BUG: Fix precise_xstrtod segfault on long exponent #38789
Conversation
mzeitlin11
added
Regression
| # GH#38753 | ||
| parser = all_parsers | ||
| data = f"data\n10E{neg_exp}" | ||
| for precision in parser.float_precision_choices: |
There was a problem hiding this comment.
you can parameterize on these too
There was a problem hiding this comment.
Issue I ran into is that the parsers don't have consistent potential float_precisions, eg PythonParser only has None. Could parameterize and just assert failure on disallowed combinations of float_precision and parser, but that seems ugly. Will try to think of a better solution.
There was a problem hiding this comment.
Some kind of fixture-within-a-fixture functionality would be nice, though not sure if that's feasible...
There was a problem hiding this comment.
Added a new fixture, all_parsers_all_precisions. Not the most elegant solution, essentially just precomputes all allowed combos and uses that as the source for the fixture. If you have a better idea please let me know!
| # GH#38753 | ||
| parser = all_parsers | ||
| data = f"data\n10E{exp}" | ||
| for precision in parser.float_precision_choices: |
|
cc @gfyoung |
|
| if precision == "round_trip": | ||
| if exp == 999999999999999999 and is_platform_linux(): | ||
| mark = pytest.mark.xfail(reason="On Linux gives object result") | ||
| request.node.add_marker(mark) |
There was a problem hiding this comment.
This discrepancy is unrelated, occurs on master, have opened #38794 for it
There was a problem hiding this comment.
kk can you put the issue number in the xfail itself.
| if precision == "round_trip": | ||
| if exp == 999999999999999999 and is_platform_linux(): | ||
| mark = pytest.mark.xfail(reason="On Linux gives object result") | ||
| request.node.add_marker(mark) |
There was a problem hiding this comment.
kk can you put the issue number in the xfail itself.
|
thanks @mzeitlin11 very nice, keep em coming! |
|
@meeseeksdev backport 1.2.x |
#38828) Co-authored-by: mzeitlin11 <[email protected]>
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff@snowman2 can you check if this branch returns behavior to what you had before 1.2?