[pypy-commit] [Git][pypy/pypy][branch/win64] 149 commits: Check whether stream is closed in BufferedMixin.truncate_w().

Matti Picus foss at heptapod.net
Mon Oct 5 05:20:26 EDT 2020 


Matti Picus pushed to branch branch/win64 at PyPy / pypy


Commits:
02d287ec by Manuel Jacob at 2020-03-08T02:21:14+01:00
Check whether stream is closed in BufferedMixin.truncate_w().

Before this, the method was failing later on a closed stream when accessing the
underlying raw stream. It is more consistent and robust to check for a closed
stream earlier.

Starting with the next changeset, closing the stream will set the buffer to
None. Given that the buffer is potentially accessed by truncate_w(), it is
a very good idea to check that the stream is not closed, although it is hard
to make truncate_w() actually access the buffer when the stream is closed
(because closing it will flush the unwritten data from the buffer).

- - - - -
8d022e19 by Manuel Jacob at 2020-03-07T23:08:57+01:00
Set buffer of buffered IO classes to None in close().

This helps to reclaim the memory used by the buffer earlier.

- - - - -
0909ebb2 by Matti Picus at 2020-08-19T10:19:08+03:00
fix import in test

- - - - -
764001ba by Matti Picus at 2020-08-21T13:02:53+03:00
fixes for running untranslated tests with pypy2

- - - - -
b209d3f2 by Matti Picus at 2020-08-21T17:10:15+03:00
partial fix: turn a crash into an error on win32

- - - - -
55653e85 by Matti Picus at 2020-08-22T14:23:05+02:00
fix

- - - - -
35452c8d by Matti Picus at 2020-08-22T23:32:01+03:00
use FdValidator not is_valid_fd

- - - - -
090bfda5 by Matti Picus at 2020-08-23T10:02:10+03:00
rename JITLOG when running untranslated since the host python may be pypy

- - - - -
e557922e by Matti Picus at 2020-08-23T11:18:36+03:00
upgrade 26691d5fd920

- - - - -
c8463f46 by Armin Rigo at 2020-08-23T11:30:50+02:00
Use "JITLOG" in programs translated with rpython/bin/rpython,
and "JITLOG_FORTESTS" in all other cases (i.e. for tests)

- - - - -
f2f7b5a0 by Matti Picus at 2020-08-23T18:37:47+03:00
when running on top of pypy, the test inside app_main fails

- - - - -
6a51a223 by Matti Picus at 2020-08-24T05:42:43+00:00
Merge branch 'topic/default/bufferedio-release-buffer' into 'branch/default'

Set buffer to None on close of buffered io classes

See merge request pypy/pypy!711

- - - - -
41d663e4 by Matti Picus at 2020-08-24T11:29:36+03:00
use an environment varialble to detect test_app_main. Is this a security risk?

- - - - -
1a3adda2 by Matti Picus at 2020-08-24T11:31:59+03:00
fix test that fails in a pypy virtualenv untranslated

- - - - -
e90b4728 by Matti Picus at 2020-08-24T12:00:00+03:00
skip test on win32

- - - - -
385e34da by Armin Rigo at 2020-08-24T13:41:21+02:00
app_main: instead of guessing in we_are_translated(), just use a global
flag which is True, but set to False when running tests

- - - - -
a36bbe54 by Matti Picus at 2020-08-24T17:44:51+03:00
suppress IPH in fclose, pclose via new Suppress_IPH context manager

This should be used where CPython uses _Py_BEGIN_SUPPRESS_IPH, _Py_END_SUPPRESS_IPH

- - - - -
9e605908 by Matti Picus at 2020-08-24T18:22:06+03:00
test, implement sq_repeat wrapper (issue 3281)

- - - - -
10ba3e18 by Matti Picus at 2020-08-24T22:50:49+03:00
test, implemenst sq_inplace_repeat wrapper; cleanup from 0fd79fda9e0a

- - - - -
b7597585 by Matti Picus at 2020-08-25T00:53:36+03:00
prevent LeakChecker from erroring on test exit by moving import to setup

