yujunjun/unstructured
1
0
Fork 0
mirror of https://github.com/Unstructured-IO/unstructured.git synced 2025-10-29 17:04:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(docx): tabulate output is non-deterministic (#2090)

Browse Source 
The test for nested tables added a few PRs ago indirectly relies on the
padding added to table-HTML by `tabulate`. The length of that padding
turns out to be non-deterministic, perhaps related to M1 vs. Intel
hardware.

Remove padding from tabulate output in the test so only actual content
is compared.
This commit is contained in:
Steve Canny 2023-11-15 23:52:16 -08:00 committed by GitHub
parent 5fa40850f4
commit 41fc55bc12
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
test_unstructured/partition/docx/test_docx.py
View File

@ -1,6 +1,7 @@
# pyright: reportPrivateUsage=false # pyright: reportPrivateUsage=false
import pathlib import pathlib
import re
from tempfile import SpooledTemporaryFile from tempfile import SpooledTemporaryFile
from typing import Dict, List, cast from typing import Dict, List, cast
@ -63,27 +64,28 @@ class Describe_DocxPartitioner:
""" """
table = docx.Document(example_doc_path("docx-tables.docx")).tables[1] table = docx.Document(example_doc_path("docx-tables.docx")).tables[1]
html = _DocxPartitioner()._convert_table_to_html(table) # -- re.sub() strips out the extra padding inserted by tabulate --
html = re.sub(r" +<", "<", _DocxPartitioner()._convert_table_to_html(table))
expected_lines = [ expected_lines = [
"<table>", "<table>",
"<thead>", "<thead>",
f"<tr><th>a </th><th>&gt;b&lt;{' ' * 96}</th><th>c </th></tr>", "<tr><th>a</th><th>&gt;b&lt;</th><th>c</th></tr>",
"</thead>", "</thead>",
"<tbody>", "<tbody>",
"<tr><td>d </td><td><table>", "<tr><td>d</td><td><table>",
"<tbody>", "<tbody>",
"<tr><td>e </td><td>f</td></tr>", "<tr><td>e</td><td>f</td></tr>",
"<tr><td>g&amp;t</td><td>h</td></tr>", "<tr><td>g&amp;t</td><td>h</td></tr>",
"</tbody>", "</tbody>",
"</table></td><td>i </td></tr>", "</table></td><td>i</td></tr>",
f"<tr><td>j </td><td>k{' ' * 104}</td><td>l </td></tr>", "<tr><td>j</td><td>k</td><td>l</td></tr>",
"</tbody>", "</tbody>",
"</table>", "</table>",
] ]
actual_lines = html.splitlines() actual_lines = html.splitlines()
for expected, actual in zip(expected_lines, actual_lines): for expected, actual in zip(expected_lines, actual_lines):
assert actual == expected assert actual == expected, f"\nexpected: {repr(expected)}\nactual: {repr(actual)}"
def it_can_convert_a_table_to_plain_text(self): def it_can_convert_a_table_to_plain_text(self):
table = docx.Document(example_doc_path("docx-tables.docx")).tables[0] table = docx.Document(example_doc_path("docx-tables.docx")).tables[0]