yujunjun/unstructured
1
0
Fork 0
mirror of https://github.com/Unstructured-IO/unstructured.git synced 2025-07-03 23:20:35 +00:00
Code Issues Packages Projects Releases Wiki Activity
unstructured/test_unstructured/documents/test_html.py
Amanda Cameron a501d1d18f
Adding table extraction to partition_html (#1324) 
Adding table extraction to HTML partitioning.

This PR utilizes 'table' HTML elements to extract and parse HTML tables
and return them in partitioning.

```
# checkout this branch, go into ipython shell
In [1]: from unstructured.partition.html import partition_html
In [2]: path_to_html = "{html sample file with table}"
In [3]: elements = partition_html(path_to_html)
```
you should see the table in the elements list!
2023-09-11 11:14:11 -07:00

688 lines
20 KiB
Python
Raw Blame History

import os
import pathlib
import pytest
from lxml import etree
from unstructured.documents import html
from unstructured.documents.base import Page
from unstructured.documents.elements import (
Address,
ListItem,
NarrativeText,
Table,
Text,
Title,
)
from unstructured.documents.html import (
HEADING_TAGS,
LIST_ITEM_TAGS,
TABLE_TAGS,
TEXT_TAGS,
HTMLDocument,
HTMLNarrativeText,
HTMLTitle,
TagsMixin,
)
DIRECTORY = pathlib.Path(__file__).parent.resolve()
TAGS = (
"<a><abbr><acronym><address><applet><area><article><aside><audio><b><base><basefont><bdi>"
"<bdo><big><blockquote><body><br><button><canvas><caption><center><cite><code><col>"
"<colgroup><data><datalist><dd><del><details><dfn><dialog><dir><div><dl><dt><em><embed>"
"<fieldset><figcaption><figure><font><footer><form><frame><frameset><h1><h2><h3><h4><h5><h6>"
"<head><header><hr><html><i><iframe><img><input><ins><kbd><label><legend><li><link><main>"
"<map><mark><meta><meter><nav><noframes><noscript><object><ol><optgroup><option><output><p>"
"<param><picture><pre><progress><q><rp><rt><ruby><s><samp><script><section><select><small>"
"<source><span><strike><strong><style><sub><summary><sup><table><tbody><td><template>"
"<textarea><tfoot><th><thead><time><title><tr><track><tt><u><ul><var><video><wbr>"
)
TAGS = TAGS.replace(">", "").split("<")[1:]
INCLUDED_TAGS = TEXT_TAGS + HEADING_TAGS + LIST_ITEM_TAGS + ["div"]
EXCLUDED_TAGS = "tag", [tag for tag in TAGS if tag not in INCLUDED_TAGS]
@pytest.fixture()
def sample_doc():
table_element = HTMLTitle(
"I'm a title in a table.",
tag="p",
ancestortags=("table", "tbody", "tr", "td"),
)
narrative = HTMLNarrativeText("I'm some narrative text", tag="p", ancestortags=())
page1 = Page(0)
page1.elements = [table_element, narrative]
header = HTMLTitle("I'm a header", tag="header", ancestortags=())
body = HTMLNarrativeText("Body text", tag="p", ancestortags=())
footer = HTMLTitle("I'm a footer", tag="footer", ancestortags=())
page2 = Page(1)
page2.elements = [header, body, footer]
doc = HTMLDocument.from_pages([page1, page2])
return doc
def test_parses_tags_correctly():
raw_html = """<html>
<body>
<table>
<tbody>
<tr>
<td><p>Hi there!</p></td>
</tr>
</tbody>
</table>
</body>
</html>"""
doc = HTMLDocument.from_string(raw_html)
el = doc.elements[0]
assert el.ancestortags + (el.tag,) == ("html", "body", "table")
def test_has_table_ancestor():
title = HTMLTitle(
"I am a Title",
tag="td",
ancestortags=["html", "body", "table", "tr"],
)
assert html.has_table_ancestor(title)
def test_has_no_table_ancestor():
title = HTMLTitle(
"I am a Title",
tag="p",
ancestortags=["html", "body"],
)
assert not html.has_table_ancestor(title)
def test_read_without_skipping_table(monkeypatch):
monkeypatch.setattr(html, "is_possible_narrative_text", lambda *args: True)
doc = """<html>
<body>
<table>
<tbody>
<tr>
<td><p>Hi there! I am Matt!</p></td>
</tr>
</tbody>
</table>
</body>
</html>"""
document = HTMLDocument.from_string(doc).doc_after_cleaners(skip_table=False)
assert document.pages[0].elements[0] == Table(text="Hi there! I am Matt!")