- - - - -
919b3097 by Matti Picus at 2020-08-25T08:40:37+03:00
typo

- - - - -
e28e4844 by Matti Picus at 2020-08-25T11:26:58+03:00
fix test, remove test for no-op function on MSVC10+

- - - - -
ed25c8a5 by Matti Picus at 2020-08-25T11:27:29+03:00
add SuppressIPH_del to be used inside __del__

- - - - -
e0492842 by Matti Picus at 2020-08-25T12:18:57+03:00
fix leak found by LeakFinder in tests

- - - - -
448ca6a9 by Matti Picus at 2020-08-25T13:02:17+03:00
typo and test rfile not posix

- - - - -
c65af24e by Matti Picus at 2020-08-25T15:15:08+03:00
add gc.collect before ALLOCATED check

- - - - -
2911ecba by Matti Picus at 2020-08-26T23:57:08+03:00
add a gc.collect for pypy

- - - - -
e3c81a24 by Matti Picus at 2020-08-27T00:14:59+03:00
dict order can change between cpython2 and pypy2. Test against both ordering

- - - - -
aca83dba by Matti Picus at 2020-08-27T12:21:09+03:00
test, fix for crash in strftime, will not crash when using msvcrt>=10

- - - - -
7fda7578 by Matti Picus at 2020-08-27T14:30:47+03:00
add missing import

- - - - -
c119b12d by Matti Picus at 2020-08-27T16:28:36+03:00
add missing import

- - - - -
14dadc2b by Matti Picus at 2020-08-28T08:19:08+03:00
properly skipif test

- - - - -
b0aaf43e by Armin Rigo at 2020-08-30T10:48:55+02:00
for now, skip tests that can't find the 'c' library (notably on Windows with PyPy)

- - - - -
f777c652 by Matti Picus at 2020-08-30T12:15:42+03:00
clean up unused import

- - - - -
be3b4f8a by Armin Rigo at 2020-08-30T11:18:44+02:00
this test is fragile on a non-translated pypy, skip it

- - - - -
b1949e10 by Armin Rigo at 2020-08-30T11:19:17+02:00
merge heads

- - - - -
b1730594 by Armin Rigo at 2020-08-30T11:24:28+02:00
add the tweaked test_c.py from cffi, which can now be run directly

- - - - -
ab4c0b0a by Matti Picus at 2020-08-30T13:28:44+03:00
make stack 3MB on windows

- - - - -
81aa90e2 by Matti Picus at 2020-08-30T13:44:13+03:00
typo

- - - - -
cb01408e by Armin Rigo at 2020-08-30T16:10:16+02:00
Trying to implement pypyjit.releaseall(), marking all machine codes as to be released

--HG--
branch : jit-releaseall

- - - - -
f28fdcda by Armin Rigo at 2020-08-30T17:01:28+02:00
add jit_hooks.stats_memmgr_release_all()

--HG--
branch : jit-releaseall

- - - - -
f6d4891c by Armin Rigo at 2020-08-30T17:10:02+02:00
explain the return value of this function in the docstring

- - - - -
a2a08253 by Armin Rigo at 2020-08-30T17:13:03+02:00
Add the app-level hook 'pypyjit.releaseall()'

--HG--
branch : jit-releaseall

- - - - -
5ae5b52f by Armin Rigo at 2020-08-30T19:10:55+00:00
Merge branch 'branch/jit-releaseall' into 'branch/default'

Add pypyjit.releaseall()

pypyjit.releaseall() is a helper that marks all current machine code objects as ready to release.  They will be released at the next GC (unless they are currently in use in the stack of one of the threads).  Doing `pypyjit.releaseall(); gc.collect()` is a heavy hammer that forces the JIT roughly to the state of a newly started PyPy.  It might be useful in some cases where our GC strategy is not good enough, e.g. if there is an ever-growing number of bridges out of the same loop, which can occur e.g. if the loop is in stable code but contains inline bridges that depend on code or classes that keep being (re-)created.

