From tracker at bugs.pypy.org  Thu Mar  1 04:31:56 2012
From: tracker at bugs.pypy.org (Victor)
Date: Thu, 01 Mar 2012 03:31:56 +0000
Subject: [pypy-issue] [issue1075] Fatal RPython error: AssertionError
In-Reply-To: <1330572716.45.0.854227167709.issue1075@bugs.pypy.org>
Message-ID: <1330572716.45.0.854227167709.issue1075@bugs.pypy.org>


New submission from Victor <victor3 at gmail.com>:

pypy 1.8.1 on Ubuntu 11.10 x64.

This is an Ant Colony Optimization algorithm for the Quadratic Assignment 
Problem.  The test problem is had12.dat from the QAP Library 
<http://www.seas.upenn.edu/qaplib/>.  Any numpy functions used are compatible 
with pypy.

When running the code, the following error is produced:

RPython traceback:
  File "jit_metainterp_compile.c", line 19027, in send_bridge_to_backend
  File "jit_backend_x86_assembler.c", line 54970, in 
Assembler386_assemble_bridge
  File "jit_backend_x86_regalloc.c", line 1041, in RegAlloc__prepare
  File "jit_backend_x86_regalloc.c", line 4984, in 
RegAlloc__compute_vars_longevity
Fatal RPython error: AssertionError
Aborted

----------
files: pypy-assertionerror.py
messages: 4035
nosy: pypy-issue, vcaligi1
priority: bug
status: unread
title: Fatal RPython error: AssertionError

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1075>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  1 06:25:12 2012
From: tracker at bugs.pypy.org (mikefc)
Date: Thu, 01 Mar 2012 05:25:12 +0000
Subject: [pypy-issue] [issue1075] Fatal RPython error: AssertionError
In-Reply-To: <1330572716.45.0.854227167709.issue1075@bugs.pypy.org>
Message-ID: <1330579512.47.0.0070355829607.issue1075@bugs.pypy.org>


mikefc <coolbutuseless at gmail.com> added the comment:

On latest pypy (f9f3b57f1300) on OSX I don't see this error.

----------
nosy: +mikefc
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1075>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  1 11:27:20 2012
From: tracker at bugs.pypy.org (Ronny Pfannschmidt)
Date: Thu, 01 Mar 2012 10:27:20 +0000
Subject: [pypy-issue] [issue1075] Fatal RPython error: AssertionError
In-Reply-To: <1330572716.45.0.854227167709.issue1075@bugs.pypy.org>
Message-ID: <1330597639.95.0.578155632571.issue1075@bugs.pypy.org>


Ronny Pfannschmidt <Ronny.Pfannschmidt at gmx.de> added the comment:

just tested on 3ea73d1d6fb1, where it works
however it failed on a older nightly i tested with before

so i think that issue got fixed recently

----------
nosy: +ronny

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1075>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  1 16:07:24 2012
From: tracker at bugs.pypy.org (Victor)
Date: Thu, 01 Mar 2012 15:07:24 +0000
Subject: [pypy-issue] [issue1075] Fatal RPython error: AssertionError
In-Reply-To: <1330572716.45.0.854227167709.issue1075@bugs.pypy.org>
Message-ID: <1330614443.9.0.981538724677.issue1075@bugs.pypy.org>


Victor <victor3 at gmail.com> added the comment:

Fixed in 53032-f9f3b57f1300 nightly

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1075>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  1 21:46:25 2012
From: tracker at bugs.pypy.org (Hakan Ardo)
Date: Thu, 01 Mar 2012 20:46:25 +0000
Subject: [pypy-issue] [issue1072] Fatal RPython error: OverflowError
In-Reply-To: <1330327260.33.0.812176274841.issue1072@bugs.pypy.org>
Message-ID: <1330634785.17.0.470341423131.issue1072@bugs.pypy.org>


Hakan Ardo <hakan at debian.org> added the comment:

Hi,
this was hopefully fixed by 90010b41b565, but needs verification on tomorrows 
nigthly.

----------
nosy: +hakanardo
status: chatting -> in-progress

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1072>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 11:14:11 2012
From: tracker at bugs.pypy.org (Tobias Oberstein)
Date: Fri, 02 Mar 2012 10:14:11 +0000
Subject: [pypy-issue] [issue1076] Finish kqueue support
In-Reply-To: <1330683251.67.0.917645290504.issue1076@bugs.pypy.org>
Message-ID: <1330683251.67.0.917645290504.issue1076@bugs.pypy.org>


New submission from Tobias Oberstein <tobias.oberstein at gmail.com>:

Alex Gaynor has started with kqueue support on branch "kqueue":

https://bitbucket.org/pypy/pypy/src/cc37b011ea0f
https://bitbucket.org/pypy/pypy/compare/kqueue..default

Finish that, make unit tests succeed, make it ready for trunk.

IOW: make kqueue on PyPy and BSDish platforms officially supported.

----------
assignedto: oberstet
messages: 4040
nosy: oberstet, pypy-issue
priority: feature
status: in-progress
title: Finish kqueue support

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1076>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 13:49:11 2012
From: tracker at bugs.pypy.org (Tobias Oberstein)
Date: Fri, 02 Mar 2012 12:49:11 +0000
Subject: [pypy-issue] [issue1076] Finish kqueue support
In-Reply-To: <1330683251.67.0.917645290504.issue1076@bugs.pypy.org>
Message-ID: <1330692551.04.0.923379000137.issue1076@bugs.pypy.org>


Tobias Oberstein <tobias.oberstein at gmail.com> added the comment:

started on https://bitbucket.org/oberstet/pypy/ branch kqueue

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1076>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 19:11:02 2012
From: tracker at bugs.pypy.org (Hakan Ardo)
Date: Fri, 02 Mar 2012 18:11:02 +0000
Subject: [pypy-issue] [issue1072] Fatal RPython error: OverflowError
In-Reply-To: <1330327260.33.0.812176274841.issue1072@bugs.pypy.org>
Message-ID: <1330711862.58.0.921559815158.issue1072@bugs.pypy.org>


Hakan Ardo <hakan at debian.org> added the comment:

Hi,
this no longer crashes on latest night (2881f17e1ffc).

----------
status: in-progress -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1072>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 19:13:11 2012
From: tracker at bugs.pypy.org (Hakan Ardo)
Date: Fri, 02 Mar 2012 18:13:11 +0000
Subject: [pypy-issue] [issue1071] jit crash
In-Reply-To: <1330303926.16.0.793903164103.issue1071@bugs.pypy.org>
Message-ID: <1330711991.52.0.483194044861.issue1071@bugs.pypy.org>


Hakan Ardo <hakan at debian.org> added the comment:

I've now verified that this no longer crashed with latest nightly (2881f17e1ffc).

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1071>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 19:19:51 2012
From: tracker at bugs.pypy.org (Hakan Ardo)
Date: Fri, 02 Mar 2012 18:19:51 +0000
Subject: [pypy-issue] [issue1045] Crash of the JIT
In-Reply-To: <1329140443.89.0.137027286035.issue1045@bugs.pypy.org>
Message-ID: <1330712391.67.0.54657686587.issue1045@bugs.pypy.org>


Hakan Ardo <hakan at debian.org> added the comment:

I've run this now several times using latest nigthly (2881f17e1ffc) and it's 
working just fine every time.

----------
status: in-progress -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1045>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  2 19:42:45 2012
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 02 Mar 2012 18:42:45 +0000
Subject: [pypy-issue] [issue1045] Crash of the JIT
In-Reply-To: <1329140443.89.0.137027286035.issue1045@bugs.pypy.org>
Message-ID: <1330713765.43.0.501789672545.issue1045@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Daid: for completeness: the "--jit enable_opts" option is indeed reserved for
internal use, but the complete list is the one you gave plus ":unroll" at the
end.  The unroll optimization is precisely where this occurs :-)

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1045>
________________________________________

From tracker at bugs.pypy.org  Sun Mar  4 01:12:30 2012
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Sun, 04 Mar 2012 00:12:30 +0000
Subject: [pypy-issue] [issue1077] Comparison with tuples and more
In-Reply-To: <1330819950.05.0.0415056089507.issue1077@bugs.pypy.org>
Message-ID: <1330819950.05.0.0415056089507.issue1077@bugs.pypy.org>


New submission from Armin Rigo <armin.rigo at gmail.com>:

A bug in the sense that PyPy (>>>>) gives a different behavior than CPython (>>>):

>>> (nan,) <= (5,)
False
>>>> (nan,) <= (5,)
True

Similarly:

>>> (set([6]),) <= (set([5]),)
False
>>>> (set([6]),) <= (set([5]),)
True

The issue is about "<=", which in PyPy is defined by default as "not >".  This
can be seen by trying this:

>>>> (nan,) <= (5,)
True
>>>> (nan,) > (5,)
False
>>> (nan,) <= (5,)
False
>>> (nan,) > (5,)
False

----------
messages: 4046
nosy: arigo, pypy-issue
priority: bug
status: unread
title: Comparison with tuples and more

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1077>
________________________________________

From tracker at bugs.pypy.org  Sun Mar  4 15:16:31 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 04 Mar 2012 14:16:31 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1330870591.76.0.00627472160535.issue1069@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

cpyext has never provided these function, but recently started to #define 
WITH_THREAD, which enabled the thread-aware version of some pyOpenSSL macros 
(MY_BEGIN_ALLOW_THREADS)

----------
nosy: +afa
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Sun Mar  4 17:58:38 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 04 Mar 2012 16:58:38 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1330880318.24.0.992149549516.issue1069@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Added these functions in fc46fb928ddb (by copying the code from CPython), tell me 
if pyOpenSSL now works.

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Sun Mar  4 20:33:48 2012
From: tracker at bugs.pypy.org (Alex Gaynor)
Date: Sun, 04 Mar 2012 19:33:48 +0000
Subject: [pypy-issue] [issue1048] pypy -m cProfile: jit bug
In-Reply-To: <1329301755.34.0.765793068872.issue1048@bugs.pypy.org>
Message-ID: <1330889627.99.0.174937901181.issue1048@bugs.pypy.org>


Alex Gaynor <alex.gaynor at gmail.com> added the comment:

Marking as resolved

----------
nosy: +agaynor
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1048>
________________________________________

From tracker at bugs.pypy.org  Mon Mar  5 11:57:53 2012
From: tracker at bugs.pypy.org (Daid)
Date: Mon, 05 Mar 2012 10:57:53 +0000
Subject: [pypy-issue] [issue1045] Crash of the JIT
In-Reply-To: <1329140443.89.0.137027286035.issue1045@bugs.pypy.org>
Message-ID: <1330945073.57.0.798005441416.issue1045@bugs.pypy.org>


Daid <daid303 at gmail.com> added the comment:

Arigo: Thanks for the extra info.
For completeness: I didn't have to chance to test it earlier, but I just tested
it and I also no longer get any crashes.

