test: avoid testing exact JSON in CSV backend (#1038)

* feat: updated verify_export
Moved verify_export to verify_utils
Reuse verify_export in tests

Signed-off-by: Matheus Abdias <matheusfabdias@gmail.com>

* feat: replace verify_export with verify_document in CSV conversion tests

Signed-off-by: Matheus Abdias <matheusfabdias@gmail.com>

---------

Signed-off-by: Matheus Abdias <matheusfabdias@gmail.com>
This commit is contained in:
Suehtam
2025-02-24 07:10:40 +00:00
committed by GitHub
parent d8a81c3168
commit 1d17e7397a
7 changed files with 22 additions and 98 deletions

View File

@@ -11,7 +11,7 @@ from docling.datamodel.document import (
)
from docling.document_converter import DocumentConverter
from .verify_utils import verify_document
from .verify_utils import verify_document, verify_export
GENERATE = False
@@ -58,22 +58,6 @@ def get_converter():
return converter
def verify_export(pred_text: str, gtfile: str):
if not os.path.exists(gtfile) or GENERATE:
with open(gtfile, "w") as fw:
fw.write(pred_text)
return True
else:
with open(gtfile) as fr:
true_text = fr.read()
assert pred_text == true_text, f"pred_text!=true_text for {gtfile}"
return pred_text == true_text
def test_e2e_html_conversions():
html_paths = get_html_paths()