[issue23408] Pickle tests use incorrect test data

New submission from Serhiy Storchaka: Pickle tests use incorrect test data for protocols 0-2. This data is not compatible with Python 2 because refers to Python 3 modules builtins and copyreg (instead of __builtin__ and copy_reg). Proposed patch fixes this (data was regenerated by

[issue17896] Move Windows external libs from src\..\ to src\externals

Martin v. Löwis added the comment: I noticed this issue today when trying to make the 3.4.3rc1 build; this broke Tools/msi.py, which fails to find the license files and the tcltk files. -- ___ Python tracker rep...@bugs.python.org

[issue23410] Document more BaseHTTPRequestHandler attributes

New submission from Martin Panter: [Padding to avoid Error: 'utf8' codec can't decode bytes in position 189-190: invalid continuation byte] This is a patch to document two attributes of http.server.BaseHTTPRequestHandler: “close_connection” and “requestline”. Normally these are set by the

[issue23410] Document more BaseHTTPRequestHandler attributes

Changes by Martin Panter vadmium...@gmail.com: -- keywords: +patch Added file: http://bugs.python.org/file38042/http-attributes.patch ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23410 ___

[issue23096] Implementation-depended pickling floats with protocol 0

Serhiy Storchaka added the comment: Here is a patch. -- keywords: +patch stage: - patch review versions: -Python 2.7, Python 3.4 Added file: http://bugs.python.org/file38039/pickle_float_repr.patch ___ Python tracker rep...@bugs.python.org

[issue11734] Add half-float (16-bit) support to struct module

Mark Dickinson added the comment: Refreshed and updated patch, cleaning up some style issues, fixing a failure to return -1 on exception in np_halffloat, and removing some endianness-determining casty expressions in favour of using PY_LITTLE_ENDIAN. -- Added file:

[issue23411] Update urllib.parse.__all__

Changes by Martin Panter vadmium...@gmail.com: Added file: http://bugs.python.org/file38045/urllib.parse-all.patch ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23411 ___

[issue17023] Subprocess does not find executable on Windows if it is PATH with quotes

Changes by Tim Golden m...@timgolden.me.uk: -- resolution: - wont fix stage: - resolved status: open - closed ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue17023 ___

[issue3213] pydoc -p should listen to [::] if IPv6 is supported

Changes by Martin Panter vadmium...@gmail.com: -- nosy: +vadmium ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue3213 ___ ___ Python-bugs-list

[issue23411] Update urllib.parse.__all__

Changes by Martin Panter vadmium...@gmail.com: Removed file: http://bugs.python.org/file38044/urllib.parse-all.patch ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23411 ___

[issue23411] Update urllib.parse.__all__

New submission from Martin Panter: +DefragResult, ParseResult, SplitResult, +DefragResultBytes, ParseResultBytes, SplitResultBytes] Also adds test case. -- components: Library (Lib) files: urllib.parse-all.patch keywords: patch messages: 235556 nosy: vadmium priority: normal severity:

[issue20680] Pickle socket enums by names

Ethan Furman added the comment: An IntEnum is just a fancy int. You can add them, subtract them, multiply them, all that kind of thing. If you attempt to unpickle a 3.4 _PickleByNameIntEnum on a 3.3 system, what's going to happen? -- ___ Python

[issue20680] Pickle socket enums by names

Serhiy Storchaka added the comment: If you attempt to unpickle a 3.4 _PickleByNameIntEnum on a 3.3 system, what's going to happen? The same as if you attempt to unpickle a 3.4 IntEnum on a 3.3 system -- AttributeError. -- ___ Python tracker

[issue23103] Reduce memory usage for ipaddress object instances

Serhiy Storchaka added the comment: Synchronized with tip. -- Added file: http://bugs.python.org/file38047/ipaddress_lightweight_4.patch ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23103

[issue23413] Incorrect division result

New submission from Martynas Brijunas: Dear Python team, when dividing 3 by 7, I expect to get the following result: 0.428571428571428571428571428571428571428571428571428571428571... What I am getting in the Python interpreter is this: 3 / 7 0.42857142857142855 Which in my opinion is

[issue23412] importlib sometimes fails to import a recently created module

SilentGhost added the comment: Looks like some sort of race condition. Sleeping for a second before import_module seem to solve the problem. -- components: +Interpreter Core nosy: +SilentGhost, brett.cannon, eric.snow, ncoghlan ___ Python tracker

[issue23412] importlib sometimes fails to import a recently created module

Changes by SilentGhost ghost@gmail.com: -- versions: +Python 3.5 ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23412 ___ ___ Python-bugs-list

[issue21793] httplib client/server status refactor

Ethan Furman added the comment: Without having looked at the code I would guess the fix is as simple as changing a %s to a %d. Having said that, it would be nice if the name was also in the log -- something like: 127.0.0.1 - - [08/Feb/2015 05:05:28] GET / HTTP/1.1 200 (OK) - --