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1045>
________________________________________

From tracker at bugs.pypy.org  Mon Mar  5 13:18:51 2012
From: tracker at bugs.pypy.org (Jean-Paul Calderone)
Date: Mon, 05 Mar 2012 12:18:51 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1330949931.37.0.77051986802.issue1069@bugs.pypy.org>


Jean-Paul Calderone <exarkun at twistedmatrix.com> added the comment:

pyOpenSSL builds without warnings now, and imports.  Thanks.  However, it looks
like at least one case isn't handled in the same way as on CPython.

$ python -c '
import OpenSSL
conn = OpenSSL.SSL.Connection(OpenSSL.SSL.Context(OpenSSL.SSL.TLSv1_METHOD), None)
conn.set_connect_state()
conn.do_handshake()
'
Traceback (most recent call last):
  File "<string>", line 5, in <module>
OpenSSL.SSL.WantReadError
$

vs PyPy:

$ ~/pypy-c-jit-53221-27d215507bad-linux64/bin/pypy -c '
import OpenSSL
conn = OpenSSL.SSL.Connection(OpenSSL.SSL.Context(OpenSSL.SSL.TLSv1_METHOD), None)
conn.set_connect_state()
conn.do_handshake()
'
pypy: thread.c:238: PyThread_set_key_value: Assertion `value != ((void *)0)' failed.
Aborted
$

You might be able to convince me that PyThread_delete key should be used instead
of PyThread_set_key_value(key, NULL), but just being compatible with CPython
would be easier for me (clearly ;).

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Mon Mar  5 13:53:24 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Mon, 05 Mar 2012 12:53:24 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1330952004.33.0.644031317298.issue1069@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Hmm, I copied the code verbatim, from CPython 2.7.  Your interpreter is certainly 
compiled with assertions disabled.

PyEval_SaveThread() always return NULL in pypy.  But I don't know if there is a 
better value, since the GIL is always released in C code.
Or we could again use an empty definition for MY_BEGIN_ALLOW_THREADS...

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Tue Mar  6 01:39:48 2012
From: tracker at bugs.pypy.org (Fraser Tweedale)
Date: Tue, 06 Mar 2012 00:39:48 +0000
Subject: [pypy-issue] [issue1078] comparison between dictview and frozendict
	is wrong
In-Reply-To: <1330994388.53.0.80561785434.issue1078@bugs.pypy.org>
Message-ID: <1330994388.53.0.80561785434.issue1078@bugs.pypy.org>


New submission from Fraser Tweedale <frasert at benon.com>:

Comparing dictview to frozendict produces wrong result (PyPy 1.8.0).

{'a':1}.viewkeys() == set(('a'))  # True in CPy and PyPy
{'a':1}.viewkeys() == frozenset(('a'))  # True in CPy, False in PyPy

----------
messages: 4053
nosy: frasertweedale, pypy-issue
priority: bug
status: unread
title: comparison between dictview and frozendict is wrong

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1078>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  8 23:49:36 2012
From: tracker at bugs.pypy.org (Justin Patrin)
Date: Thu, 08 Mar 2012 22:49:36 +0000
Subject: [pypy-issue] [issue1080] jit errors (infinite loop and str instead
	of token object)
In-Reply-To: <1331246976.69.0.646014776853.issue1080@bugs.pypy.org>
Message-ID: <1331246976.69.0.646014776853.issue1080@bugs.pypy.org>


New submission from Justin Patrin <papercrane at reversefold.com>:

The attached tarball has a scheme interpreter I wrote for fun and two scripts 
which have problems when run with pypy with a jit. If the jit is turned off they 
work as expected. I've narrowed the jit options down to 
enable_opts=intbounds:rewrite:virtualize:string:heap:unroll which cause the 
errors to occur. If all jit options except for unroll are specified 
(enable_opts=intbounds:rewrite:virtualize:string:earlyforce:pure:heap:ffi) then 
the issues go away, so it stands to reason that this is "unroll" related.

Running str_instead_of_token.py gets an error where a 'str' object is somehow in 
the place of one of my 'token' objects.

Running infinite_loop.py gets into an infinite loop when running the last enabled 
test.

I have tested this with the newest version of pypy in hg:
$ hg summary                                                                                                                                
(jpatrin at jpatrin-mbp-lol:..ypy/translator/goal)
parent: 53272:d8c92dd23635 tip
 (fijal, arigo)
branch: default
commit: 1511 unknown (clean)
update: (current)

built with:
./pypy-c translate.py --opt=jit --output=pypy-c.new --continuation --thread --
gc=minimark --listcompr --clever-malloc-removal --insist --
gcrootfinder=shadowstack targetpypystandalone.py

----------
files: jit_errors.tbz2
messages: 4055
nosy: papercrane, pypy-issue
priority: bug
status: unread
title: jit errors (infinite loop and str instead of token object)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1080>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  9 17:40:39 2012
From: tracker at bugs.pypy.org (Simon Sapin)
Date: Fri, 09 Mar 2012 16:40:39 +0000
Subject: [pypy-issue] [issue1081] string.Formatter does not unescape double
	braces
In-Reply-To: <1331311239.95.0.375206161935.issue1081@bugs.pypy.org>
Message-ID: <1331311239.95.0.375206161935.issue1081@bugs.pypy.org>


New submission from Simon Sapin <simon.sapin at kozea.fr>:

The Formatter class of the string module is supposed to implement the same
functionality as the str.format method, but on PyPy 1.8.0 it (incorrectly) does
not unescape double braces.

Test case:

    import string; print string.Formatter().format("{{a}}", a=42)

Expected (CPython) output: {a}
PyPy output: {{a}}

Note that str.format('{{a}}', a=42) works fine in both CPython and PyPy.

The string.py stdlib module is identical in CPython 2.7.2 and PyPy 1.8.0, but it
calls the str._formatter_parser() method, where the results differ:

CPython: [('{', None, None, None), ('a}', None, None, None)]
PyPy: [('{{a}}', None, None, None)]

(This output is documented here:
http://docs.python.org/release/2.7/library/string.html#string.Formatter.parse )

----------
messages: 4056
nosy: SimonSapin, pypy-issue
priority: bug
status: unread
title: string.Formatter does not unescape double braces

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1081>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 10 22:50:41 2012
From: tracker at bugs.pypy.org (rednaks)
Date: Sat, 10 Mar 2012 21:50:41 +0000
Subject: [pypy-issue] [issue1082] [PATCH]HTMLParser decode issue
In-Reply-To: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>
Message-ID: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>


New submission from rednaks <salexandre.bm at gmail.com>:

while parsing a HTML code i got an decode Error :

but this issue can be fixed by replacing  the last string by s.decode() like in
the diff file.
I also tried to execute my script under python3.2 and it does not parsing any thing 

  File "/usr/lib/python2.7/HTMLParser.py", line 111, in feed
    self.goahead(0)
  File "/usr/lib/python2.7/HTMLParser.py", line 155, in goahead
    k = self.parse_starttag(i)
  File "/usr/lib/python2.7/HTMLParser.py", line 260, in parse_starttag
    attrvalue = self.unescape(attrvalue)
  File "/usr/lib/python2.7/HTMLParser.py", line 410, in unescape
    return re.sub(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));", replaceEntities, s)
  File "/usr/lib/python2.7/re.py", line 151, in sub
    return _compile(pattern, flags).sub(repl, string, count)
UnicodeDecodeError: 'ascii' codec can't decode byte 0x97 in position 1: ordinal
not in range(128)

----------
files: patch.txt
messages: 4057
nosy: pypy-issue, rednaks
priority: bug
status: unread
title: [PATCH]HTMLParser decode issue

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1082>
________________________________________
-------------- next part --------------
diff -ru1 old/HTMLParser.py new/HTMLParser.py
--- old/HTMLParser.py	2012-03-10 23:40:16.000000000 +0100
+++ new/HTMLParser.py	2012-03-10 23:46:36.000000000 +0100
@@ -409,2 +409,2 @@
 
-        return re.sub(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));", replaceEntities, s)
+        return re.sub(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));", replaceEntities, s.decode())

From tracker at bugs.pypy.org  Sun Mar 11 01:28:19 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 11 Mar 2012 00:28:19 +0000
Subject: [pypy-issue] [issue1082] [PATCH]HTMLParser decode issue
In-Reply-To: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>
Message-ID: <1331425699.3.0.488067291686.issue1082@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Hum, from the traceback it seems that you use CPython and not PyPy.
I this is the case, you should report the issue to the CPython bug tracker: 
bugs.python.org.
In any case, please provide a small test script that reproduces the problem.

----------
nosy: +afa
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1082>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 03:12:29 2012
From: tracker at bugs.pypy.org (Jean-Paul Calderone)
Date: Sun, 11 Mar 2012 02:12:29 +0000
Subject: [pypy-issue] [issue1083] Py_buffer-related formats not really
	supported by	cpyext's PyArg_ParseTuple
In-Reply-To: <1331431949.73.0.11762789711.issue1083@bugs.pypy.org>
Message-ID: <1331431949.73.0.11762789711.issue1083@bugs.pypy.org>


New submission from Jean-Paul Calderone <exarkun at twistedmatrix.com>:

Using the Py_buffer-filling formats with cpyext's PyArg_ParseTuple leads to 
random stack corruption.

Overall, I think the solution to this is to rewrite getargs.c in Python (with 
unit tests, of course).  The existing code, copied from CPython, is mega-gross, 
not nearly completely tested, and it would just be all around better if it were 
Python code instead of nasty C code.

A more complete description of the way Py_buffer-usage causes stack corruption 
is at http://bazaar.launchpad.net/~exarkun/pyopenssl/some-fixed-
segfaults/revision/173

----------
messages: 4059
nosy: calderone, pypy-issue
priority: bug
status: unread
title: Py_buffer-related formats not really supported by cpyext's PyArg_ParseTuple

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1083>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 03:19:20 2012
From: tracker at bugs.pypy.org (rednaks)
Date: Sun, 11 Mar 2012 02:19:20 +0000
Subject: [pypy-issue] [issue1082] [PATCH]HTMLParser decode issue
In-Reply-To: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>
Message-ID: <1331432360.32.0.628661928366.issue1082@bugs.pypy.org>


rednaks <salexandre.bm at gmail.com> added the comment:

Okey thank's afa, i'll report this on the CPython bug tracker !
does PyPy using standard libs of CPython ? did you ever get such issues ?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1082>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 11:12:23 2012
From: tracker at bugs.pypy.org (Stefan Behnel)
Date: Sun, 11 Mar 2012 10:12:23 +0000
Subject: [pypy-issue] [issue1084] Implement PyErr_GetExcInfo() and
	PyErr_SetExcInfo() in	cpyext
In-Reply-To: <1331460743.68.0.269550106582.issue1084@bugs.pypy.org>
Message-ID: <1331460743.68.0.269550106582.issue1084@bugs.pypy.org>


New submission from Stefan Behnel <stefan_ml at behnel.de>:

Provide a portable way to read and write sys.get_excinfo() from C code using
cpyext, to be used especially by Cython.

