],
+    ["A.O.P Côteaux du Languedoc", ""],
+    ["Villa Tempora « Un temps pour elle » 2014", "9 €"],
+    ["A.O.P Côtes de Provence", ""],
+    ["Château Grand Boise 2017", "9 €"],
+    ["Les Rosés", "12,5 CL"],
+    ["A.O.P Côtes du Rhône", ""],
+    ["Domaine de la Florane « A fleur de Pampre » 2016", "8 €"],
+    ["Famille Coulon (Domaine Beaurenard) Biotifulfox 2017", "8 €"],
+    ["A.O.P Vacqueyras", ""],
+    ["Domaine de Montvac 2017", "9 €"],
+    ["A.O.P Languedoc", ""],
+    ["Domaine de Joncas « Nébla » 2015", "8 €"],
+    ["Villa Tempora « L’arroseur arrosé » 2015", "9 €"],
+    ["A.O.P Côtes de Provence", ""],
+    ["Château Grand Boise « Sainte Victoire » 2017", "9 €"],
+    ["Château Léoube 2016", "10 €"],
+    ["Les Rouges", "12,CL"],
+    ["A.O.P Côtes du Rhône", ""],
+    ["Domaine de Dionysos « La Cigalette »", "8 €"],
+    ["Château Saint Estève d’Uchaux « Grande Réserve » 2014", "9 €"],
+    ["Domaine de la Guicharde « Cuvée Massillan » 2016", "9 €"],
+    ["Domaine de la Florane « Terre Pourpre » 2014", "10 €"],
+    ["L’Oratoire St Martin « Réserve des Seigneurs » 2015", "11 €"],
+    ["A.O.P Saint Joseph", ""],
+    ["Domaine Monier Perréol « Châtelet » 2015", "13 €"],
+    ["A.O.P Châteauneuf du Pape", ""],
+    ["Domaine de Beaurenard 2011", "15 €"],
+    ["A.O.P Cornas", ""],
+    ["Domaine Lionnet « Terre Brûlée » 2012", "15 €"]
 ]

tests/test_common.py

@@ -81,7 +81,7 @@ def test_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)
+        filename, flavor="stream", columns=["67,180,230,425,475"], row_tol=10)
     assert df.equals(tables[0].df)
 
 
@@ -102,6 +102,31 @@ def test_stream_flag_size():
     assert df.equals(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 df.equals(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 df.equals(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 df.equals(tables[0].df)
+
+
 def test_lattice():
     df = pd.DataFrame(data_lattice)
 
@@ -179,7 +204,15 @@ def test_repr():
     tables = camelot.read_pdf(filename)
     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=120.48 y1=218.43 x2=164.64 y2=233.77>"
+
+
+def test_url():
+    url = "https://camelot-py.readthedocs.io/en/master/_static/pdf/foo.pdf"
+    tables = camelot.read_pdf(url)
+    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=218.43 x2=164.64 y2=233.77>"
 
 
 def test_arabic():