See merge request pypy/pypy!749

- - - - -
893217a9 by Matti Picus at 2020-08-31T09:01:00+03:00
document new function, fix translation

- - - - -
f8c00c2e by Armin Rigo at 2020-09-02T12:48:49+02:00
Rewrite code like the app-level "lst[i+1]" to fold the "+1" inside the
constant offset inside the low-level array (the overflow check, if any,
remains)

- - - - -
b5fb435b by Armin Rigo at 2020-09-02T15:56:28+02:00
Some intbound checking around the 'uint' comparisons

- - - - -
15f87937 by Ronan Lamy at 2020-09-02T16:39:42+01:00
reduce diff with py3

- - - - -
84f37171 by Armin Rigo at 2020-09-02T21:16:56+02:00
fix for e11b82085fc9 (hard to test, it requires careful consideration of
when to call get_box_replacement and when not to call it)

- - - - -
d3d85756 by Armin Rigo at 2020-09-03T08:17:02+02:00
update to cffi/108825c76286

- - - - -
9603e120 by Armin Rigo at 2020-09-03T08:33:54+02:00
fix for 90c5a06b0923

- - - - -
6ec3fd76 by Armin Rigo at 2020-09-03T08:40:50+02:00
add some prints in an attempt to understand this failure, which doesn't show up
locally

- - - - -
2f249602 by Armin Rigo at 2020-09-03T11:14:55+02:00
Disable the work done on optimize_UINT_LT & friends, with explanation

- - - - -
0c7c37b1 by Matti Picus at 2020-09-06T14:00:20+03:00
update release note

- - - - -
e7f7ad3b by Matti Picus at 2020-09-06T14:04:48+03:00
update the HEAD version to 7.3.3

- - - - -
6bab6f69 by Matti Picus at 2020-09-06T14:09:53+03:00
restart whatsnew

- - - - -
72bac5e9 by Matti Picus at 2020-09-06T16:21:02+03:00
fix some documentation build warnings and errors, comment out hpy from release

- - - - -
cea16152 by Matti Picus at 2020-09-06T16:24:07+03:00
change all http to https, fix some broken links

- - - - -
dbb5c4b6 by Ronan Lamy at 2020-09-08T17:10:47+01:00
Use rffi.ptradd() instead of a weird cast

- - - - -
8e89b9a3 by Matti Picus at 2020-09-08T19:44:04+03:00
Added tag release-pypy3.6-v7.3.2rc1 for changeset 3d817608d63b

- - - - -
70fd1164 by Matti Picus at 2020-09-08T19:44:19+03:00
Added tag release-pypy3.7-v7.3.2rc1 for changeset 55adbaed6eda

- - - - -
27dd7e9a by Matti Picus at 2020-09-08T19:44:41+03:00
Added tag release-pypy2.7-v7.3.2rc1 for changeset 9d418ec24144

- - - - -
84c31c50 by cjmcdonald at google.com at 2020-09-08T15:25:11-06:00
Respect system PKG_CONFIG environment variable

Many cross-compilation setups use custom pkg-config wrappers to ensure that
the proper version of libraries get linked against. The conventional way to
specify such a wrapper is via the PKG_CONFIG environment variable, so change
the default pkg_config logic to use the contents of that variable if available.

--HG--
branch : cross_compilation_fixes

- - - - -
948a417e by cjmcdonald at google.com at 2020-09-08T15:30:57-06:00
Only set CC variable if otherwise unspecified

This probing logic tests for the existence of `gcc` and sets the value of
the 'CC' variable based on the results of that test, but many build pipelines
already define a value for 'CC' that they don't want overridden. Change this
probing logic to only set 'CC' if the environment variable doesn't already have
a specified value.