See:

http://bugs.python.org/issue14098

http://thread.gmane.org/gmane.comp.python.pypy/9507/focus=9644

----------
messages: 4061
nosy: pypy-issue, sbehnel
priority: feature
status: unread
title: Implement PyErr_GetExcInfo() and PyErr_SetExcInfo() in cpyext

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1084>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 16:08:06 2012
From: tracker at bugs.pypy.org (Stefan Behnel)
Date: Sun, 11 Mar 2012 15:08:06 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>


New submission from Stefan Behnel <stefan_ml at behnel.de>:

Calling PyString_AsString(Py_None) returns 0x7 for me instead of raising an
exception and returning NULL. This triggers a sure crash.

Example code is in Cython's "literal_lists.pyx" test.

----------
messages: 4062
nosy: pypy-issue, sbehnel
priority: bug
status: unread
title: cpyext: PyString_AsString() returns bogus results for non-bytes input

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 16:48:16 2012
From: tracker at bugs.pypy.org (Stefan Behnel)
Date: Sun, 11 Mar 2012 15:48:16 +0000
Subject: [pypy-issue] [issue1086] cpyext: calling PyIter_Next() on a
	"dictionaryiterator"	object crashes when the dict was modified
In-Reply-To: <1331480896.67.0.851470392018.issue1086@bugs.pypy.org>
Message-ID: <1331480896.67.0.851470392018.issue1086@bugs.pypy.org>


New submission from Stefan Behnel <stefan_ml at behnel.de>:

The iterdict.pyx test crashes in the PyPy adapted Cython.

https://github.com/scoder/cython/blob/master/tests/run/iterdict.pyx#L362

    cdef int count = 0
    for i in d.iterkeys():
        d[i+1] = 5
        count += 1
        if count > 5:
            break # safety

Readable parts of the GDB stack trace:

#2  0x0000000000f2afbc in PyIter_Next ()
#3  0x00007ffff4c6ad72 in __Pyx_dict_iter_next (dict_or_iter=0x7ffff7e45350,
orig_length=140737315342408, ppos=0x7fffffffb428, pkey=0x7fffffffb418,
pvalue=0x0, pitem=0x0, is_dict=0) at iterdict.c:5714
#4  0x00007ffff4c651f6 in __pyx_pf_8iterdict_40iterdict_change_size
(__pyx_self=0x0, __pyx_v_d=0x7ffff7e45320) at iterdict.c:4007
#5  0x00007ffff4c65079 in __pyx_pw_8iterdict_41iterdict_change_size
(__pyx_self=0x0, __pyx_v_d=0x7ffff7e45320) at iterdict.c:3929

----------
messages: 4063
nosy: pypy-issue, sbehnel
priority: bug
status: unread
title: cpyext: calling PyIter_Next() on a "dictionaryiterator" object crashes when the dict was modified

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1086>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 22:24:54 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 11 Mar 2012 21:24:54 +0000
Subject: [pypy-issue] [issue1082] [PATCH]HTMLParser decode issue
In-Reply-To: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>
Message-ID: <1331501094.62.0.0875917005395.issue1082@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Yes, PyPy uses the same stdlib as CPython.  Sometimes they are slightly modified to cope 
with pypy differences, but this is not the case here.

I've seen that you filed http://bugs.python.org/issue14251
Closing this one -- pypy will take whatever changes are made to the CPython stdlib.

----------
status: chatting -> wontfix

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1082>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 22:31:53 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 11 Mar 2012 21:31:53 +0000
Subject: [pypy-issue] [issue1081] string.Formatter does not unescape double
	braces
In-Reply-To: <1331311239.95.0.375206161935.issue1081@bugs.pypy.org>
Message-ID: <1331501513.36.0.946157943631.issue1081@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Fixed with 186605be5a0a, thanks for the report and the analysis!

----------
nosy: +afa
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1081>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 22:54:25 2012
From: tracker at bugs.pypy.org (rednaks)
Date: Sun, 11 Mar 2012 21:54:25 +0000
Subject: [pypy-issue] [issue1082] [PATCH]HTMLParser decode issue
In-Reply-To: <1331416241.28.0.454005848678.issue1082@bugs.pypy.org>
Message-ID: <1331502865.76.0.168389135715.issue1082@bugs.pypy.org>


rednaks <salexandre.bm at gmail.com> added the comment:

Okey thank's :)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1082>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 23:06:41 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sun, 11 Mar 2012 22:06:41 +0000
Subject: [pypy-issue] [issue1083] Py_buffer-related formats not really
	supported by	cpyext's PyArg_ParseTuple
In-Reply-To: <1331431949.73.0.11762789711.issue1083@bugs.pypy.org>
Message-ID: <1331503601.06.0.368598196727.issue1083@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

I fear it won't be easy: C vararg functions are not friendly with any kind of 
wrapper, this is why we copied the C version from CPython.

----------
nosy: +afa
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1083>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 11 23:54:18 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sun, 11 Mar 2012 22:54:18 +0000
Subject: [pypy-issue] [issue1083] Py_buffer-related formats not really
	supported by	cpyext's PyArg_ParseTuple
In-Reply-To: <1331431949.73.0.11762789711.issue1083@bugs.pypy.org>
Message-ID: <1331506458.25.0.36375424332.issue1083@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

I hate to say that but maybe we should fix C? Like copy the stringbuffer if we 
exit correctly or so?

----------
nosy: +fijal

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1083>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 14:13:11 2012
From: tracker at bugs.pypy.org (Tobias Oberstein)
Date: Mon, 12 Mar 2012 13:13:11 +0000
Subject: [pypy-issue] [issue878] too many open files
In-Reply-To: <1316725510.98.0.71219979546.issue878@bugs.pypy.org>
Message-ID: <1331557991.45.0.754710267742.issue878@bugs.pypy.org>


Tobias Oberstein <tobias.oberstein at gmail.com> added the comment:

For people running into this issue, here is the recipe for Ubuntu:

1.
sudo vi /etc/pam.d/su

=> uncomment the line:

session    required   pam_limits.so

2.
sudo vi /etc/security/limits.conf

=> add the following 2 lines at the end of file:

*       soft  nofile   65535                                                     
*       hard  nofile   65535

3.
reboot.

4.
check:

pypy =>
>>>> import resource
>>>> resource.getrlimit(resource.RLIMIT_NOFILE)
(65535L, 65535L)

===

Caveat: This will set limit to 64k for _all_ users. i.e. don't do that on a
production box.

----------
nosy: +oberstet

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue878>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 18:13:03 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 17:13:03 +0000
Subject: [pypy-issue] [issue1079] str.join + setdefaultencoding problem
In-Reply-To: <1331220655.08.0.829278773481.issue1079@bugs.pypy.org>
Message-ID: <1331572383.28.0.957232530577.issue1079@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Confirmed the bug, in pypy-1.8 on x86 Linux

----------
nosy: +dripton
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1079>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 18:21:54 2012
From: tracker at bugs.pypy.org (Alex Gaynor)
Date: Mon, 12 Mar 2012 17:21:54 +0000
Subject: [pypy-issue] [issue1078] comparison between dictview and frozendict
	is wrong
In-Reply-To: <1330994388.53.0.80561785434.issue1078@bugs.pypy.org>
Message-ID: <1331572914.74.0.680285296052.issue1078@bugs.pypy.org>


Alex Gaynor <alex.gaynor at gmail.com> added the comment:

Fixed in bf600fa56140

----------
nosy: +agaynor
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1078>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 18:31:48 2012
From: tracker at bugs.pypy.org (Alex Gaynor)
Date: Mon, 12 Mar 2012 17:31:48 +0000
Subject: [pypy-issue] [issue1079] str.join + setdefaultencoding problem
In-Reply-To: <1331220655.08.0.829278773481.issue1079@bugs.pypy.org>
Message-ID: <1331573508.96.0.233548348353.issue1079@bugs.pypy.org>


Alex Gaynor <alex.gaynor at gmail.com> added the comment:

Cause seems to be 
https://bitbucket.org/pypy/pypy/src/default/pypy/objspace/std/stringobject.py#cl-
58 (note the XXX)

----------
nosy: +agaynor

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1079>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 21:14:35 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 20:14:35 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>


New submission from Andrew McNabb <amcnabb at mcnabbs.org>:

In PyPy 1.8.0, I came across a race condition when calling getattr. This race 
condition appears to only occur pypy; I cannot reproduce it in cPython 2.7. 
Specifically, in multithreaded code, I occasionally encounter the following 
traceback:

2012-03-12 13:56:37,004: ERROR: Traceback: Traceback (most recent call last):    
  File "/home/amcnabb/python/mrs/http.py", line 157, in _marshaled_dispatch        
    response = dispatch_method(method, params)                                     
  File "/home/amcnabb/python/mrs/http.py", line 128, in _dispatch                  
    return self.server._dispatch(method, params, host)                             
  File "/home/amcnabb/python/mrs/http.py", line 177, in _dispatch                  
    func = getattr(self.instance, 'xmlrpc_' + method)                              
LookupError: unknown encoding: ascii

I can work around this by adding the following lines at the module level:

import codecs
codecs.lookup('ascii')

----------
messages: 4073
nosy: amcnabb, pypy-issue
priority: bug
status: unread
title: Race condition when calling getattr

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 21:54:17 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 20:54:17 +0000
Subject: [pypy-issue] [issue1079] str.join + setdefaultencoding problem
In-Reply-To: <1331220655.08.0.829278773481.issue1079@bugs.pypy.org>
Message-ID: <1331585657.45.0.687937816965.issue1079@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Fix in
https://bitbucket.org/pypy/pypy/pull-request/35/fix-for-issue-1079-strjoin-unicode-utf-8

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1079>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 22:01:09 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 21:01:09 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331586069.03.0.0890534329071.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

amcnabb, can you submit a minimal program that triggers the bug?

----------
nosy: +dripton
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 22:21:28 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 21:21:28 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331587288.08.0.354908199462.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

My attempt to reproduce the bug is attached.  It does not exhibit the bug, using
PyPy 1.8 on x86 Linux.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 22:53:07 2012
From: tracker at bugs.pypy.org (mikefc)
Date: Mon, 12 Mar 2012 21:53:07 +0000
Subject: [pypy-issue] [issue878] too many open files
In-Reply-To: <1316725510.98.0.71219979546.issue878@bugs.pypy.org>
Message-ID: <1331589187.58.0.897106703296.issue878@bugs.pypy.org>


mikefc <coolbutuseless at gmail.com> added the comment:

I found that changing everything to the "with open() as f" syntax in pypy (rather 
than relying on cpython's refcounting to close a file) resolved this issue for 
me.

