[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Change by Josh Rosenberg : -- nosy: +josh.r ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Armin Rigo added the comment: An update to Serhiy's proposed fix: #if PY_VERSION_HEX < 0x0307 && defined(PySlice_GetIndicesEx) #if !defined(PYPY_VERSION) #undef PySlice_GetIndicesEx #endif #endif All PyXxx functions are macros on PyPy, and undefining a macro

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: Looks good to me, thanks Serhiy. -- ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Ned Deily added the comment: As a followup, Nathanial, are you satisfied with the resolution here for the upcoming 3.6.2? -- ___ Python tracker ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Dino Viehland : -- pull_requests: +2082 ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Serhiy Storchaka : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: New changeset f43b293f2fee91578e28c7aa566510a0cd6e33cb by Serhiy Storchaka in branch '3.6': [3.6] bpo-29943: Do not replace the function PySlice_GetIndicesEx() with a macro (GH-1049) (#1813)

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Serhiy Storchaka : -- pull_requests: +1899 ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: I'll apply to 3.6 the same patch as for 3.5 tomorrow. -- assignee: -> serhiy.storchaka ___ Python tracker ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Ned Deily added the comment: Thanks everyone for the input on this issue. We've had some discussions here at PyCon US and I think the consensus is that we all agree with Nathaniel's comment above that, for 3.6.x, that modules built on 3.6.2 (and later 3.6.x) should be importable on any 3.6.x

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: @Jonathan: Even 3.6.1 was careful to retain compatibility with code built by 3.6.0. And your proposed 3.6.1-patched will generate binaries equivalent to the ones 3.6.0 generates. So I don't think you need to worry; 3.6.2 is not going to add a new and worse

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Jonathan Helmus added the comment: In Anaconda we ship both Python 3.6.0 and 3.6.1 and have run into this issue when building packages with 3.6.1 and importing them on 3.6.0. We are discussing adding #undef PySlice_GetIndicesEx to the Python.h we ship with Python 3.6.1. From the discussion

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Mark Dickinson : -- nosy: +mark.dickinson ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Miro Hrončok added the comment: > For what it's worth, in Fedora 26 we already rebased Python to 3.6.1, so this > issue now is non existent for our ecosystem, and we are not shipping 3.6.0 in > any way now. Except of course if 3.6.2 would be fixed in a way that 3.6.1 would be considered

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Charalampos Stratakis added the comment: For what it's worth, in Fedora 26 we already rebased Python to 3.6.1, so this issue now is non existent for our ecosystem, and we are not shipping 3.6.0 in any way now. -- ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: I don't find it helpful to think of it as declaring 3.6.0 broken vs declaring 3.6.1 broken. 3.6.0 is definitely good in the sense that if you build a module on it then it will import on both 3.6.0 and 3.6.1, and 3.6.1 is definitely good in the sense that if

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: Either 3.6.1 or 3.6.0 should be officially declared broken. If declare 3.6.1 broken we must port PR 1049 to 3.6 and release 3.6.1.1 or 3.6.2 as early as possible. Many third-party extensions built with 3.6.1 should be rebuild with a new release for

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: More collateral damage: apparently the workaround that Pandas used for this bug (#undef'ing PySlice_GetIndicesEx) broke PyPy, so now they need a workaround for the workaround: https://github.com/pandas-dev/pandas/pull/16192 Recording this here partly as a

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Stefan Krah added the comment: I'm with Nathaniel here: The fixed-bug is probably too obscure to warrant ABI breakage. -- nosy: +skrah ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: Pillow also had broken wheels up on pypi for a while; they've now put out a bug fix release that #undef's PySlice_GetIndicesEx, basically monkeypatching out the bugfix to get back to the 3.6.0 behavior: https://github.com/python-pillow/Pillow/issues/2479

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: Apparently this also broke pyqt for multiple users; here's the maintainers at conda-forge struggling to figure out the best workaround: https://github.com/conda-forge/pyqt-feedstock/pull/25 I really think it would be good to fix this in 3.6 sooner rather

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: FYI: https://github.com/pandas-dev/pandas/pull/16066 -- ___ Python tracker ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: New changeset 49a905958ffc2fcd5d1d1a293ae453d45deeb884 by Serhiy Storchaka in branch '3.5': [3.5] bpo-29943: Do not replace the function PySlice_GetIndicesEx() with a macro (#1049)

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: New changeset 89f9eb5b192b875c017d37cac16bd514aad9a801 by Serhiy Storchaka in branch '2.7': bpo-29943: Remove the PySlice_GetIndicesEx() macro. (#1050) https://github.com/python/cpython/commit/89f9eb5b192b875c017d37cac16bd514aad9a801 --

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: It looks like PyTorch got bitten by this: https://discuss.pytorch.org/t/pyslice-adjustindices-error/1475/11 -- ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Nathaniel Smith added the comment: > Can we consider 3.6.0 rather than 3.6.1 as broken release? In the last week, pypi downloads were about evenly split between 3.6.0 and 3.6.1 (2269969 for "3.6.1", 1927189 for "3.6.0", and those two were ~2 orders of magnitude more common than other strings

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: PR 1049 and PR 1050 restore ABI compatibility in 3.5 and 2.7. Unfortunately this restores the original bug with using PySlice_GetIndicesEx in third-party code (but CPython core and standard extensions no longer use PySlice_GetIndicesEx). Can we consider

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Serhiy Storchaka : -- pull_requests: +1202 ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Serhiy Storchaka : -- pull_requests: +1201 ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Christoph Gohlke : -- nosy: +cgohlke ___ Python tracker ___ ___ Python-bugs-list

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Charalampos Stratakis added the comment: Currently we haven't updated to Python 3.6.1 at Fedora 26 due to this issue. While it is a release blocker for 3.6.2, what can be done for 3.6.1? -- ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Miro Hrončok : -- nosy: +hroncok ___ Python tracker ___ ___ Python-bugs-list

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Charalampos Stratakis : -- nosy: +cstratak ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: 2.7 is affected too. Nathaniel, could you add references to this issue and to original issue27867 on the Fedora bug tracker? -- nosy: +benjamin.peterson versions: +Python 2.7 ___ Python tracker

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Serhiy Storchaka added the comment: My apologies for breaking the world. The workaround is to undefine PySlice_GetIndicesEx after #include "Python.h". #if PY_VERSION_HEX < 0x0307 && defined(PySlice_GetIndicesEx) #undef PySlice_GetIndicesEx #endif But this restores the initial bug. Other

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Changes by Steve Dower : -- nosy: +steve.dower ___ Python tracker ___ ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

Larry Hastings added the comment: Let's make it a release blocker for now. -- priority: normal -> release blocker ___ Python tracker ___

[issue29943] PySlice_GetIndicesEx change broke ABI in 3.5 and 3.6 branches

New submission from Nathaniel Smith: In the process of fixing issue 27867, a new function PySlice_AdjustIndices was added, and PySlice_GetIndicesEx was converted into a macro that calls this new function. The patch was backported to both the 3.5 and 3.6 branches, was released in 3.6.1, and is