--HG--
branch : cross_compilation_fixes

- - - - -
71141863 by Matti Picus at 2020-09-09T11:56:46+03:00
update release process

- - - - -
18d9c67f by Matti Picus at 2020-09-09T11:57:21+03:00
backport fix for bpo-39017

- - - - -
d52d46ac by Matti Picus at 2020-09-09T13:30:52+03:00
add sphinx-affiliates which allows searching across other sphinx sites

- - - - -
cdb4d846 by Matti Picus at 2020-09-09T16:15:15+03:00
add the extension to rpython as well

- - - - -
bd337d33 by Armin Rigo at 2020-09-09T15:02:16+00:00
Merge branch 'branch/cross_compilation_fixes' into 'branch/default'

Cross-compilation cleanups

See merge request pypy/pypy!750

- - - - -
029a00da by Armin Rigo at 2020-09-09T17:45:30+02:00
A fix for a corner case (shown by test_ztranslation in branch 'hpy').
A non-fix as a skipped test for a similar, even-more-obscure case.

- - - - -
30ddf8f9 by Armin Rigo at 2020-09-09T22:47:47+02:00
next similar issue (73e7c4e0b67d)

- - - - -
e7e2dead by Matti Picus at 2020-09-10T07:35:59+03:00
typo, update whatsnew

- - - - -
8f27f38a by Matti Picus at 2020-09-10T07:40:25+03:00
add test file

- - - - -
b6da8a63 by Matti Picus at 2020-09-10T12:07:19+03:00
test, implement easy part of PyMemoryView_GetContiguous

- - - - -
85a16ef3 by Michał Górny at 2020-09-10T15:55:03+02:00
sync httplib2 and urllib2 to cpython 2.7 git with security backports

Sync httplib and urllib2 stdlib modules and the respective tests
to the current state of CPython 2.7 git (EOL-ed) + two patches
backported from 3.6 that are present in the Gentoo patchset.  This has
only trivial changes compared to backporting the four relevant patches
and should make it easier to apply a final stdlib update post-release.

The Gentoo patches can be found as the two top patches on
https://gitweb.gentoo.org/fork/cpython.git/log/?h=gentoo-2.7.18-r2

- - - - -
50c53a77 by Armin Rigo at 2020-09-10T17:07:02+02:00
Issue #3297

A corner case that produces a bogus loop containing getarrayitem_gc with
index -1, or with whatever index is given in the initial value of 'pos'.
Negative values would segfault; positive values would allocate '8 * pos'
bytes of memory however large the value is.  Try to fix all these cases.

- - - - -
dd9cdb9e by Armin Rigo at 2020-09-10T18:00:01+02:00
A test that fails on some backends, will fix

- - - - -
a837a163 by Armin Rigo at 2020-09-10T18:06:18+02:00
fix aarch64 for 3495b2305437

- - - - -
e47fa51e by Armin Rigo at 2020-09-10T18:11:50+02:00
fix arm for 3495b2305437 (probably, it's like aarch64, but not tested)

- - - - -
b2a983a3 by Matti Picus at 2020-09-10T20:14:56+00:00
Merge branch 'topic/default/py27-backport-secfixes' into 'branch/default'

sync httplib2 and urllib2 to cpython 2.7 git with security backports

See merge request pypy/pypy!754

- - - - -
a246d030 by Matti Picus at 2020-09-10T23:45:40+03:00
add latest changes to release note

- - - - -
cd731cf0 by Matti Picus at 2020-09-11T10:50:40+03:00
fix translation

- - - - -
16161f05 by Armin Rigo at 2020-09-11T14:44:38+02:00
oops sorry, fix

- - - - -
50e2497e by Armin Rigo at 2020-09-11T14:51:35+02:00
Fix this test for 32-bit

- - - - -
8795a5a6 by Armin Rigo at 2020-09-11T15:03:40+02:00
Test fix: workaround for limited length of UNIXAddresses

- - - - -
9c36f683 by Armin Rigo at 2020-09-12T10:02:35+02:00
backport bpo-38243 because it is classified as a CVE

- - - - -
3145a409 by Matti Picus at 2020-09-13T00:27:13+03:00
test, implement PyNumber_ToBase

- - - - -
1e9b3172 by Matti Picus at 2020-09-13T11:39:24+03:00
skip test that crashes uptranslated