----------
nosy: +mikefc

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue878>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:13:41 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 22:13:41 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331590421.77.0.103760994318.issue1087@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

I was able to adapt boom.py to produce the bug. I am attaching the script which 
triggers the bug more than 50% of the time.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:19:14 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 22:19:14 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331590754.78.0.509630696218.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

boomboom.py won't crash for me, either.  (32-bit Linux, PyPy 1.8.)  I suspect we
have an OS-specific or PyPy-version specific problem.  What OS are you using,
and what version of PyPy?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:25:10 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 22:25:10 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331591110.87.0.842150481505.issue1087@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

I'm on 64-bit Fedora Linux.

uname -a:
Linux sage.aml.cs.byu.edu 3.1.9-1.fc16.x86_64 #1 SMP Fri Jan 13 16:37:42 UTC 2012 
x86_64 x86_64 x86_64 GNU/Linux

pypy --version:
Python 2.7.2 (2346207d99463f299f09f3e151c9d5fa9158f71b, Feb 10 2012, 16:19:53)
[PyPy 1.8.0 with GCC 4.7.0]

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:34:43 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 22:34:43 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331591683.6.0.902164753878.issue1087@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

Another possibility might be a difference in PYTHONPATH. Mine is fairly large: 
PYTHONPATH="$HOME:/usr/lib/python2.7/site-packages". If the race condition has 
something to do with something like how long it takes to import the codecs 
module, then having a larger PYTHONPATH might make it easier to reproduce.

Just a thought.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:38:49 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 22:38:49 +0000
Subject: [pypy-issue] [issue1079] str.join + setdefaultencoding problem
In-Reply-To: <1331220655.08.0.829278773481.issue1079@bugs.pypy.org>
Message-ID: <1331591929.56.0.568167003648.issue1079@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Fixed in f07d9c4d0671

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1079>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 12 23:55:50 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 22:55:50 +0000
Subject: [pypy-issue] [issue1047] CVE-2012-0845 SimpleXMLRPCServer DoS
	(excessive CPU	usage) via malformed XML-RPC / HTTP POST request
In-Reply-To: <1329152174.59.0.255188204929.issue1047@bugs.pypy.org>
Message-ID: <1331592950.78.0.967897675852.issue1047@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Fix at
https://bitbucket.org/pypy/pypy/pull-request/39/update-simplexmlrpcserverpy-to-latest-from

----------
assignedto:  -> dripton
nosy: +dripton

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1047>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 00:02:50 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 23:02:50 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331593370.2.0.826128534807.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

If your PYTHONPATH includes CPython modules that include C code, that's likely
the problem.  Try running with a clear PYTHONPATH.  (If you need to install
third-party Python code, install it inside PyPy using pip, or make a virtualenv
using PyPy.)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 00:13:58 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 23:13:58 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331594038.12.0.698833625525.issue1087@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

On Fedora, the /usr/lib/python2.7 only contains pure-Python modules.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 00:17:55 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Mon, 12 Mar 2012 23:17:55 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331594275.64.0.324502183201.issue1087@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

On Fedora, the /usr/lib/python2.7/site-packages directory only contains pure-
Python modules. And only one or two packages are actually being imported from 
that directory.

Using pip isn't ideal because it doesn't play well with yum. Virtualenv is an 
option, but it really shouldn't be necessary here. In any case, there aren't any 
C modules on the PYTHONPATH, so that can't explain the bug.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 00:39:32 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Mon, 12 Mar 2012 23:39:32 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331595572.12.0.162159758752.issue1087@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

boomboom.py also fails for me almost every time.

It's indeed a race condition: when the first thread is executing code in 
encodings/ascii.py, the second thread starts to "import encoding.ascii", finds it in 
sys.modules, and returns it... even when the module is not complete.

Because the module is incomplete, it is skipped (search for "mod.getregentry" in 
encodings/__init__.py), and codec lookup fails.

This is a bug in pypy: the check in sys.modules should be done only with the import lock 
held. This will impose a speed penalty though...

----------
nosy: +afa

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 00:50:19 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Mon, 12 Mar 2012 23:50:19 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331596219.01.0.270738528353.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

It makes sense that failing to use the import lock results in a race.  But it's
interesting that I can't get this to fail at all on 32-bit Linux in dozens of
attempts, while you can both see it usually fail on 64-bit Linux.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 01:00:45 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Tue, 13 Mar 2012 00:00:45 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331596845.33.0.435751559213.issue1087@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

It's possible that encodings.ascii is already imported at startup in your 32-bit 
environment (because of locales, etc.).
To be sure you can try to print sorted(sys.modules) before your threaded code.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 02:05:12 2012
From: tracker at bugs.pypy.org (Ken Watford)
Date: Tue, 13 Mar 2012 01:05:12 +0000
Subject: [pypy-issue] [issue1088] numpypy array scalars do not hash nicely
In-Reply-To: <1331600712.72.0.0717696928313.issue1088@bugs.pypy.org>
Message-ID: <1331600712.72.0.0717696928313.issue1088@bugs.pypy.org>


New submission from Ken Watford <kwatford at gmail.com>:

>>>> from numpypy import int8
>>>> assert hash(int8(0)) == hash(int8(0))
Traceback (most recent call last):
  File "<console>", line 1, in <module>
AssertionError

Array scalars in real numpy normally hash to the same value as the equivalent 
python scalars. Especially since their reprs are the same. While debugging some 
code of mine I came across something like this:

>>>> print {0: 'wtf', int8(0): 'foo', int8(0): 'bar'}
{0: 'wtf', 0: 'foo', 0: 'bar'}

Except with several thousand pairs. It was slightly surreal until I thought about 
where the supposed zeros were coming from.

----------
messages: 4090
nosy: kw, pypy-issue
priority: bug
status: unread
title: numpypy array scalars do not hash nicely

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1088>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 03:18:29 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Tue, 13 Mar 2012 02:18:29 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331605109.59.0.568889969086.issue1087@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

I hope it's fixed in
https://bitbucket.org/pypy/pypy/pull-request/44/hold-import-lock-longer-to-fix-an-import

I haven't translated this code yet (I only have a 1 GB Netbook at PyCon) so I
can't say how bad the performance impact is.  Most code doesn't import inside
loops so probably not too bad.

----------
assignedto:  -> dripton

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 10:07:21 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Tue, 13 Mar 2012 09:07:21 +0000
Subject: [pypy-issue] [issue1087] Race condition when calling getattr
In-Reply-To: <1331583275.77.0.620668848256.issue1087@bugs.pypy.org>
Message-ID: <1331629641.41.0.930500435294.issue1087@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

> Most code doesn't import inside loops so probably not too bad.

Unfortunately this is not true, see the past discussion on IRC about the Django 
benchmark:
http://tismerysoft.de/pypy/irc-logs/pypy/pypy.2011-04-07.log.html
(search for "import lock", and "escape")

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1087>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 18:22:10 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Tue, 13 Mar 2012 17:22:10 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <1331659330.91.0.127603854988.issue1085@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

My test shows that PyString_AsString(Py_None) raises TypeError as expected.
See
https://bitbucket.org/pypy/pypy/pull-request/50/issue1085-add-a-test-to-show-that

----------
nosy: +dripton
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 18:45:38 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Tue, 13 Mar 2012 17:45:38 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <1331660738.47.0.32087022699.issue1085@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Amaury told me in IRC that it segfaults on his machine, so it appears this
depends on OS, PyPy version, phase of the moon, etc.  FWIW I can't make the test
fail on x86 Ubuntu with PyPy 1.8.  When I get to a box where it segfaults, I'll
add a fix.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 13 18:42:02 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Tue, 13 Mar 2012 17:42:02 +0000
Subject: [pypy-issue] [issue1045] Crash of the JIT
In-Reply-To: <1329140443.89.0.137027286035.issue1045@bugs.pypy.org>
Message-ID: <1331660522.8.0.464666333034.issue1045@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Everyone seems to agree it's fixed, so closing.

----------
nosy: +dripton
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1045>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 14 07:08:52 2012
From: tracker at bugs.pypy.org (Stefan Behnel)
Date: Wed, 14 Mar 2012 06:08:52 +0000
Subject: [pypy-issue] [issue1084] Implement PyErr_GetExcInfo() and
	PyErr_SetExcInfo() in	cpyext
In-Reply-To: <1331460743.68.0.269550106582.issue1084@bugs.pypy.org>
Message-ID: <1331705332.66.0.385819070498.issue1084@bugs.pypy.org>


Stefan Behnel <stefan_ml at behnel.de> added the comment:

I should note that this is currently a blocker for getting Cython to work on PyPy.

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1084>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 14 18:38:24 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Wed, 14 Mar 2012 17:38:24 +0000
Subject: [pypy-issue] [issue1047] CVE-2012-0845 SimpleXMLRPCServer DoS
	(excessive CPU	usage) via malformed XML-RPC / HTTP POST request
In-Reply-To: <1329152174.59.0.255188204929.issue1047@bugs.pypy.org>
Message-ID: <1331746703.98.0.309760659083.issue1047@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Fix was merged in 452b9e9ac4b8

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1047>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 14 21:21:18 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Wed, 14 Mar 2012 20:21:18 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1331756478.4.0.361650574746.issue1069@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Sorry, my previous comment is completely wrong: the GIL is not released in C code, except 
by the functions explicitly designed for this.

Instead, the solution is for PyEval_SaveThread() to return the current thread state, and 
have PyThreadState_GET() do the right thing depending on the GIL being current or not.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 15 15:19:49 2012
From: tracker at bugs.pypy.org (montag451)
Date: Thu, 15 Mar 2012 14:19:49 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>


New submission from montag451 <montag451 at laposte.net>:

When I try to bind a packet socket using ETH_P_ALL as the proto, I get an
"unknown address family" error. The code attached to this issue work with
CPython but doesn't work with PyPy.

----------
nosy: +montag451
priority: bug -> wish
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 15 16:19:43 2012
From: tracker at bugs.pypy.org (Richard Barrell)
Date: Thu, 15 Mar 2012 15:19:43 +0000
Subject: [pypy-issue] [issue1091] list(xrange) is slower in PyPy 1.8 than in
	CPython 2.6.5
In-Reply-To: <1331824783.12.0.608099911457.issue1091@bugs.pypy.org>
Message-ID: <1331824783.12.0.608099911457.issue1091@bugs.pypy.org>


New submission from Richard Barrell <rchrd at brrll.co.uk>:

I'm using the Linux 64-bit binary PyPy 1.8 from here: https://bitbucket.org/pypy/pypy/downloads/pypy-
1.8-linux64.tar.bz2
and Python 2.6.5 from the repositories on Ubuntu 10.04. The machine that I'm running on is a Core 2 Duo 
laptop with 4GB of RAM, running at 2.10GHz.

richyb at figgy:~$ python
Python 2.6.5 (r265:79063, Apr 16 2010, 13:57:41)
[GCC 4.4.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import time; t1 = time.time(); _ = list(xrange(10000000)); print time.time() - t1
0.385615825653
>>>

richyb at figgy:~$ pypy
Python 2.7.2 (0e28b379d8b3, Feb 09 2012, 19:41:03)
[PyPy 1.8.0 with GCC 4.4.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``PyPy is Shirt Origami research''
>>>> import time; t1 = time.time(); _ = list(xrange(10000000)); print time.time() - t1
2.09597301483
>>>>

----------
messages: 4101
nosy: RichyB, pypy-issue
priority: performance bug
status: unread
title: list(xrange) is slower in PyPy 1.8 than in CPython 2.6.5

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1091>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 15 19:45:37 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Thu, 15 Mar 2012 18:45:37 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1331837137.83.0.637738254291.issue1090@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

In pypy/rlib/rsocket.py, the class PacketAddress is missing a from_object() 
method.  See 
http://hg.python.org/cpython/file/v2.7.2/Modules/socketmodule.c#l1422 for the 
equivalent implementation in CPython.

