mirror of
https://github.com/docling-project/docling-core.git
synced 2026-11-18 20:27:00 +00:00
test: rename 'test' folder to 'tests' To align with other Docling repositories, use 'tests' for test modules. Signed-off-by: Cesar Berrospi Ramis <ceb@zurich.ibm.com>
180 lines
5.7 KiB
Python
180 lines
5.7 KiB
Python
"""Test LaTeX serialization.
|
|
|
|
These tests leverage DOCLING_GEN_TEST_DATA via GEN_TEST_DATA to optionally
|
|
generate or update the golden files. Additionally, if a golden file does not
|
|
exist, the test will create it to bootstrap the baseline.
|
|
"""
|
|
|
|
from pathlib import Path
|
|
|
|
import yaml
|
|
|
|
from docling_core.transforms.serializer.latex import LaTeXDocSerializer, LaTeXParams
|
|
from docling_core.types.doc.base import ImageRefMode
|
|
from docling_core.types.doc.document import DoclingDocument, Formatting
|
|
|
|
from .test_data_gen_flag import GEN_TEST_DATA
|
|
|
|
|
|
def verify_or_update(exp_file: Path, actual: str):
|
|
exp_file.parent.mkdir(parents=True, exist_ok=True)
|
|
# If GEN_TEST_DATA is enabled or the expected file is missing, write/update it
|
|
if GEN_TEST_DATA or not exp_file.exists():
|
|
with open(exp_file, "w", encoding="utf-8") as f:
|
|
f.write(f"{actual}\n")
|
|
else:
|
|
with open(exp_file, encoding="utf-8") as f:
|
|
expected = f.read().rstrip()
|
|
assert expected == actual
|
|
|
|
|
|
def test_latex_basic_activities():
|
|
src = Path("./tests/data/doc/activities.json")
|
|
doc = DoclingDocument.load_from_json(src)
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
# Do not add page-break replacements by default
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_latex_inline_and_formatting():
|
|
src = Path("./tests/data/doc/inline_and_formatting.yaml")
|
|
doc = DoclingDocument.load_from_yaml(src)
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_dummy_doc():
|
|
src = Path("tests/data/doc/dummy_doc.yaml")
|
|
|
|
# Read YAML file of manual reference doc
|
|
with open(src, encoding="utf-8") as fp:
|
|
dict_from_yaml = yaml.safe_load(fp)
|
|
|
|
doc = DoclingDocument.model_validate(dict_from_yaml)
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_constructed_doc(sample_doc: DoclingDocument):
|
|
doc = sample_doc
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
src = Path("tests/data/doc/construct_doc.yaml")
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_constructed_rich_table_doc(rich_table_doc: DoclingDocument):
|
|
doc = rich_table_doc
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
src = Path("tests/data/doc/construct_rich_table_doc.yaml")
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_latex_paper():
|
|
src = Path("./tests/data/doc/2408.09869v3_enriched.json")
|
|
doc = DoclingDocument.load_from_json(src)
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_latex_nested_lists():
|
|
src = Path("./tests/data/doc/polymers.json")
|
|
doc = DoclingDocument.load_from_json(src)
|
|
|
|
ser = LaTeXDocSerializer(
|
|
doc=doc,
|
|
params=LaTeXParams(
|
|
image_mode=ImageRefMode.PLACEHOLDER,
|
|
page_break_command=None,
|
|
),
|
|
)
|
|
actual = ser.serialize().text
|
|
verify_or_update(exp_file=src.with_suffix(".gt.tex"), actual=actual)
|
|
|
|
|
|
def test_inline_group_no_duplication():
|
|
r"""Regression test: inline-group content must not appear twice.
|
|
|
|
When a SectionHeaderItem, TitleItem or ListItem delegates its text to a
|
|
child InlineGroup (the shape produced by Markdown/DOCX/HTML import for any
|
|
formatted heading or list item), the serialized text must appear exactly
|
|
once — inside the LaTeX command (\section{}, \item, …) and *not* again
|
|
as a standalone paragraph immediately after it.
|
|
"""
|
|
doc = DoclingDocument(name="t")
|
|
|
|
# Formatted section heading via InlineGroup
|
|
heading = doc.add_heading(text="", level=1)
|
|
ig = doc.add_inline_group(parent=heading)
|
|
doc.add_text(
|
|
label="text",
|
|
text="Partially formatted",
|
|
formatting=Formatting(italic=True),
|
|
parent=ig,
|
|
)
|
|
doc.add_text(label="text", text="heading", parent=ig)
|
|
|
|
# Formatted list item via InlineGroup
|
|
lg = doc.add_list_group()
|
|
li = doc.add_list_item(text="", parent=lg)
|
|
ig2 = doc.add_inline_group(parent=li)
|
|
doc.add_text(label="text", text="Term", formatting=Formatting(bold=True), parent=ig2)
|
|
doc.add_text(label="text", text=": definition", parent=ig2)
|
|
|
|
body = LaTeXDocSerializer(doc=doc).serialize().text.split("\\begin{document}")[1]
|
|
|
|
# Each piece of content must appear exactly once in the body
|
|
assert body.count("\\textit{Partially formatted} heading") == 1
|
|
assert body.count("\\textbf{Term} : definition") == 1
|
|
|
|
# The content must be inside the LaTeX command, not on a standalone line
|
|
assert "\\section{\\textit{Partially formatted} heading}" in body
|
|
assert "\\item \\textbf{Term} : definition" in body
|