- - - - -
84352e45 by Matti Picus at 2020-09-13T18:30:30+03:00
refactor force-builds script

- - - - -
28615e94 by Matti Picus at 2020-09-13T23:59:08+03:00
use win32_14x or win64_14x branch name for exernals

- - - - -
af67d261 by Matti Picus at 2020-09-14T18:32:55+03:00
update release note, add note about release status

- - - - -
f1377391 by Matti Picus at 2020-09-14T20:49:26+03:00
typo

- - - - -
7ff7bd6c by Matti Picus at 2020-09-15T00:13:17+03:00
typo

- - - - -
5830135d by Michał Górny at 2020-09-15T08:00:57+02:00
sync test_xmlrpc to CPython 2.7 to fix test failure

- - - - -
0ac5b28d by Matti Picus at 2020-09-15T09:47:51+03:00
typo (again). which will happen first: correctly formatting or the release

- - - - -
2bf6d207 by Matti Picus at 2020-09-16T14:19:10+03:00
Added tag release-pypy2.7-v7.3.2rc2 for changeset 513d750d64de

- - - - -
a7742911 by Matti Picus at 2020-09-16T14:19:32+03:00
Added tag release-pypy3.6-v7.3.2rc2 for changeset bef50b0f3fe7

- - - - -
d728bc1a by Matti Picus at 2020-09-16T14:19:46+03:00
Added tag release-pypy3.7-v7.3.2rc2 for changeset ade3eeb8331f

- - - - -
fa54da53 by Dan Villiom Podlaski Christiansen at 2020-09-17T12:38:45+02:00
darwin: don't define _POSIX_C_SOURCE, as it hides e.g. getpagesize()

- - - - -
74e9314f by Dan Villiom Podlaski Christiansen at 2020-09-17T12:51:25+02:00
darwin: sysctlbyname is in sys/sysctl.h

- - - - -
a5639069 by Dan Villiom Podlaski Christiansen at 2020-09-17T12:55:34+02:00
fix inconsistent #ifdef's surrounding get_vmprof_stack()

- - - - -
8dfa45d2 by Dan Villiom Podlaski Christiansen at 2020-09-17T12:56:22+02:00
darwin: strlen() is in <string.h>

- - - - -
d0441a61 by Dan Villiom Podlaski Christiansen at 2020-09-17T13:33:16+02:00
boehm gc: try using pkg-config to find it

- - - - -
e4f103a2 by Dan Villiom Podlaski Christiansen at 2020-09-17T14:17:03+02:00
rvmprof: avoid missing prototype, which is an error in Clang 11

- - - - -
017dc84d by Dan Villiom Podlaski Christiansen at 2020-09-17T16:38:55+02:00
fix curses build

- - - - -
0f38a218 by Dan Villiom Podlaski Christiansen at 2020-09-17T17:35:25+02:00
build_cffi_imports: use an HTTP mirror; fix building gdbm on Darwin

- - - - -
de23b457 by Matti Picus at 2020-09-22T10:38:52+03:00
mangle the _Py_subtype_dealloc name so tests on a pypy2 host will not be confused

This is needed due to the code using the function pointer internally

- - - - -
8232fed9 by Matti Picus at 2020-09-22T11:22:31+03:00
lzma is not available via http, only https

- - - - -
567a5f14 by Antonio Cuni at 2020-09-22T10:35:52+02:00
remove duplicate entry and ignore the content of the bin/ directory (which is useful if you ./pypy -m pip install stuff inside your working copy

- - - - -
fc501ec5 by Matti Picus at 2020-09-22T12:09:34+03:00
fix link, http works but the suffix is bz2

- - - - -
c101bf47 by Antonio Cuni at 2020-09-22T11:48:51+02:00
Suppress the cpyext warnings like:
   missing slot '__setattr__'/'tp_setattr', discovered on 'object'
   ...