@pytest.mark.parametrize(
("doc", "expected"),
[
(
"<p>Hi there <span>my name is</span> <b><i>Matt</i></i></p>",
"Hi there my name is Matt",
),
("<p>I have a</p> tail", "I have a tail"),
],
)
def test_construct_text(doc, expected):
document_tree = etree.fromstring(doc, etree.HTMLParser())
para = document_tree.find(".//p")
text = html._construct_text(para)
assert text == expected
@pytest.mark.parametrize(
("doc", "root", "expected"),
[
(
"<p>Hello <strong>there</strong> I <em>am</em> a <b>very</b> <i>important</i> text</p>",
"p",
[
{"text": "there", "tag": "strong"},
{"text": "am", "tag": "em"},
{"text": "very", "tag": "b"},
{"text": "important", "tag": "i"},
],
),
(
"<p>Here is a <span>list</span> of <b>my <i>favorite</i> things</b></p>",
"p",
[
{"text": "list", "tag": "span"},
{"text": "my favorite things", "tag": "b"},
{"text": "favorite", "tag": "i"},
],
),
(
"<strong>A lone strong text!</strong>",
"strong",
[{"text": "A lone strong text!", "tag": "strong"}],
),
("<span>I have a</span> tail", "span", [{"text": "I have a", "tag": "span"}]),
("<span>Empty result</span> ", "p", []),
],
)
def test_get_emphasized_texts_from_tag(doc, expected, root):
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(f".//{root}")
emphasized_texts = html._get_emphasized_texts_from_tag(el)
assert emphasized_texts == expected
def test_parse_nothing():
doc = """<p></p>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//p")
parsed_el = html._parse_tag(el)
assert parsed_el is None
def test_read_with_existing_pages():
page = Page(number=0)
html_document = HTMLDocument.from_pages([page])
html_document._read()
assert html_document.pages == [page]
def test_parse_not_anything(monkeypatch):
monkeypatch.setattr(html, "is_narrative_tag", lambda *args: False)
monkeypatch.setattr(html, "is_possible_title", lambda *args: False)
doc = """<p>This is nothing</p>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//p")
parsed_el = html._parse_tag(el)
assert parsed_el == Text(text="This is nothing")
def test_parse_bullets(monkeypatch):
doc = """<p>● An excellent point!</p>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//p")
parsed_el = html._parse_tag(el)
assert parsed_el == ListItem("An excellent point!")
def test_parse_tag_ignores_lonely_bullets():
doc = """<p>●</p>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//p")
parsed_el = html._parse_tag(el)
assert parsed_el is None
def test_parse_tag_ignores_stubs():
doc = """<p>$</p>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//p")
parsed_el = html._parse_tag(el)
assert parsed_el is None
def test_adjacent_spans_are_text_tags():
doc = """<div><span>&#8226;</span><span>A bullet!</span></div>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//div")
assert html._is_text_tag(el) is True
def test_process_list_item_gets_next_section():
doc = """
<div>
<p>●</p>
<p>●</p>
</div>
<div>
<p>An excellent point!</p>
</div>
"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//div")
parsed_el, _ = html._process_list_item(el, max_predecessor_len=10)
assert parsed_el == ListItem(text="An excellent point!")
