paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-04-02 13:45:10 -05:00
Code Issues Packages Projects Releases Wiki Activity

Don't consider better OCR as failing

Browse Source 
Tesseract 5.3.0 does a better job at OCR, and correctly
reads "a webp" instead of "awebp", this is good, so we
don't want the test to fail.
This commit is contained in:
Dennis Brakhane 2023-07-11 16:41:31 +02:00
parent db48d4c576
commit 93009c1eed
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/paperless_tesseract/tests/test_parser.py
View File

@ -861,8 +861,9 @@ class TestParserFileTypes(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
parser = RasterisedDocumentParser(None)
parser.parse(os.path.join(self.SAMPLE_FILES, "document.webp"), "image/webp")
self.assertIsFile(parser.archive_path)
# OCR consistent mangles this space, oh well
self.assertIn(
"this is awebp document, created 11/14/2022.",
# Older tesseracts consistently mangle the space between "a webp",
# tesseract 5.3.0 seems to do a better job, so we're accepting both
self.assertRegex(
parser.get_text().lower(),
r"this is a ?webp document, created 11/14/2022.",
)