They have been around forever and nobody looks at them anyway

- - - - -
afa884f1 by Matti Picus at 2020-09-22T15:57:31+03:00
follow CPython and include wchar.h in unicodeobject.h, it is needed elsewhere

- - - - -
05bcd838 by Stefano Rivera at 2020-09-22T14:35:38-07:00
Update path to build_cffi_imports.py

- - - - -
a66270a3 by Matti Picus at 2020-09-23T10:29:56+03:00
python is no longer a valid command on some systems (tumbleweed)

- - - - -
e2a2b739 by Armin Rigo at 2020-09-23T14:19:20+02:00
update to cffi/144ac54d9bff

- - - - -
460470ea by Matti Picus at 2020-09-24T20:44:28+03:00
Added tag release-pypy2.7-v7.3.2rc3 for changeset 6abe2e00c51d

- - - - -
81c5481d by Matti Picus at 2020-09-24T20:44:49+03:00
Added tag release-pypy3.6-v7.3.2rc3 for changeset d38cd66c14b8

- - - - -
6433fe08 by Matti Picus at 2020-09-24T20:45:05+03:00
Added tag release-pypy3.7-v7.3.2rc3 for changeset 87875bf2dfd8

- - - - -
4a7d1478 by Matti Picus at 2020-09-24T20:50:47+03:00
guess at a fix

- - - - -
70c354b6 by Matti Picus at 2020-09-24T23:02:57+03:00
get rid of mentions of bitbucket

- - - - -
e00ab6a1 by Matti Picus at 2020-09-24T23:09:34+03:00
Added tag release-pypy2.7-v7.3.2 for changeset 6abe2e00c51d

- - - - -
f29e139a by Matti Picus at 2020-09-24T23:09:53+03:00
Added tag release-pypy3.6-v7.3.2 for changeset d38cd66c14b8

- - - - -
a745bc08 by Matti Picus at 2020-09-24T23:10:19+03:00
Added tag release-pypy3.7-v7.3.2 for changeset 87875bf2dfd8

- - - - -
72234cba by Matti Picus at 2020-09-24T23:27:11+03:00
update release note

- - - - -
b94c3c6d by Matti Picus at 2020-09-25T09:28:26+03:00
redo the rvmprof changes in a way that I think can be upstreamed

- - - - -
63b4501a by Matti Picus at 2020-09-25T10:57:11+03:00
Rework, based on comments from LarstiQ on IRC

- - - - -
95d42207 by Matti Picus at 2020-09-25T11:23:53+03:00
close resources in test

- - - - -
f293b45b by Matti Picus at 2020-09-25T11:38:38+03:00
increase the tolerance a bit for latency inside a docker

- - - - -
913e0dae by Armin Rigo at 2020-09-26T09:26:24+02:00
oops, fix for test_gc_indexed_box_plus_large_offset

- - - - -
c9be894f by Armin Rigo at 2020-09-26T10:41:02+02:00
skip a segfaulting test on ppc big-endian.  more complex tests using the whole
backend work fine, and also going back in time doesn't stop this segfault, so
maybe it is some unrelated issue

- - - - -
525479ae by Armin Rigo at 2020-09-26T10:57:52+02:00
skip another test that now crashes on ppc big-endian

- - - - -
30c1deda by Matti Picus at 2020-09-27T01:00:52+03:00
try to add a flag to support MacOS<=10.13

- - - - -
fdf249b1 by Matti Picus at 2020-09-27T01:05:29+03:00
typo

- - - - -
bb007914 by Stefano Rivera at 2020-09-26T21:13:09-07:00
Replace another python call. Debian removed the /usr/bin/python binary

- - - - -
71d608cb by Armin Rigo at 2020-09-28T13:42:11+02:00
issue3312: some limited support for longdouble constants

- - - - -
d91f3c43 by Matti Picus at 2020-09-28T23:01:57+03:00
reapply PyPy fixes to urllib2.py after 1e7b78d6f418, fixes issue 3313

- - - - -
12083530 by Dan Villiom Podlaski Christiansen at 2020-09-22T18:30:24+02:00
build_cffi_imports: don't assume archive suffix is .gz

- - - - -
8333e438 by Dan Villiom Podlaski Christiansen at 2020-09-23T23:48:00+02:00
rposix: implement sendfile on darwin (and freebsd?)