def test_get_bullet_descendants():
div_1 = "<div><p>●</p><p>●</p></div>"
document_tree_1 = etree.fromstring(div_1, etree.HTMLParser())
element = document_tree_1.find(".//div")
div_2 = "<div><p>An excellent point!</p></div>"
document_tree_2 = etree.fromstring(div_2, etree.HTMLParser())
next_element = document_tree_2.find(".//div")
descendants = html._get_bullet_descendants(element, next_element)
assert len(descendants) == 1
def test_process_list_item_returns_none_if_next_blank():
doc = """
<div>
<p>●</p>
<p>●</p>
</div>
"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//div")
parsed_el, _ = html._process_list_item(el)
assert parsed_el is None
def test_process_list_item_returns_none_if_next_has_no_text():
doc = """
<div>
<p>●</p>
<p>●</p>
</div>
<div>
</div>
"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//div")
assert html.is_list_item_tag(el) is True
parsed_el, _ = html._process_list_item(el)
assert parsed_el is None
def test_process_list_item_ignores_deep_divs():
doc = """
<div>
<p>●</p>
<p>●</p>
<p>●</p>
<p>●</p>
<p>●</p>
</div>
<div>
<p>An excellent point!</p>
</div>
"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
el = document_tree.find(".//div")
parsed_el, _ = html._process_list_item(el, max_predecessor_len=2)
assert parsed_el is None
def test_read_html_doc(tmpdir, monkeypatch):
TITLE1 = "A Great and Glorious Section"
SECTION1 = "Dear Leader is the best. He is such a wonderful engineer!"
TITLE2 = "Another Magnificent Title"
SECTION2 = "The last section is a title because of its capitalization patterns!"
TABLE_SECTION = "Skip me because I'm in a table"
TITLE3 = "A New Beginning"
SECTION3 = "Here is the start of a new page."
doc = f"""<html>
<body>
<header>
<p>Here is a header. We want to ignore anything that is in this section.</p>
</header>
<h1>{TITLE1}</h1>
<p>{SECTION1}</p>
<p></p>
<p>{TITLE2}</p>
<p><b>{SECTION2}</b></p>
<table>
<tbody>
<tr>
<td><p>{TABLE_SECTION}</p></td>
</tr>
</tbody>
</table>
<hr>
<h2>{TITLE3}</h2>
<div>{SECTION3}</div>
<footer>
<p>Here is a footer. We want to ignore anything that is in this section</p>
</footer>
<div>
<p>Let's ignore anything after the footer too since it's probably garbage.</p>
</div>
</body>
</html>"""
filename = os.path.join(tmpdir.dirname, "sample-doc.html")
with open(filename, "w") as f:
f.write(doc)
html_document = HTMLDocument.from_file(filename=filename).doc_after_cleaners(
skip_headers_and_footers=True,
skip_table=True,
)
print("original pages: ", HTMLDocument.from_file(filename=filename).pages)
print("filtered pages: ", html_document.pages)
print([el.text for el in html_document.pages[0].elements])
assert len(html_document.pages) == 2
page_one = html_document.pages[0]
assert len(page_one.elements) == 4
assert page_one.elements == [
Title(text=TITLE1),
NarrativeText(text=SECTION1),
Title(text=TITLE2),
NarrativeText(text=SECTION2),
]
page_two = html_document.pages[1]
assert len(page_two.elements) == 2
assert page_two.elements == [
Title(text=TITLE3),
NarrativeText(text=SECTION3),
]
pages = html_document.pages
assert all(isinstance(page, Page) for page in pages)
def test_find_main():
html_str = """<header></header>
