Briq
/
camelot-py
6
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Use assert_frame_equal for more informative errors in tests

pull/48/head
Joel Nothman 2019-08-06 11:38:44 +10:00
parent 7ecfcad239
commit 9eb15c09dc
1 changed files with 22 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
tests/test_common.py
View File

@ -3,6 +3,7 @@
import os import os
import pandas as pd import pandas as pd
from pandas.testing import assert_frame_equal
import camelot import camelot
from camelot.core import Table, TableList from camelot.core import Table, TableList
@ -26,10 +27,10 @@ def test_password():
filename = os.path.join(testdir, "health_protected.pdf") filename = os.path.join(testdir, "health_protected.pdf")
tables = camelot.read_pdf(filename, password="ownerpass", 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") tables = camelot.read_pdf(filename, password="userpass", flavor="stream")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream(): def test_stream():
@ -37,7 +38,7 @@ def test_stream():
filename = os.path.join(testdir, "health.pdf") filename = os.path.join(testdir, "health.pdf")
tables = camelot.read_pdf(filename, flavor="stream") tables = camelot.read_pdf(filename, flavor="stream")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_table_rotated(): def test_stream_table_rotated():
@ -45,11 +46,11 @@ def test_stream_table_rotated():
filename = os.path.join(testdir, "clockwise_table_2.pdf") filename = os.path.join(testdir, "clockwise_table_2.pdf")
tables = camelot.read_pdf(filename, flavor="stream") tables = camelot.read_pdf(filename, flavor="stream")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
filename = os.path.join(testdir, "anticlockwise_table_2.pdf") filename = os.path.join(testdir, "anticlockwise_table_2.pdf")
tables = camelot.read_pdf(filename, flavor="stream") tables = camelot.read_pdf(filename, flavor="stream")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_two_tables(): def test_stream_two_tables():
@ -71,7 +72,7 @@ def test_stream_table_regions():
tables = camelot.read_pdf( tables = camelot.read_pdf(
filename, flavor="stream", table_regions=["320,460,573,335"] filename, flavor="stream", table_regions=["320,460,573,335"]
) )
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_table_areas(): def test_stream_table_areas():
@ -81,7 +82,7 @@ def test_stream_table_areas():
tables = camelot.read_pdf( tables = camelot.read_pdf(
filename, flavor="stream", table_areas=["320,500,573,335"] filename, flavor="stream", table_areas=["320,500,573,335"]
) )
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_columns(): def test_stream_columns():
@ -91,7 +92,7 @@ def test_stream_columns():
tables = camelot.read_pdf( tables = camelot.read_pdf(
filename, flavor="stream", columns=["67,180,230,425,475"], row_tol=10 filename, flavor="stream", columns=["67,180,230,425,475"], row_tol=10
) )
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_split_text(): def test_stream_split_text():
@ -104,7 +105,7 @@ def test_stream_split_text():
columns=["72,95,209,327,442,529,566,606,683"], columns=["72,95,209,327,442,529,566,606,683"],
split_text=True, split_text=True,
) )
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_flag_size(): def test_stream_flag_size():
@ -112,7 +113,7 @@ def test_stream_flag_size():
filename = os.path.join(testdir, "superscript.pdf") filename = os.path.join(testdir, "superscript.pdf")
tables = camelot.read_pdf(filename, flavor="stream", flag_size=True) tables = camelot.read_pdf(filename, flavor="stream", flag_size=True)
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_strip_text(): def test_stream_strip_text():
@ -120,7 +121,7 @@ def test_stream_strip_text():
filename = os.path.join(testdir, "detect_vertical_false.pdf") filename = os.path.join(testdir, "detect_vertical_false.pdf")
tables = camelot.read_pdf(filename, flavor="stream", strip_text=" ,\n") tables = camelot.read_pdf(filename, flavor="stream", strip_text=" ,\n")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_edge_tol(): def test_stream_edge_tol():
@ -128,7 +129,7 @@ def test_stream_edge_tol():
filename = os.path.join(testdir, "edge_tol.pdf") filename = os.path.join(testdir, "edge_tol.pdf")
tables = camelot.read_pdf(filename, flavor="stream", edge_tol=500) tables = camelot.read_pdf(filename, flavor="stream", edge_tol=500)
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_stream_layout_kwargs(): def test_stream_layout_kwargs():
@ -138,7 +139,7 @@ def test_stream_layout_kwargs():
tables = camelot.read_pdf( tables = camelot.read_pdf(
filename, flavor="stream", layout_kwargs={"detect_vertical": False} filename, flavor="stream", layout_kwargs={"detect_vertical": False}
) )
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice(): def test_lattice():
@ -148,7 +149,7 @@ def test_lattice():
testdir, "tabula/icdar2013-dataset/competition-dataset-us/us-030.pdf" testdir, "tabula/icdar2013-dataset/competition-dataset-us/us-030.pdf"
) )
tables = camelot.read_pdf(filename, pages="2") tables = camelot.read_pdf(filename, pages="2")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice_table_rotated(): def test_lattice_table_rotated():
@ -156,11 +157,11 @@ def test_lattice_table_rotated():
filename = os.path.join(testdir, "clockwise_table_1.pdf") filename = os.path.join(testdir, "clockwise_table_1.pdf")
tables = camelot.read_pdf(filename) tables = camelot.read_pdf(filename)
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
filename = os.path.join(testdir, "anticlockwise_table_1.pdf") filename = os.path.join(testdir, "anticlockwise_table_1.pdf")
tables = camelot.read_pdf(filename) tables = camelot.read_pdf(filename)
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice_two_tables(): def test_lattice_two_tables():
@ -179,7 +180,7 @@ def test_lattice_table_regions():
filename = os.path.join(testdir, "table_region.pdf") filename = os.path.join(testdir, "table_region.pdf")
tables = camelot.read_pdf(filename, table_regions=["170,370,560,270"]) tables = camelot.read_pdf(filename, table_regions=["170,370,560,270"])
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice_table_areas(): def test_lattice_table_areas():
@ -187,7 +188,7 @@ def test_lattice_table_areas():
filename = os.path.join(testdir, "twotables_2.pdf") filename = os.path.join(testdir, "twotables_2.pdf")
tables = camelot.read_pdf(filename, table_areas=["80,693,535,448"]) tables = camelot.read_pdf(filename, table_areas=["80,693,535,448"])
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice_process_background(): def test_lattice_process_background():
@ -195,7 +196,7 @@ def test_lattice_process_background():
filename = os.path.join(testdir, "background_lines_1.pdf") filename = os.path.join(testdir, "background_lines_1.pdf")
tables = camelot.read_pdf(filename, process_background=True) tables = camelot.read_pdf(filename, process_background=True)
assert df.equals(tables[1].df) assert_frame_equal(df, tables[1].df)
def test_lattice_copy_text(): def test_lattice_copy_text():
@ -203,7 +204,7 @@ def test_lattice_copy_text():
filename = os.path.join(testdir, "row_span_1.pdf") filename = os.path.join(testdir, "row_span_1.pdf")
tables = camelot.read_pdf(filename, line_scale=60, copy_text="v") tables = camelot.read_pdf(filename, line_scale=60, copy_text="v")
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_lattice_shift_text(): def test_lattice_shift_text():
@ -271,7 +272,7 @@ def test_arabic():
filename = os.path.join(testdir, "tabula/arabic.pdf") filename = os.path.join(testdir, "tabula/arabic.pdf")
tables = camelot.read_pdf(filename) tables = camelot.read_pdf(filename)
assert df.equals(tables[0].df) assert_frame_equal(df, tables[0].df)
def test_table_order(): def test_table_order():