[issue23413] Incorrect division result

Benjamin Peterson added the comment: https://docs.python.org/3/tutorial/floatingpoint.html -- nosy: +benjamin.peterson resolution: - not a bug status: open - closed ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23413

[issue23412] importlib sometimes fails to import a recently created module

New submission from Richard Dymond: importlib.import_module() sometimes fails to import a module that has just been written to the filesystem, aborting with an ImportError. Example output when executing the attached file with Python 3.3 or 3.4: Wrote tmpwbzb35.py Successfully imported

[issue23411] Update urllib.parse.__all__

Senthil Kumaran added the comment: Thanks for the tests. Reviewed the patch and looks good to me. -- nosy: +orsenthil stage: - commit review ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23411

[issue23397] PEP 431 implementation

Changes by Berker Peksag berker.pek...@gmail.com: -- nosy: +lregebro ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23397 ___ ___ Python-bugs-list

[issue23414] seek(count, whence) accepts bogus whence on windows, python2.7

New submission from mattip: f=open('abc.txt', 'w+') f.seek(0, 42) does not raise an exception on windows, python2.7 -- components: Windows messages: 235568 nosy: mattip, steve.dower, tim.golden, zach.ware priority: normal severity: normal status: open title: seek(count, whence) accepts

[issue17911] traceback: add a new thin class storing a traceback without storing local variables

Robert Collins added the comment: Sorry xonatius I wasn't clear: AFAICT your patch is going to require changes to the traceback tests, and this issue is changing the implementation substantially: I was suggesting that you make sure your patch applies on top of this issue, not that you merge

[issue17911] traceback: add a new thin class storing a traceback without storing local variables

Robert Collins added the comment: Nick, Antoine - I'm now stuck between you. Options going forward: - I can JFDI realising you won't both be happy :) - you two can reach consensus! I could cripple __init__ by switching to __new__, but I think thats massive overcomplication and not needed.

[issue10320] printf %qd is nonstandard

John Malmberg added the comment: OpenVMS needs %lld or % PY_FORMAT_LONG_LONG d in order to build the _ctypes module. -- nosy: +John.Malmberg ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue10320

[issue10320] printf %qd is nonstandard

Mark Lawrence added the comment: As VMS is not supported I doubt that people here are too bothered about what is needed to build the _ctypes module on it. -- nosy: +BreamoreBoy versions: +Python 3.4, Python 3.5 -Python 3.1, Python 3.2 ___ Python

[issue23415] add-to-pydotorg does not support .exe installers for Windows

New submission from Larry Hastings: Steve is using new technology to make the installers for Windows. He generates four installers now: * .exe for Win32 * -amd64.exe for Win64 * -webinstall.exe for Win32 web-based installers * -amd64-webinstall.exe for Win64 web-based installers

[issue23414] seek(count, whence) accepts bogus whence on windows, python2.7

eryksun added the comment: The whence argument is used in a switch statement that handles SEEK_END and SEEK_CUR. It doesn't raise an error for an invalid whence value. It just falls through to the fsetpos call. _portable_fseek

[issue23415] add-to-pydotorg does not support .exe installers for Windows

Larry Hastings added the comment: I should add, adding the files by hand worked fine. (Which means that if you experiment with the script, when it blows away the files and re-adds them, you'll be blowing away the files I added by hand. So if you do so, it's up to you to ensure the files are

[issue17911] traceback: add a new thin class storing a traceback without storing local variables

Nick Coghlan added the comment: Escalating to python-dev for API design feedback is the usual path forward if we reach an impasse in the tracker comments. I'll make one more attempt at persuading Antoine here though: the fact that we're being tempted to add do not use this API the way you would

[issue22235] httplib: TypeError with file() object in ssl.py

Etienne Robillard added the comment: problem is fixed with python 2.7.9. it seem ssl.py module has been greatly improved with this release. Thanks! -- resolution: - fixed ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue22235

[issue20680] Pickle socket enums by names

Serhiy Storchaka added the comment: For the socket module constants there is even better way, which supports forward compatibility, i.e. SOCK_STREAM pickled in 3.5 could be unpickled on 3.3 (replacing ints with enums breaks forward compatibility). Here is a patch. -- Added file:

[issue10697] host and port attributes not documented well in function urllib.parse.urlparse and urlsplit

Martin Panter added the comment: I don’t understand where the work needs to be done for this one. Even in the 3.1 and 2.7 documentation, the urlparse() and urlsplit() entries both list “port” as one of the returned attributes, and urlparse() has example code for it. -- nosy: +vadmium

[issue15009] urlsplit can't round-trip relative-host urls.