<body>
<p>Lots preamble stuff yada yada yada</p>
<main>
<article>
<section>
<h2>A Wonderful Section!</h2>
<p>Look at this amazing section!</p>
</section>
<section>
<h2>Another Wonderful Section!</h2>
<p>Look at this other amazing section!</p>
</section>
</article>
</main>
</body>"""
html_document = HTMLDocument.from_string(html_str)
document_tree = html_document.document_tree
main_tag = html._find_main(document_tree)
assert main_tag.tag == "main"
def test_find_main_returns_doc_when_main_not_present():
html_str = """<header></header>
<body>
<p>Lots preamble stuff yada yada yada</p>
<article>
<section>
<h2>A Wonderful Section!</h2>
<p>Look at this amazing section!</p>
</section>
<section>
<h2>Another Wonderful Section!</h2>
<p>Look at this other amazing section!</p>
</section>
</article>
</body>"""
html_document = HTMLDocument.from_string(html_str)
document_tree = html_document.document_tree
root = html._find_main(document_tree)
assert root.tag == "html"
def test_find_articles():
html_str = """<header></header>
<body>
<p>Lots preamble stuff yada yada yada</p>
<article>
<h2>A Wonderful Section!</h2>
<p>Look at this amazing section!</p>
</article>
<article>
<h2>Another Wonderful Section!</h2>
<p>Look at this other amazing section!</p>
</article>
</body>"""
html_document = HTMLDocument.from_string(html_str)
document_tree = html_document.document_tree
articles = html._find_articles(document_tree)
assert len(articles) == 2
def test_find_articles_returns_doc_when_none_present():
html_str = """<header></header>
<body>
<p>Lots preamble stuff yada yada yada</p>
<section>
<h2>A Wonderful Section!</h2>
<p>Look at this amazing section!</p>
</section>
<section>
<h2>Another Wonderful Section!</h2>
<p>Look at this other amazing section!</p>
</section>
</body>"""
html_document = HTMLDocument.from_string(html_str)
document_tree = html_document.document_tree
articles = html._find_articles(document_tree)
assert len(articles) == 1
def test_include_headers_and_footers(sample_doc):
html_document = sample_doc.doc_after_cleaners(skip_headers_and_footers=False)
assert len(html_document.pages[1].elements) == 3
def test_include_table_text(sample_doc):
html_document = sample_doc.doc_after_cleaners(skip_table=False)
assert len(html_document.pages[0].elements) == 2
@pytest.mark.parametrize("tag", [tag for tag in TEXT_TAGS if tag not in TABLE_TAGS])
def test_tag_types(tag):
html_str = f"""
<body>
<{tag}>
There is some text here.
</{tag}>
</body>
"""
html_document = HTMLDocument.from_string(html_str)
assert len(html_document.pages[0].elements) == 1
@pytest.mark.parametrize("tag", EXCLUDED_TAGS)
def test_exclude_tag_types(tag):
html_str = f"""
<body>
<{tag}>
There is some text here.
</{tag}>
</body>
"""
html_document = HTMLDocument.from_string(html_str)
assert len(html_document.pages) == 0
def test_tag_types_table(sample_doc):
html_document = sample_doc.doc_after_cleaners(skip_table=True)
assert len(html_document.pages[0].elements) == 2
def test_nested_text_tags():
tag1, tag2 = [tag for tag in TEXT_TAGS if tag not in TABLE_TAGS][:2]
html_str = f"""
<body>
<{tag1}>
<{tag2}>
There is some text here.