----------
nosy: +afa

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 15 21:44:50 2012
From: tracker at bugs.pypy.org (Jean-Paul Calderone)
Date: Thu, 15 Mar 2012 20:44:50 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1331844290.29.0.690712847464.issue1069@bugs.pypy.org>


Jean-Paul Calderone <exarkun at twistedmatrix.com> added the comment:

I put some PyThreadState fixes in the branch non-null-threadstate.  Anyone want
to take a look and see if they're sane?  They make pyOpenSSL work.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 02:46:16 2012
From: tracker at bugs.pypy.org (behackett)
Date: Fri, 16 Mar 2012 01:46:16 +0000
Subject: [pypy-issue] [issue1092] 1.8.0 datetime.timedelta issue
In-Reply-To: <1331862376.54.0.35663691756.issue1092@bugs.pypy.org>
Message-ID: <1331862376.54.0.35663691756.issue1092@bugs.pypy.org>


New submission from behackett <bernie at 10gen.com>:

pypy 1.7.0:

[PyPy 1.7.0 with GCC 4.5.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``"anyway, python is a vast
language" (fijal)''
>>>> import datetime
>>>> d = datetime.datetime.utcnow()
>>>> d = d - datetime.timedelta(microseconds=d.microsecond)
>>>> d
datetime.datetime(2012, 3, 16, 1, 43, 51)

pypy 1.8.0:

[PyPy 1.8.0 with GCC 4.5.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``we have much more interpreters
than users''
>>>> import datetime
>>>> d = datetime.datetime.utcnow()
>>>> d = d - datetime.timedelta(microseconds=d.microsecond)
Traceback (most recent call last):
  File "<console>", line 1, in <module>
  File "/usr/lib64/pypy1.8/lib_pypy/datetime.py", line 1859, in __sub__
    return self + -other
  File "/usr/lib64/pypy1.8/lib_pypy/datetime.py", line 1850, in __add__
    t.microsecond, tzinfo=self._tzinfo)
  File "/usr/lib64/pypy1.8/lib_pypy/datetime.py", line 1451, in __new__
    _check_time_fields(hour, minute, second, microsecond)
  File "/usr/lib64/pypy1.8/lib_pypy/datetime.py", line 286, in _check_time_fields
    raise TypeError('int expected')
TypeError: int expected

----------
messages: 4104
nosy: behackett, pypy-issue
priority: bug
status: unread
title: 1.8.0 datetime.timedelta issue

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1092>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 08:23:19 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Fri, 16 Mar 2012 07:23:19 +0000
Subject: [pypy-issue] [issue1092] 1.8.0 datetime.timedelta issue
In-Reply-To: <1331862376.54.0.35663691756.issue1092@bugs.pypy.org>
Message-ID: <1331882599.23.0.426743969672.issue1092@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

This was already fixed in 8a9843bb99ad.
This part of the diff is relevant, if you want to modify your copy of datetime.py:

--- a/lib_pypy/datetime.py	Sun Jan 22 10:33:03 2012 +0100
+++ b/lib_pypy/datetime.py	Fri Feb 10 01:19:56 2012 +0100
@@ -1520,7 +1520,7 @@
     def utcfromtimestamp(cls, t):
         "Construct a UTC datetime from a POSIX timestamp (like time.time())."
         t, frac = divmod(t, 1.0)
-        us = round(frac * 1e6)
+        us = int(round(frac * 1e6))
 
         # If timestamp is less than one microsecond smaller than a
         # full second, us can be rounded up to 1000000.  In this case,

----------
nosy: +afa
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1092>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 09:49:54 2012
From: tracker at bugs.pypy.org (montag451)
Date: Fri, 16 Mar 2012 08:49:54 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1331887794.32.0.826451061305.issue1090@bugs.pypy.org>


montag451 <montag451 at laposte.net> added the comment:

So it's a bug ?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 11:05:31 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Fri, 16 Mar 2012 10:05:31 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1331892331.99.0.544448462126.issue1090@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

yes, an unimplemented feature :)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 14:10:12 2012
From: tracker at bugs.pypy.org (montag451)
Date: Fri, 16 Mar 2012 13:10:12 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1331903412.56.0.176180130212.issue1090@bugs.pypy.org>


montag451 <montag451 at laposte.net> added the comment:

I changed the priority to "bug". Thanks for your replies !

----------
priority: wish -> bug

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 18:49:47 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Fri, 16 Mar 2012 17:49:47 +0000
Subject: [pypy-issue] [issue912] ZipFile.read() leaks file handles
In-Reply-To: <1318930736.76.0.381437213231.issue912@bugs.pypy.org>
Message-ID: <1331920187.15.0.437311326258.issue912@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

Shouldn't the file object eventually get garbage collected after the ZipFile is 
deleted? I ran into a similar problem in my own code where file descriptors were 
being leaked under pypy 1.8. Yes, I have now fixed the missing close on the file 
object, but I'm still surprised that files weren't eventually garbage collected 
and the file descriptors closed. Is this garbage collection behavior a bug, or 
am I misunderstanding something?

----------
nosy: +amcnabb

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue912>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 16 21:40:30 2012
From: tracker at bugs.pypy.org (Jean-Paul Calderone)
Date: Fri, 16 Mar 2012 20:40:30 +0000
Subject: [pypy-issue] [issue1069] PyThread_set_key_value and
	PyThread_delete_key_value	missing from cpyext support
In-Reply-To: <1330193098.61.0.219372830064.issue1069@bugs.pypy.org>
Message-ID: <1331930430.33.0.076045858347.issue1069@bugs.pypy.org>


Jean-Paul Calderone <exarkun at twistedmatrix.com> added the comment:

Merged that branch.  pyOpenSSL should be happy now.

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1069>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 18 07:49:07 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sun, 18 Mar 2012 06:49:07 +0000
Subject: [pypy-issue] [issue1091] list(xrange) is slower in PyPy 1.8 than in
	CPython 2.6.5
In-Reply-To: <1331824783.12.0.608099911457.issue1091@bugs.pypy.org>
Message-ID: <1332053347.62.0.00981888947574.issue1091@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

This is very likely due to lack of sizehint. Maybe now it's a good time to 
introduce it.

----------
nosy: +fijal
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1091>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 18 07:53:18 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sun, 18 Mar 2012 06:53:18 +0000
Subject: [pypy-issue] [issue1090] Can't bind packet socket
In-Reply-To: <1331821189.63.0.719460787992.issue1090@bugs.pypy.org>
Message-ID: <1332053598.3.0.184414381229.issue1090@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

This is maybe worth noting that this is a nice "easy" feature for a newcomer. We 
don't however have a way to agregate such things.

----------
nosy: +fijal

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1090>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 20 04:26:59 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Tue, 20 Mar 2012 03:26:59 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <1332214018.12.0.463909805921.issue1085@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

Still can't make this test segfault.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 20 07:34:31 2012
From: tracker at bugs.pypy.org (raymin)
Date: Tue, 20 Mar 2012 06:34:31 +0000
Subject: [pypy-issue] [issue1093] PyPy list element del insert to slow
In-Reply-To: <1332225271.38.0.963212152982.issue1093@bugs.pypy.org>
Message-ID: <1332225271.38.0.963212152982.issue1093@bugs.pypy.org>


New submission from raymin <guo.raymin at gmail.com>:

# testList.py
L = []

i = 0
while i < 100000:
    L.append('k%d' % i)
    i += 1

i = 0
while i < 100000:
    if i & 1:
        k = L[0]
        del L[0]
        L.insert(i, k)
    else:
        k = L[i]
        del L[i]
        L.append(k)
    i += 1

print i, len(L)

time python testList.py   : 0m6.710s
time pypy testList.py     : 0m41.702s

----------
messages: 4114
nosy: pypy-issue, raymin
priority: wish
status: unread
title: PyPy list element del insert to slow

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1093>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 20 07:38:27 2012
From: tracker at bugs.pypy.org (raymin)
Date: Tue, 20 Mar 2012 06:38:27 +0000
Subject: [pypy-issue] [issue1093] PyPy list element del insert too slow
In-Reply-To: <1332225507.83.0.0822621714333.issue1093@bugs.pypy.org>
Message-ID: <1332225507.83.0.0822621714333.issue1093@bugs.pypy.org>


New submission from raymin <guo.raymin at gmail.com>:

# testList.py
L = []

i = 0
while i < 100000:
    L.append('k%d' % i)
    i += 1

i = 0
while i < 100000:
    if i & 1:
        k = L[0]
        del L[0]
        L.insert(i, k)
    else:
        k = L[i]
        del L[i]
        L.append(k)
    i += 1

print i, len(L)

time python testList.py   : 0m6.710s
time pypy testList.py     : 0m41.702s

----------
title: PyPy list element del insert to slow -> PyPy list element del insert too slow

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1093>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 20 08:05:02 2012
From: tracker at bugs.pypy.org (Stefan Behnel)
Date: Tue, 20 Mar 2012 07:05:02 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <4F682AE9.3070608@behnel.de>


Stefan Behnel <stefan_ml at behnel.de> added the comment:

I'm on Linux 64bit, in case that helps.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 20 15:36:19 2012
From: tracker at bugs.pypy.org (Simon Sapin)
Date: Tue, 20 Mar 2012 14:36:19 +0000
Subject: [pypy-issue] [issue1094] Non-BPM unicode literals decoded as
	surrogate pairs
In-Reply-To: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>
Message-ID: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>


New submission from Simon Sapin <simon.sapin at kozea.fr>:

PyPy decodes each non-BMP character (ie. U+10000 and beyond) in unicode string
literals as a surrogate pair (as used in UTF-16) instead of a single character.

