paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-10-30 03:56:23 -05:00
Code Issues Packages Projects Releases Wiki Activity

tests

Browse Source
This commit is contained in:
jonaswinkler
2021-02-22 00:17:16 +01:00
parent 30b0a30146
commit 99a18516b2
1 changed files with 41 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

41
src/paperless_tesseract/tests/test_parser.py
View File

@@ -300,6 +300,47 @@ class TestParser(DirectoriesMixin, TestCase):
"If you read this, it’s your own fault. Also check your screen orientation."
])
def test_ocrmypdf_parameters(self):
parser = RasterisedDocumentParser(None)
params = parser.construct_ocrmypdf_parameters(input_file="input.pdf", output_file="output.pdf",
sidecar_file="sidecar.txt", mime_type="application/pdf",
safe_fallback=False)
self.assertEqual(params['input_file'], "input.pdf")
self.assertEqual(params['output_file'], "output.pdf")
self.assertEqual(params['sidecar'], "sidecar.txt")
with override_settings(OCR_CLEAN="none"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertNotIn("clean", params)
self.assertNotIn("clean_final", params)
with override_settings(OCR_CLEAN="clean"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertTrue(params['clean'])
self.assertNotIn("clean_final", params)
with override_settings(OCR_CLEAN="clean-final", OCR_MODE="skip"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertTrue(params['clean_final'])
self.assertNotIn("clean", params)
with override_settings(OCR_CLEAN="clean-final", OCR_MODE="redo"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertTrue(params['clean'])
self.assertNotIn("clean_final", params)
with override_settings(OCR_DESKEW=True, OCR_MODE="skip"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertTrue(params['deskew'])
with override_settings(OCR_DESKEW=True, OCR_MODE="redo"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertNotIn('deskew', params)
with override_settings(OCR_DESKEW=False, OCR_MODE="skip"):
params = parser.construct_ocrmypdf_parameters("", "", "", "")
self.assertNotIn('deskew', params)
class TestParserFileTypes(DirectoriesMixin, TestCase):