luciano/markitdown
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Linked to Magika issue.

Browse source
This commit is contained in:
Adam Fourney 2025-03-12 03:51:23 -07:00
parent 677ce9132b
commit 8938eb84dc
1 changed files with 3 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
packages/markitdown/tests/test_module_vectors.py
View file

@ -47,7 +47,7 @@ def test_guess_stream_info(test_vector):
# mimetype or extension, so we'll special-case them here.
if test_vector.filename in [
"test_outlook_msg.msg",
"test_mskanji.csv", # This works locally but not on the CI. Maybe the charset is different?
"test_mskanji.csv", # See: https://github.com/google/magika/issues/983
]:
return
@ -99,10 +99,8 @@ def test_convert_stream_without_hints(test_vector):
# For some limited exceptions, we can't guarantee the exact
# mimetype or extension, so we'll special-case them here.
if test_vector.filename in [
# This appears to be a subtle bug in magika where, in some
# environments (like CI and Docker), this file is misclassified
# as random bytes, only when read from stdin. (MD5 hashses
# confirm byte-for-byte equality)
# This appears to be a subtle bug in magika.
# See: https://github.com/google/magika/issues/983
"test_mskanji.csv",
]:
return