Test case:

# coding: utf8
string1 = u'\U00010083'
string2 = u'?'  # should be the same as string1
codepoints1 = [ord(c) for c in string1]
codepoints2 = [ord(c) for c in string2.encode('utf16').decode('utf16')]
codepoints3 = [ord(c) for c in string2]
assert codepoints1 == [0x10083], codepoints1
assert codepoints2 == [0x10083], codepoints2
assert codepoints3 == [0x10083], codepoints3

This test pass on CPython compiled with "wide unicode". On PyPy 1.8.0,
codepoints3 is [0xd800, 0xdc83] (codepoints1 and codepoints2 are correct)

----------
messages: 4117
nosy: SimonSapin, pypy-issue
priority: bug
status: unread
title: Non-BPM unicode literals decoded as surrogate pairs

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1094>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 15 06:29:00 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Thu, 15 Mar 2012 05:29:00 +0000
Subject: [pypy-issue] [issue1089] logparser.py crashes with nesting error
In-Reply-To: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>
Message-ID: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>


New submission from Andrew McNabb <amcnabb at mcnabbs.org>:

I encountered the following error after running pypy 1.8 with PYPYLOG=pypylog55:

../pypy/pypy/tool/logparser.py print-summary pypylog55 -
93% ...................................................................       
Traceback (most recent call last):
  File "../pypy/pypy/tool/logparser.py", line 437, in <module>
    log = parse_log_file(args[0])
  File "../pypy/pypy/tool/logparser.py", line 26, in parse_log_file
    return parse_log(lines, verbose=verbose)
  File "../pypy/pypy/tool/logparser.py", line 74, in parse_log
    record(match.group(2), time=int(match.group(1), 16))
  File "/aml/home/amcnabb/c/pypy/pypy/rlib/debug.py", line 42, in debug_stop
    (self[i][1], category))
AssertionError: nesting error: starts with 'gc-minor' but stops with 'gc-set-
nursery-size'

I'm trying to think of anything unusual about my situation. The program shut down 
cleanly, so that rules out pypy crashing mid-write. The only slightly unusual 
situation I can think of is that the program creates a second process with 
multiprocessing.Process. One of the processes has two or three threads, including 
a daemon thread.

----------
files: pypylog55
messages: 4099
nosy: amcnabb, pypy-issue
priority: bug
status: unread
title: logparser.py crashes with nesting error

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1089>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 21 09:24:40 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Wed, 21 Mar 2012 08:24:40 +0000
Subject: [pypy-issue] [issue1089] logparser.py crashes with nesting error
In-Reply-To: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>
Message-ID: <1332318280.42.0.337281063183.issue1089@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

Uh. So indeed what happened (I think) is that the child process inherited parent 
env variable and started writing to the same file. Maybe we should automatically 
add PID to the log filename?

----------
nosy: +fijal
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1089>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 21 16:50:47 2012
From: tracker at bugs.pypy.org (Andrew McNabb)
Date: Wed, 21 Mar 2012 15:50:47 +0000
Subject: [pypy-issue] [issue1089] logparser.py crashes with nesting error
In-Reply-To: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>
Message-ID: <1332345047.44.0.102368541936.issue1089@bugs.pypy.org>


Andrew McNabb <amcnabb at mcnabbs.org> added the comment:

Adding the PID to the filename sounds like an easy fix. The only drawback would 
be that a directory could get really big after a few runs if it's not cleaned up, 
but I don't think it would be a big deal. Any alternative solutions I can think 
of would be much more complex.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1089>
________________________________________

From amauryfa at gmail.com  Wed Mar 21 16:55:35 2012
From: amauryfa at gmail.com (Amaury Forgeot d'Arc)
Date: Wed, 21 Mar 2012 16:55:35 +0100
Subject: [pypy-issue] [issue1089] logparser.py crashes with nesting error
In-Reply-To: <1332345047.44.0.102368541936.issue1089@bugs.pypy.org>
References: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>
	<1332345047.44.0.102368541936.issue1089@bugs.pypy.org>
Message-ID: <CAGmFidZRHidFtrwF1fjkvbUdhU-Mt-vuOoJ_GjDZO4MbbFuq3w@mail.gmail.com>

2012/3/21 Andrew McNabb <tracker at bugs.pypy.org>:
>
> Andrew McNabb <amcnabb at mcnabbs.org> added the comment:
>
> Adding the PID to the filename sounds like an easy fix. The only drawback would
> be that a directory could get really big after a few runs if it's not cleaned up,
> but I don't think it would be a big deal. Any alternative solutions I can think
> of would be much more complex.

Make it optional, then?
   PYPYLOG=pypylog55-PID:
and have pypy replace "PID" with the actual one.

-- 
Amaury Forgeot d'Arc

From tracker at bugs.pypy.org  Wed Mar 21 18:04:31 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Wed, 21 Mar 2012 17:04:31 +0000
Subject: [pypy-issue] [issue1089] logparser.py crashes with nesting error
In-Reply-To: <1331789340.11.0.0530114615506.issue1089@bugs.pypy.org>
Message-ID: <1332349471.88.0.779243450938.issue1089@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

Optional would still end up in obscure errors, unfortunately.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1089>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 18:12:59 2012
From: tracker at bugs.pypy.org (Mike Auty)
Date: Thu, 22 Mar 2012 17:12:59 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>


New submission from Mike Auty <mike.auty at gmail.com>:

On both 1.8, and trunk (53900:4cb3a7f01438), I receive the following error:

[translation:ERROR]     Traceback (most recent call last):
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 2013, in <module>
[translation:ERROR]         tracker.process(f, g, filename=fn)
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 1906, in process
[translation:ERROR]         tracker = parser.process_function(lines, filename)
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 1421, in process_function
[translation:ERROR]         table = tracker.computegcmaptable(self.verbose)
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 52, in computegcmaptable
[translation:ERROR]         self.parse_instructions()
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 204, in parse_instructions
[translation:ERROR]         self.find_missing_visit_method(opname)
[translation:ERROR]       File
"/var/tmp/portage/dev-python/pypy-9999/work/pypy-9999/pypy/translator/c/gcc/trackgcroot.py",
line 234, in find_missing_visit_method
[translation:ERROR]         raise UnrecognizedOperation(opname)
[translation:ERROR]     __main__.UnrecognizedOperation: vmulsd

I've attached the complete build log, and will include some additional
information about my build system that may be of use at the end of this message.

If there's any further information you need from me, or further tests you'd like
me to run, please just let me know...  5:)


Portage 2.2.0_alpha90 (default/linux/amd64/10.0, gcc-4.5.3, glibc-2.14.1-r2,
3.3.0 x86_64)
=================================================================
System uname:
Linux-3.3.0-x86_64-Intel-R-_Core-TM-_i7-2600K_CPU_ at _3.40GHz-with-gentoo-2.1
Timestamp of tree: Fri, 16 Mar 2012 08:00:01 +0000
ccache version 3.1.7 [enabled]
app-shells/bash:          4.2_p24
dev-java/java-config:     2.1.11-r3
dev-lang/python:          2.7.2-r3, 3.2.2
dev-util/ccache:          3.1.7
dev-util/cmake:           2.8.7-r5
dev-util/pkgconfig:       0.26
sys-apps/baselayout:      2.1
sys-apps/openrc:          0.9.9.3
sys-apps/sandbox:         2.5
sys-devel/autoconf:       2.13, 2.68
sys-devel/automake:       1.9.6-r3, 1.10.3, 1.11.3
sys-devel/binutils:       2.22-r1
sys-devel/gcc:            4.5.3-r2
sys-devel/gcc-config:     1.5.1-r1
sys-devel/libtool:        2.4.2
sys-devel/make:           3.82-r3
sys-kernel/linux-headers: 3.2-r1 (virtual/os-headers)
sys-libs/glibc:           2.14.1-r2
Repositories: gentoo autogenerated ikelos uncon personal
Installed sets: 
ACCEPT_KEYWORDS="amd64 ~amd64"
ACCEPT_LICENSE="*"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=native -O2 -pipe -ggdb"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/gnupg/qualified.txt"
CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d
/etc/env.d/java/ /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release
/etc/revdep-rebuild /etc/sandbox.d /etc/terminfo"
CXXFLAGS="-march=native -O2 -pipe -ggdb"
DISTDIR="/usr/portage/distfiles"
EMERGE_DEFAULT_OPTS="--quiet-build n"
FEATURES="assume-digests binpkg-logs ccache collision-protect distlocks
ebuild-locks fixlafiles news parallel-fetch protect-owned sandbox sfperms sign
splitdebug strict unknown-features-warn unmerge-logs unmerge-orphans userfetch
userpriv usersandbox"
FFLAGS=""
GENTOO_MIRRORS="http://distfiles.gentoo.org"
LANG="en_GB.UTF-8"
LDFLAGS="-Wl,-O1 -Wl,--as-needed"
LINGUAS="en en_GB"
MAKEOPTS="-j9"
PKGDIR="/usr/portage/packages"
PORTAGE_CONFIGROOT="/"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress
--force --whole-file --delete --stats --human-readable --timeout=180
--exclude=/distfiles --exclude=/local --exclude=/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/overlays/autogenerated /usr/local/overlays/ikelos
/usr/local/overlays/uncon /usr/local/overlays/personal"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="X a52 aac acl amd64 apng bash-completion berkdb bluetooth bzip2 cairo cdda
cli clutter consolekit cracklib crypt cups cxx dbus divx djvu dri exif
fontconfig fortran fts gcj gdbm gdu git gnome gnome-keyring
gnome-online-accounts gpm graphviz gstreamer gtk gtk3 gudev hwdb iconv ipv6 irc
jabber jpeg libffi libnotify mmx modules mp3 msn mudflap multilib nautilus
ncurses networkmanager nls nptl nptlonly opengl openmp pam pcre png policykit
pppd pulseaudio python readline resolvconf session sip sqlite sse sse2 ssl
subversion svg sysfs tcpd theora truetype unicode vaapi vala vorbis x264 xorg
xvid zeitgeist zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x
ca0106 cmipci emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0
intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci"
ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file
hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug
rate route share shm softvol" APACHE2_MODULES="actions alias auth_basic
authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm
authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache
cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter
file_cache filter headers include info log_config logio mem_cache mime
mime_magic negotiation rewrite setenvif speling status unique_id userdir
usertrack vhost_alias" CALLIGRA_FEATURES="kexi words flow plan stage tables
krita karbon braindump" CAMERAS="ptp2" COLLECTD_PLUGINS="df interface irq load
memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm
earthmate evermore fv18 garmin garmintxt gpsclock itrax mtk3301 nmea ntrip
navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf superstar2 timing
tsip tripmate tnt ubx" INPUT_DEVICES="keyboard mouse evdev wacom" KERNEL="linux"
LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses
text" LINGUAS="en en_GB" PHP_TARGETS="php5-3" RUBY_TARGETS="ruby18"
USERLAND="GNU" VIDEO_CARDS="intel" XTABLES_ADDONS="quota2 psd pknock lscan
length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq
steal rawnat logmark ipmark dhcpmac delude chaos account"
Unset:  CPPFLAGS, CTARGET, INSTALL_MASK, LC_ALL, PORTAGE_BUNZIP2_COMMAND,
PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS

----------
files: build.log.gz
messages: 4122
nosy: ikelos, pypy-issue
priority: bug
status: unread
title: __main__.UnrecognizedOperation: vmulsd

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 19:20:47 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Thu, 22 Mar 2012 18:20:47 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332440447.28.0.902177640287.issue1095@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

should be fixed in f18c75ec9e0f

----------
nosy: +fijal
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 20:45:54 2012
From: tracker at bugs.pypy.org (Mike Auty)
Date: Thu, 22 Mar 2012 19:45:54 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332445554.27.0.70369977241.issue1095@bugs.pypy.org>


Mike Auty <mike.auty at gmail.com> added the comment:

Thanks, I do seem to have gotten further, but am now getting:

__main__.UnrecognizedOperation: vcvttsd2siq

Would you like me to file a separate bug, or just keep reporting these as I run
into them on this bug?