Martin Panter added the comment: Fixing Issue 22852 or Issue 5843 should help fixing this. -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue15009 ___

[issue23418] Keep http.server.__all__ up to date

New submission from Martin Panter: Here is a patch that adds the Simple/CGIHTTPRequestHandler classes to __all__, and a test case that should help keep it up to date in the future. -- components: Library (Lib) files: http.server-all.patch keywords: patch messages: 235594 nosy: vadmium

[issue17896] Move Windows external libs from src\..\ to src\externals

Roundup Robot added the comment: New changeset 7d22dbf3a0dc by Martin v. Löwis in branch '3.4': Issue #17896: Update msi.py to new externals dir. https://hg.python.org/cpython/rev/7d22dbf3a0dc -- ___ Python tracker rep...@bugs.python.org

[issue23328] urllib2 fails for proxy credentials that contain a '/' character

Martin Panter added the comment: The relevant code looks like it is _parse_proxy() at Lib/urllib/request.py:693. It has custom code to search for a slash (/), so it wouldn’t be hard to make it search after the last at (@) symbol. (I previously assumed it would use urlsplit() or similar, which

[issue23072] 2.7.9 multiprocessing compile conflict

Davin Potts added the comment: It's a little unclear if one or more of the following is true: (1) you are having trouble getting the multiprocessing module to build properly on Solaris 2.8; (2) you are having trouble getting your own code to build in a similar way to the multiprocessing

[issue23417] Windows 8.1 and Windows Server 2012 R2 are not displayed properly

New submission from George Schizas: Python on Windows can now can understand that it's on Windows 8.1 and Windows Server 2012 R2, but platform.py hasn't been updated, and claims it's on post2012Server: import platform print(platform.win32_ver()) ('post2012Server', '6.3.9600', '',

[issue21429] Input.output error with multiprocessing

Davin Potts added the comment: The provided links to the relevant code are unfortunately pointing at the master branch's copy of those files and not a specific version of the code making it that much harder to now discern where the issue arises. (A number of things have changed in the master

[issue23415] add-to-pydotorg does not support .exe installers for Windows

Georg Brandl added the comment: Can you attach the diff you made? -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue23415 ___ ___ Python-bugs-list

[issue22235] httplib: TypeError with file() object in ssl.py

Etienne Robillard added the comment: hg push worked like a charm... but for some reasons i get another error trying to install things with setuptools: erob@nguns:~/Desktop/pyaml-14.12.10$ sudo python setup.py install --prefix=/usr/local Processing dependencies for pyaml==14.12.10 Searching

[issue12955] urllib.request example should use with ... as:

Martin Panter added the comment: Issue 22755 is about the example arms race for contextlib.closing(). -- nosy: +vadmium ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue12955 ___

[issue17911] traceback: add a new thin class storing a traceback without storing local variables

Antoine Pitrou added the comment: You can also switch back to a single constructor, taking either an exception instance or an exception triple. That way everyone is happy - except perhaps if you are ideologically opposed to polymorphic signatures :-) Switching to python-dev would probably be

[issue3647] urlparse - relative url parsing and joins to be RFC3986 compliance

Changes by Martin Panter vadmium...@gmail.com: -- nosy: +vadmium ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue3647 ___ ___ Python-bugs-list

[issue5843] Normalization error in urlunparse

Martin Panter added the comment: Looks like this duplicates Issue 22852, which has a patch, although its author had second thoughts on the implementation -- nosy: +vadmium ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue5843

[issue18140] urlparse.urlsplit confused to fragment when password include #

Martin Panter added the comment: Possibly related: Issue 23328. I would have thought any special characters in the password etc field should be encoded. Where do such unencoded URLs come from? It seems that this would change the behaviour of parsing a URL such as

[issue23328] urllib2 fails for proxy credentials that contain a '/' character

Martin Panter added the comment: Related: Issue 18140. The slash character is meant to be a reserved character in URLs, so why hasn’t it been encoded? Where does the environment variable come from? -- nosy: +vadmium ___ Python tracker

[issue23328] urllib2 fails for proxy credentials that contain a '/' character

Andy Reitz added the comment: The proxy credentials are supplied by our sysadmin. My understanding is that the http_proxy env variable doesn't require URI encoding. In addition, the same credentials work fine with curl. -- ___ Python tracker

[issue22852] urllib.parse wrongly strips empty #fragment

Martin Panter added the comment: I also liked the idea of returning None to distinguish a missing URL component from an empty-but-present component, and it would make them more consistent with the “username” and “password” fields. But I agree it would break backwards compabitility too much.

[issue23416] Make urllib.parse.SplitResult etc arguments optional

New submission from Martin Panter: This would be a simple API enhancement and would allow easier building of URLs, like SplitResult(rtp, address, query=urlencode(query)).geturl() rtp://localhost:5004?rtcpport=5005 It seems the best way to do this at the moment is annoyingly verbose:

[issue12955] urllib.request example should use with ... as:

Martin Panter added the comment: Here is a patch to change the urlopen() examples to use context managers where appropriate. There were also a few examples of handling HTTPError which I didn’t touch, because the whole file object versus exception object thing is probably a separate can of