</{tag2}>
</{tag1}>
</body>
"""
html_document = HTMLDocument.from_string(html_str).doc_after_cleaners(skip_table=False)
assert len(html_document.pages[0].elements) == 1
def test_containers_with_text_are_processed():
html_str = """<div dir=3D"ltr">Hi All,<div><br></div>
<div>Get excited for our first annual family day!</div>
<div>Best.<br clear=3D"all">
<div><br></div>
-- <br>
<div dir=3D"ltr">
<div dir=3D"ltr">Dino the Datasaur<div>Unstructured Technologies<br><div>Data Scientist
</div>
<div>Doylestown, PA 18901</div>
<div><br></div>
</div>
</div>
</div>
</div>
</div>"""
html_document = HTMLDocument.from_string(html_str)
html_document._read()
assert html_document.elements == [
Text(text="Hi All,"),
NarrativeText(text="Get excited for our first annual family day!"),
Title(text="Best."),
Title(text="Dino the Datasaur"),
Title(text="Unstructured Technologies"),
Title(text="Data Scientist"),
Address(text="Doylestown, PA 18901"),
]
def test_html_grabs_bulleted_text_in_tags():
html_str = """<html>
<body>
<ol>
<li>Happy Groundhog's day!</li>
<li>Looks like six more weeks of winter ...</li>
</ol>
</body>
</html>"""
html_document = HTMLDocument.from_string(html_str)
html_document._read()
assert html_document.elements == [
ListItem(text="Happy Groundhog's day!"),
ListItem(text="Looks like six more weeks of winter ..."),
]
def test_html_grabs_bulleted_text_in_paras():
html_str = """<html>
<body>
<p>
<span>&#8226; Happy Groundhog's day!</span>
</p>
<p>
<span>&#8226; Looks like six more weeks of winter ...</span>
</p>
</body>
</html>"""
html_document = HTMLDocument.from_string(html_str)
html_document._read()
assert html_document.elements == [
ListItem(text="Happy Groundhog's day!"),
ListItem(text="Looks like six more weeks of winter ..."),
]
def test_bulletized_bulleted_text_from_table():
doc = """<html>
<body>
<table>
<tbody>
<tr>
<td>•</td>
<td><p>Happy Groundhog's day!</p></td>
</tr>
<tr>
<td>•</td>
<td><p>Looks like six more weeks of winter ...</p></td>
</tr>
</tbody>
</table>
</body>
</html>"""
document_tree = etree.fromstring(doc, etree.HTMLParser())
table = document_tree.find(".//table")
bulleted_text = html._bulleted_text_from_table(table)
assert bulleted_text == [
ListItem(text="Happy Groundhog's day!"),
ListItem(text="Looks like six more weeks of winter ..."),
]
def test_html_grabs_bulleted_text_in_tables():
html_str = """<html>
<body>
<table>
<tbody>
<tr>
<td>&#8226;</td>
<td><p>Happy Groundhog's day!</p></td>
</tr>
<tr>
<td>&#8226;</td>
<td><p>Looks like six more weeks of winter ...</p></td>
</tr>
</tbody>
</table>
</body>
</html>"""
html_document = HTMLDocument.from_string(html_str)
html_document._read()
assert html_document.elements == [
ListItem(text="Happy Groundhog's day!"),
ListItem(text="Looks like six more weeks of winter ..."),
]
def test_raises_error_no_tag():
with pytest.raises(TypeError):
TagsMixin(tag=None)
with pytest.raises(TypeError):
TagsMixin()
def test_raises_error_wrong_elements(monkeypatch, sample_doc):
page = Page(0)
page.elements = ["this should def not be a string"]
monkeypatch.setattr(sample_doc, "_pages", [page])
with pytest.raises(ValueError):
sample_doc.doc_after_cleaners()
def test_filter_in_place():
html_doc = """
<table><tbody><tr><td>A table thing.</td></tr></tbody></table>
<p>A non-table thing</p>
"""
doc = HTMLDocument.from_string(html_doc)
assert len(doc.elements) == 2
doc.doc_after_cleaners(skip_table=True, inplace=True)
assert len(doc.elements) == 1
def test_joins_tag_text_correctly():
raw_html = "<p>Hello again peet mag<i>ic</i>al</p>"
doc = HTMLDocument.from_string(raw_html)
el = doc.elements[0]
assert el.text == "Hello again peet magical"
def test_sample_doc_with_scripts():
filename = os.path.join(DIRECTORY, "..", "..", "example-docs", "example-with-scripts.html")
doc = HTMLDocument.from_file(filename=filename)
assert all("function (" not in element.text for element in doc.elements)
def test_sample_doc_with_emoji():
raw_html = """
<html charset="unicode">
<p>Hello again 😀</p>
</html>
"""
doc = HTMLDocument.from_string(raw_html)
# NOTE(robinson) - unclear why right now, but the output is the emoji on the test runners
# and the byte string representation when running locally on mac
assert doc.elements[0].text in ["Hello again ð\x9f\x98\x80", "Hello again 😀"]
Reference in New Issue View Git Blame Copy Permalink