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 21:44:53 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Thu, 22 Mar 2012 20:44:53 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332449093.93.0.179084340649.issue1095@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

If you look at f18c75ec9e0f (file pypy/translator/c/gcc/trackgcroot.py) you can 
see the list of ignored functions. just add them one after another. I'm not sure 
how to do it in gentoo, but typically you have /tmp/usession-.../testing_1 
directories where you can go and type make just after you modified 
trackgcroot.py to have quicker turnaround times.

Do you feel like providing a patch with all the things that should be ignored?

Cheers,
fijal

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 21:48:25 2012
From: tracker at bugs.pypy.org (Mike Auty)
Date: Thu, 22 Mar 2012 20:48:25 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332449305.46.0.788246837926.issue1095@bugs.pypy.org>


Mike Auty <mike.auty at gmail.com> added the comment:

Yep, I'm happy to, I just wasn't sure how to decide the length of the prefix. 
I'll produce a patch with the complete specific operations to ignore, and then
someone who knows if it'll cause problems can reduce those to a sensible prefix...

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 21:56:27 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Thu, 22 Mar 2012 20:56:27 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332449787.19.0.470165253163.issue1095@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

Sounds good to me, although prefix is to be found in intel specification of 
instructions.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 22 22:21:13 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Thu, 22 Mar 2012 21:21:13 +0000
Subject: [pypy-issue] [issue1050] socket attribute MSG_WAITALL under win32
In-Reply-To: <1329330388.67.0.22055127816.issue1050@bugs.pypy.org>
Message-ID: <1332451273.65.0.141286551986.issue1050@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

It's certainly because PyPy 1.8 was compiled with VS2010, whereas CPython2.7 is 
still compiled with VS2008.
Does this definition cause a problem?

----------
nosy: +afa
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1050>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 23 16:25:52 2012
From: tracker at bugs.pypy.org (Senyai)
Date: Fri, 23 Mar 2012 15:25:52 +0000
Subject: [pypy-issue] [issue1050] socket attribute MSG_WAITALL under win32
In-Reply-To: <1329330388.67.0.22055127816.issue1050@bugs.pypy.org>
Message-ID: <1332516352.36.0.941007684504.issue1050@bugs.pypy.org>


Senyai <senyai at gmail.com> added the comment:

Well, Pyro4 has this code:

if hasattr(socket, "MSG_WAITALL"):
    while True:
        try:
            data=sock.recv(size, socket.MSG_WAITALL)

That fails on windows XP, but I guess should work starting with Windows 2003.

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1050>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 23 16:37:39 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Fri, 23 Mar 2012 15:37:39 +0000
Subject: [pypy-issue] [issue1050] socket attribute MSG_WAITALL under win32
In-Reply-To: <1329330388.67.0.22055127816.issue1050@bugs.pypy.org>
Message-ID: <1332517059.61.0.702883740434.issue1050@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

This test is wrong: hasattr(socket, "MSG_WAITALL") is a condition on the compiler/SDK used to compile Python; instead it should test 
whether the socket transport supports this option.
On Windows, platform.win32_ver() could be used; alternatively, allow the operation to fail and fall back to the other behavior.

(btw, CPython3.3 is likely to be compiled with VS2010 as well)

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1050>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 24 21:02:22 2012
From: tracker at bugs.pypy.org (Mike Auty)
Date: Sat, 24 Mar 2012 20:02:22 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332619342.86.0.632997728343.issue1095@bugs.pypy.org>


Mike Auty <mike.auty at gmail.com> added the comment:

Ok, so it seems almost all of these are existing ignored prefixes, but starting
with a v.  Anyway, here's the patch/list I had to use to successfully compile
pypy on my box.  Hope it's of some use...

Please note, I was testing this against both 1.8 and the hg tip, so I've include
vmul even though that was already fixed in the hg.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 24 21:19:41 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sat, 24 Mar 2012 20:19:41 +0000
Subject: [pypy-issue] [issue1095] __main__.UnrecognizedOperation: vmulsd
In-Reply-To: <1332436379.97.0.2613513595.issue1095@bugs.pypy.org>
Message-ID: <1332620381.34.0.205581358525.issue1095@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

If it is what I think it is (http://en.wikipedia.org/wiki/VEX_prefix), then it's 
safe to ignore all 'v*' instructions. should be done in 03bbbea4dbff

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1095>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 01:39:33 2012
From: tracker at bugs.pypy.org (hm)
Date: Sun, 25 Mar 2012 00:39:33 +0000
Subject: [pypy-issue] [issue1097] data loss in file.write()
In-Reply-To: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>
Message-ID: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>


New submission from hm <hacman0 at gmail.com>:

http://pypy.org/compat.html mentions that 

open("filename", "w").write("stuff")

will not work as expected because there is no reference counting in pypy.  In fact, that line can cause data loss, as demonstrated in the 
attached file.

$ pypy pypyfile.py 
$ python pypyfile.py
$ cat CPython-output.txt
11111
234
$ cat PyPy-output.txt
11111

# my system:
$ pypy --version
Python 2.7.2 (2346207d99463f299f09f3e151c9d5fa9158f71b, Feb 20 2012, 04:52:48)
[PyPy 1.8.0]
$ uname -mrsv
Darwin 10.8.0 Darwin Kernel Version 10.8.0: Tue Jun  7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 i386
$ python --version
Python 2.7.2
# (also ran this on linux/python2.6 with the same results)

----------
files: pypyfile.py
messages: 4135
nosy: hacman, pypy-issue
priority: bug
status: unread
title: data loss in file.write()

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1097>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 01:51:22 2012
From: tracker at bugs.pypy.org (hm)
Date: Sun, 25 Mar 2012 00:51:22 +0000
Subject: [pypy-issue] [issue1098] sys.stdout.close() from interactive
	session crashes pypy	and messes up terminal
In-Reply-To: <1332636682.08.0.8292785494.issue1098@bugs.pypy.org>
Message-ID: <1332636682.08.0.8292785494.issue1098@bugs.pypy.org>


New submission from hm <hacman0 at gmail.com>:

Those last few $s are the result of hitting return a few times after pypy has quit.  The terminal does not seem to want to print newlines anymore.  
This happened on a Mac using the standard terminal program, but I have observed the same behavior in using iterm and on linux as well (the crash and 
the lack of newlines, although this is from an ssh session initiated on the same mac--I can check from a proper linux box at some point as well).

$ pypy
Python 2.7.2 (2346207d99463f299f09f3e151c9d5fa9158f71b, Feb 20 2012, 04:52:48)
[PyPy 1.8.0] on darwin
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``ECOOP is happening''
>>>> import sys
>>>> sys.stdout.close()
Traceback (most recent call last):
                                    File "app_main.py", line 51, in run_toplevel
                                                                                  File "/opt/local/lib/pypy/lib_pypy/_pypy_interact.py", line 40, in 
interactive_console
                       run_multiline_interactive_console(mainmodule)
                                                                      File "/opt/local/lib/pypy/lib_pypy/pyrepl/simple_interact.py", line 58, in 
run_multiline_interactive_console
                                 returns_unicode=True)
                                                        File "/opt/local/lib/pypy/lib_pypy/pyrepl/readline.py", line 210, in multiline_input
                                                                                                                                                return 
reader.readline(returns_unicode=returns_unicode)
                                                    File "/opt/local/lib/pypy/lib_pypy/pyrepl/reader.py", line 594, in readline
                                                                                                                                   self.restore()
                                                                                                                                                   File 
"/opt/local/lib/pypy/lib_pypy/pyrepl/reader.py", line 494, in restore
                                                                            self.console.restore()
                                                                                                    File 
"/opt/local/lib/pypy/lib_pypy/pyrepl/unix_console.py", line 396, in restore
                                   self.flushoutput()
                                                       File "/opt/local/lib/pypy/lib_pypy/pyrepl/unix_console.py", line 497, in flushoutput
                                                                                                                                               
self.__tputs(text)
              File "/opt/local/lib/pypy/lib_pypy/pyrepl/unix_console.py", line 515, in __tputs
                                                                                                  os.write(self.output_fd, fmt)
                                                                                                                               OSError: [Errno 9] Bad 
file descriptor
                $ $ $ $

----------
messages: 4136
nosy: hacman, pypy-issue
priority: bug
status: unread
title: sys.stdout.close() from interactive session crashes pypy and messes up terminal

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1098>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 05:29:31 2012
From: tracker at bugs.pypy.org (cactuss)
Date: Sun, 25 Mar 2012 03:29:31 +0000
Subject: [pypy-issue] [issue1099] distutils not correctly using g++,
	cplusplus compiler
In-Reply-To: <1332646171.02.0.679929203502.issue1099@bugs.pypy.org>
Message-ID: <1332646171.02.0.679929203502.issue1099@bugs.pypy.org>


New submission from cactuss <mgs33 at cornell.edu>:

I have a C++ module that I am currently using in a python application. I have 
bindings that use Python.h and compile it through the attached setup.py file. My 
build process in my current cPython app simply executes python setup.py 
build_ext --inplace. When I compile using ./pypy setup.py build_ext --inplace 
the resulting compilation uses 'cc' with pypy and 'g++' with python2.7. When I 
compile the module to pypy using ./pypy setup.py build_ext --inplace, the 
resulting binary does not work because it was not compiled with a c++ compiler. 
When I change 'cc' to 'g++' the module is successfully imported by the pypy 
interpreter. tl;dr cpython seems to automatically detect through distutils that 
it should use a c++ compiler and pypy does not when i run python setup.py 
build_ext --inplace

----------
files: setup.py
messages: 4137
nosy: cactuss, pypy-issue
priority: feature
status: unread
title: distutils not correctly using g++, cplusplus compiler

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1099>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 15:16:26 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sun, 25 Mar 2012 13:16:26 +0000
Subject: [pypy-issue] [issue1097] data loss in file.write()
In-Reply-To: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>
Message-ID: <1332681386.81.0.811220345561.issue1097@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

Yes, this is documented and also acknowledged as being a reasonable 
implementation-detail (refcounting or no refcounting). Closing as wontfix

----------
nosy: +fijal
status: unread -> wontfix

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1097>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 15:18:03 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Sun, 25 Mar 2012 13:18:03 +0000
Subject: [pypy-issue] [issue1098] sys.stdout.close() from interactive
	session crashes pypy	and messes up terminal
In-Reply-To: <1332636682.08.0.8292785494.issue1098@bugs.pypy.org>
Message-ID: <1332681483.28.0.434649223343.issue1098@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

Well, you just messed up with internals, what would be the expected behavior? I 
think this is also a wontfix.