This is not available in CPython 2.7 whether from the OS or MacPorts,
so I've applied this against the `py3.6` branch.

--HG--
branch : darwin-sendfile-2.7

- - - - -
869a2990 by Matti Picus at 2020-09-29T01:14:03+03:00
revert be9ad752d782, set MACOSX_DEPLOYMENT_TARGET in sysconfig to fix issue 3311

- - - - -
76802dfc by Matti Picus at 2020-09-30T09:13:08+03:00
document preference for named branches

- - - - -
98652c40 by Matti Picus at 2020-10-02T09:55:02+00:00
Merge branch 'branch/darwin-sendfile-2.7' into 'branch/default'

rposix: implement sendfile on darwin (and freebsd?)

See merge request pypy/pypy!763

- - - - -
fe78f8cf by Matti Picus at 2020-10-02T17:57:58+03:00
update repackage script to create .hg_archival.txt, issue 3315

- - - - -
7461549e by Matti Picus at 2020-10-03T19:20:51+03:00
move translated os.* tests to extra_tests

- - - - -
64e7a519 by Matti Picus at 2020-10-03T19:32:57+03:00
redo moved tests, flake8

- - - - -
609406d4 by Matti Picus at 2020-10-03T19:33:09+03:00
small fixes

- - - - -
21f2bcd1 by Matti Picus at 2020-10-03T20:07:58+03:00
restart whatsnew-pypy3-head.rst

- - - - -
e3549d0f by Matti Picus at 2020-10-03T20:33:35+03:00
simplify source repackaging via hg archive, creates .hg_archival.txt, issue 3315

- - - - -
7f4f78a8 by Antonio Cuni at 2020-10-04T15:24:36+02:00
improve the fake objspace so that it fails if you pass a resizable list to newtuple, and to catch the problem which was fixed by 0726e91a1ce1

- - - - -
8f4a1ba5 by Matti Picus at 2020-10-05T12:19:00+03:00
merge default into branch

--HG--
branch : win64

- - - - -


30 changed files:

- .hgignore
- .hgtags
- Makefile
- + extra_tests/cffi_tests/test_c.py
- extra_tests/test_binop_overriding.py
- + extra_tests/test_os.py
- extra_tests/test_semlock.py
- lib-python/2.7/DocXMLRPCServer.py
- lib-python/2.7/httplib.py
- lib-python/2.7/site.py
- lib-python/2.7/subprocess.py
- lib-python/2.7/sysconfig.py
- lib-python/2.7/tarfile.py
- + lib-python/2.7/test/recursion.tar
- lib-python/2.7/test/test_docxmlrpc.py
- lib-python/2.7/test/test_httplib.py
- lib-python/2.7/test/test_tarfile.py
- lib-python/2.7/test/test_urllib2.py
- lib-python/2.7/test/test_xmlrpc.py
- lib-python/2.7/urllib.py
- lib-python/2.7/urllib2.py
- lib_pypy/_curses_build.py
- lib_pypy/cffi.egg-info/PKG-INFO
- lib_pypy/cffi/__init__.py
- lib_pypy/cffi/_embedding.h
- lib_pypy/pypy_tools/build_cffi_imports.py
- pypy/doc/architecture.rst
- pypy/doc/build.rst
- pypy/doc/coding-guide.rst
- pypy/doc/conf.py


View it on GitLab: https://foss.heptapod.net/pypy/pypy/-/compare/246d52d8da60a8913e3107bd3fb7b7b137775416...8f4a1ba5be22d83a086f8c2a2f978176484b21d9

-- 
View it on Heptapod: https://foss.heptapod.net/pypy/pypy/-/compare/246d52d8da60a8913e3107bd3fb7b7b137775416...8f4a1ba5be22d83a086f8c2a2f978176484b21d9
You're receiving this email because of your account on foss.heptapod.net.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/pypy-commit/attachments/20201005/4890ecdb/attachment-0001.html>

More information about the pypy-commit mailing list