BUG: read_table and read_csv crash (#22748) #22750
Conversation
|
Hello @troels! Thanks for submitting the PR.
|
troels
force-pushed
the
dont-crash-on-badly-encoded-data
branch
from
b92bc82 to
6d5ecc7
Compare
Codecov Report
@@ Coverage Diff @@
## master #22750 +/- ##
==========================================
+ Coverage 92.18% 92.18% +<.01%
==========================================
Files 169 169
Lines 50812 50820 +8
==========================================
+ Hits 46842 46850 +8
Misses 3970 3970
Continue to review full report at Codecov.
|
troels
force-pushed
the
dont-crash-on-badly-encoded-data
branch
from
6d5ecc7 to
b7ddc09
Compare
| t = TextIOWrapper(b, encoding='ascii', errors='surrogateescape') | ||
| with pytest.raises(UnicodeEncodeError): | ||
| pd.read_csv(t) | ||
|
|
There was a problem hiding this comment.
What happens when you run this test against the Python engine?
There was a problem hiding this comment.
Hi @gfyoung
I've moved this to common.py now. I guess that even if the interpreter crash is only reproducible in the c-parser, the parsers should behave the same.
There was a problem hiding this comment.
Yes, that is correct. Rarely do we have tests that specific to a given parser, unless the parser does not support the behavior we are testing.
troels
force-pushed
the
dont-crash-on-badly-encoded-data
branch
from
b7ddc09 to
55e6e43
Compare
|
tiny change. ping on green. @gfyoung merge when pushed and green. |
troels
force-pushed
the
dont-crash-on-badly-encoded-data
branch
from
55e6e43 to
93fb9d5
Compare
A missing null-pointer check made read_table and read_csv prone to crash on badly encoded text. Add null-pointer check.
|
Thanks @troels ! |
A missing null-pointer check made read_table and read_csv prone to crash on badly encoded text. Add null-pointer check. Closes pandas-devgh-22748.
A missing null-pointer check made read_table and read_csv prone
to crash on badly encoded text. Add null-pointer check.
git diff upstream/master -u -- "*.py" | flake8 --diff