The real reason is that pypy uses pyrepl for repl which is a piece of python code 
and it uses sys.stdout and not writing to file descriptors.

----------
nosy: +fijal
status: unread -> wontfix

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1098>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 25 15:37:35 2012
From: tracker at bugs.pypy.org (Ronny Pfannschmidt)
Date: Sun, 25 Mar 2012 13:37:35 +0000
Subject: [pypy-issue] [issue1098] sys.stdout.close() from interactive
	session crashes pypy	and messes up terminal
In-Reply-To: <1332636682.08.0.8292785494.issue1098@bugs.pypy.org>
Message-ID: <1332682655.53.0.695888971622.issue1098@bugs.pypy.org>


Ronny Pfannschmidt <Ronny.Pfannschmidt at gmx.de> added the comment:

the issue for that is 
https://bitbucket.org/pypy/pyrepl/issue/1/bad-breakage-when-sysstdout-is-closed

----------
nosy: +ronny

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1098>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 26 01:41:56 2012
From: tracker at bugs.pypy.org (hm)
Date: Sun, 25 Mar 2012 23:41:56 +0000
Subject: [pypy-issue] [issue1097] data loss in file.write()
In-Reply-To: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>
Message-ID: <1332718916.3.0.976974456137.issue1097@bugs.pypy.org>


hm <hacman0 at gmail.com> added the comment:

The order of the output might be considered an implementation detail (although I'd argue that 
it isn't), but whether output is recorded at all hardly seems an implementation detail.  If 
each works on its own (which is the case), they should be expected to work together.  If the 
flush of the write is delayed until garbage collection, then the seek of the file pointer for 
proper "append" behavior should also be delayed.  Otherwise, calling open(..., 'a') is behaving 
like open(..., 'w').  That is the behavior I would call a bug.  I can understand if this is not 
an urgent item for you, as I know you are busy with other issues and making pypy even more 
awesome, but I'd urge you to reconsider "wontfix." At least answer this, then I'll leave you 
alone: if someone were to submit a patch modifying this behavior without breaking or slowing 
anything else, would you reject it?  I'm not offering at the moment, although I'd think about 
looking into what it would take at least, but take it as a rhetorical question for now.  jython, for instance, outputs all of the data, but out of order from how it is specified in the 
code.

I read the further explanation at 
http://pypy.readthedocs.org/en/latest/cpython_differences.html and I can see the point of view 
there, but it seems wrong in the case of appends.  The following C program, for instance, 
outputs 234511111111111111111
#include <stdio.h>
int main()
{
    char a1[] = "11111111111111111";
    char a2[] = "2345";
    FILE *f1 = fopen("bye", "a");
    fwrite(a1, sizeof(a1[0]), sizeof(a1)-1, f1);
    FILE *f2 = fopen("bye", "a");
    fwrite(a2, sizeof(a2[0]), sizeof(a2)-1, f2);
    fflush(f2);
}

Whether or not we agree on this particular issue, thanks for all the great work on the project!

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1097>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 26 09:47:31 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Mon, 26 Mar 2012 07:47:31 +0000
Subject: [pypy-issue] [issue1097] data loss in file.write()
In-Reply-To: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>
Message-ID: <1332748051.08.0.864085934372.issue1097@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Is it because streamio.py never uses the O_APPEND flag?

----------
nosy: +afa

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1097>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 26 10:04:49 2012
From: tracker at bugs.pypy.org (Fijal)
Date: Mon, 26 Mar 2012 08:04:49 +0000
Subject: [pypy-issue] [issue1097] data loss in file.write()
In-Reply-To: <1332635973.49.0.384966847778.issue1097@bugs.pypy.org>
Message-ID: <1332749089.2.0.946937938834.issue1097@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

It comes from unix file semantics. You have two open files and 234 is written 
*before* 1111 is written. There are possible workarounds, if you want this 
particular usecase, but I suggest closing files is probably a better idea.

As far as patches go - I think we would accept a reasonable patch, but I still 
consider this as a wontfix.

Cheers,
fijal

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1097>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 26 15:28:58 2012
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Mon, 26 Mar 2012 13:28:58 +0000
Subject: [pypy-issue] [issue1100] Dict iterators can return deleted entries
In-Reply-To: <1332768538.31.0.460128219214.issue1100@bugs.pypy.org>
Message-ID: <1332768538.31.0.460128219214.issue1100@bugs.pypy.org>


New submission from Armin Rigo <armin.rigo at gmail.com>:

Dict iterators (and sets in the set-strategies branch) may return deleted
entries if the dict is "cleverly" mutated while iterating in a way that changes
its strategy.  While it's known that iterating over a mutated dict should not be
done, the result can still be surprizing:

>>>> d = {3:4, 6:7}
>>>> it = d.iteritems()
>>>> d['foo'] = 'bar'            
>>>> del d[6]
>>>> it.next()
(3, 4)
>>>> it.next()
(6, 7)

Maybe the dict iterator should check not only that the dict size didn't change,
but also that its strategy is still the same one.

----------
messages: 4144
nosy: arigo, pypy-issue
priority: bug
status: unread
title: Dict iterators can return deleted entries

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1100>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 27 15:18:30 2012
From: tracker at bugs.pypy.org (Simon Sapin)
Date: Tue, 27 Mar 2012 13:18:30 +0000
Subject: [pypy-issue] [issue1094] Non-BPM unicode literals decoded as
	surrogate pairs
In-Reply-To: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>
Message-ID: <1332854310.27.0.550301248836.issue1094@bugs.pypy.org>


Simon Sapin <simon.sapin at kozea.fr> added the comment:

CPython 3.1.4 and 3.2.2 behave as described above. CPython 3.0.1 has the same
bug as PyPy 1.8.0.

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1094>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 27 15:20:21 2012
From: tracker at bugs.pypy.org (Simon Sapin)
Date: Tue, 27 Mar 2012 13:20:21 +0000
Subject: [pypy-issue] [issue1094] Non-BPM unicode literals decoded as
	surrogate pairs
In-Reply-To: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>
Message-ID: <1332854421.66.0.501603905834.issue1094@bugs.pypy.org>


Simon Sapin <simon.sapin at kozea.fr> added the comment:

CPython 2.7.2, 3.1.4 and 3.2.2 behave as described above.
CPython 2.6.7 and 3.0.1 have the same bug as PyPy 1.8.0.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1094>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 27 15:33:23 2012
From: tracker at bugs.pypy.org (Simon Sapin)
Date: Tue, 27 Mar 2012 13:33:23 +0000
Subject: [pypy-issue] [issue1094] Non-BPM unicode literals decoded as
	surrogate pairs
In-Reply-To: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>
Message-ID: <1332855203.73.0.942123881244.issue1094@bugs.pypy.org>


Simon Sapin <simon.sapin at kozea.fr> added the comment:

Assuming UCS4 (wide unicode):
CPython 2.6.7, 2.7.2, 3.1.4 and 3.2.2 behave as described above.
CPython 3.0.1 has the same bug as PyPy 1.8.0: only codepoints3 fails.

(Sorry for the duplicate message, I had some of my pythons in UCS2.)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1094>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 28 10:33:59 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Wed, 28 Mar 2012 08:33:59 +0000
Subject: [pypy-issue] [issue1094] Non-BPM unicode literals decoded as
	surrogate pairs
In-Reply-To: <1332254179.6.0.235268017197.issue1094@bugs.pypy.org>
Message-ID: <1332923639.45.0.0777017037508.issue1094@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

Most probably related to CPython bug http://bugs.python.org/issue3297 which was 
fixed in 2.7 with changeset http://hg.python.org/cpython/rev/032d9749047d

PyPy has very similar code in pypy/interpreter/pyparser/parsestring.py.

----------
nosy: +afa

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1094>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 29 17:36:23 2012
From: tracker at bugs.pypy.org (David Ripton)
Date: Thu, 29 Mar 2012 15:36:23 +0000
Subject: [pypy-issue] [issue1085] cpyext: PyString_AsString() returns bogus
	results for	non-bytes input
In-Reply-To: <1331478486.45.0.687703104839.issue1085@bugs.pypy.org>
Message-ID: <1333035383.11.0.457678926513.issue1085@bugs.pypy.org>


David Ripton <d+pypy at ripton.net> added the comment:

I translated PyPy on 64-bit Ubuntu 11.04, and ./pypy-c pytest.py
pypy/module/cpyext/test/test_stringobject.py gives errors in a bunch of
app-level tests, then segfaults in TestString.test_string_resize

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1085>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 31 10:18:42 2012
From: tracker at bugs.pypy.org (papaeye)
Date: Sat, 31 Mar 2012 08:18:42 +0000
Subject: [pypy-issue] [issue1101] Can't use unicode string as key in
	__dict__ attribute
In-Reply-To: <1333181922.06.0.897522719468.issue1101@bugs.pypy.org>
Message-ID: <1333181922.06.0.897522719468.issue1101@bugs.pypy.org>


New submission from papaeye <papaeye at gmail.com>:

In PyPy 1.8 on Mac OS X 10.7.3 and nightly binary 
build(http://buildbot.pypy.org/nightly/trunk/pypy-c-jit-54086-4f1f458bedae-
linux64.tar.bz2) on Debian 6.0.4:

>>> class Foo(object): attribute = 'foo'
>>> 'attribute' in Foo.__dict__
True
>>> u'attribute' in Foo.__dict__
False
>>> d = {'foo': 'bar'}
>>> 'foo' in d
True
>>> u'foo' in d
True

In CPython 2.5.6/2.6.7/2.7.2 and Jython 2.5.2 on Mac OS X 10.7.3, all of the 
above return True.

----------
nosy: +papaeye
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1101>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 31 13:11:51 2012
From: tracker at bugs.pypy.org (Amaury Forgeot d Arc)
Date: Sat, 31 Mar 2012 11:11:51 +0000
Subject: [pypy-issue] [issue1101] Can't use unicode string as key in
	__dict__ attribute
In-Reply-To: <1333181922.06.0.897522719468.issue1101@bugs.pypy.org>
Message-ID: <1333192311.12.0.943823970936.issue1101@bugs.pypy.org>


Amaury Forgeot d Arc <amauryfa at gmail.com> added the comment:

This is now fixed (c07e2a2ba2f8)
Thanks for reporting this use case!

----------
nosy: +afa
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1101>
________________________________________