nulano pushed to branch branch/win64 at PyPy / pypy


Commits:
f44c09a9 by Armin Rigo at 2020-07-30T15:30:00+02:00
get rid of the pre-version-80 support.  It was buggy because it expected output 
from cl.exe in English and wouldn't recognize the output in French, for 
example

- - - - -
81e0c554 by Armin Rigo at 2020-07-30T16:18:24+02:00
let the signal handler set a windows event, like it does in CPython 3.6

- - - - -
5c9c6385 by Matti Picus at 2020-07-30T17:37:29+03:00
tweak c4f64ef3a071 (version is shown on stdout when running tests)

- - - - -
17fcee55 by Matti Picus at 2020-07-30T17:33:16+03:00
fix skip to allow tests to run

- - - - -
bdefcbb1 by Matti Picus at 2020-07-31T07:22:54+03:00
initialize value

- - - - -
9044806a by Armin Rigo at 2020-07-31T09:05:47+02:00
add a define that we can test for elsewhere

- - - - -
5df33baa by Matti Picus at 2020-07-31T18:09:26+03:00
change interface for rposix_scandir - on win32 use utf8

- - - - -
ca920b27 by Matti Picus at 2020-08-03T08:04:51+03:00
/O2 or /Ox on MSVC does tail call optimization

- - - - -
c78bdbab by Matti Picus at 2020-08-02T19:52:22+03:00
rsocket tests - win32 has no recvmsg; socketgetsockname needs to be bound

- - - - -
73300620 by Matti Picus at 2020-08-03T21:14:16+03:00
backout 690ed94d970b and do PYPY_INHIBIT_TAIL_CALL for MSVC properly

- - - - -
d145555d by Armin Rigo at 2020-08-07T12:13:38+02:00
When "pypy -m pip" fails to find pip, give an error message that 
hints at "pypy
-m ensurepip"

- - - - -
2eb845af by Matti Picus at 2020-08-08T22:50:48+03:00
start release note, optimistically include hpy Universal mode

- - - - -
95a26954 by Julian Berman at 2020-07-18T15:52:08-04:00
Use timespec on macOS too.

It's present, the fields are just named slightly differently.

I think this fixes all but one test in rlib.rposix_stat, which
previously were failing.

The last one just needs some bifurcation due again to the
differing names in the timespec struct.

More broadly, the bug here is that this code has different
behavior on PyPy and CPython + macOS:

    touch foo
    for each in pypy3 python3.6; do
        echo $each; $each -c 'import os; 
print(os.stat("foo").st_mtime)'
    done

where on PyPy the *time attributes only give integer precision.

And yeah this actually came up -- a test I wrote for other code failed
on PyPy because the mtime of a file was changing by less than a second
of precision.

- - - - -
4d69d1aa by Julian Berman at 2020-07-18T16:03:40-04:00
Uglily fix test_high_precision_stat_time on macOS

Copypaste + modify the differing struct field names.

- - - - -
99a24228 by Julian Berman at 2020-07-18T16:12:47-04:00
Fixup the comments around timespec field naming.

- - - - -
d785b6cf by Matti Picus at 2020-08-09T20:59:31+03:00
update contributors list, 8 new names

- - - - -
78a847a9 by nulano at 2020-08-10T02:25:56+02:00
fix rffi.LONG / rffi.ULONG vs rffi.SIGNED / rffi.UNSIGNED in _pypyjson module

--HG--
branch : win64

- - - - -
db35dbf7 by nulano at 2020-08-10T02:39:21+02:00
Merge default

--HG--
branch : win64

- - - - -
1882b547 by nulano at 2020-08-10T02:40:44+02:00
win64 MSVC doesn't support __asm

--HG--
branch : win64

- - - - -


17 changed files:

- LICENSE
- lib-python/2.7/runpy.py
- pypy/doc/contributor.rst
- pypy/doc/index-of-release-notes.rst
- + pypy/doc/release-v7.3.2.rst
- pypy/module/_pypyjson/interp_decoder.py
- pypy/module/_pypyjson/simd.py
- rpython/rlib/rposix_scandir.py
- rpython/rlib/rposix_stat.py
- rpython/rlib/rsignal.py
- rpython/rlib/test/test_rposix_scandir.py
- rpython/rlib/test/test_rposix_stat.py
- rpython/rlib/test/test_rsocket.py
- rpython/rtyper/lltypesystem/rffi.py
- rpython/translator/c/src/signals.c
- rpython/translator/c/src/stack.h
- rpython/translator/platform/windows.py


View it on GitLab: 
https://foss.heptapod.net/pypy/pypy/-/compare/489f9edbc6c659a516470093bd037761c6e8efe8...1882b5473db82c90a33cf85ce468e320ef5db7fc

-- 
View it on GitLab: 
https://foss.heptapod.net/pypy/pypy/-/compare/489f9edbc6c659a516470093bd037761c6e8efe8...1882b5473db82c90a33cf85ce468e320ef5db7fc
You're receiving this email because of your account on foss.heptapod.net.


_______________________________________________
pypy-commit mailing list
pypy-commit@python.org
https://mail.python.org/mailman/listinfo/pypy-commit

Reply via email to