[issue12113] test_packaging fails when run twice

Alexis Metaireau added the comment: Same here, cannot reproduce the issue. I wasn't able to reproduce it even before Tarek's commit so I suppose his commit didnt changed anything regarding this hash problem. Antoine & Tarek, does the attached additional test reproduces the

[issue12391] packaging install fails to clean up temp files

Alexis Metaireau added the comment: I'm +1 on applying this patch as well. Removing files in the tmp directory is far better than letting the OS doing so. -- ___ Python tracker <http://bugs.python.org/is

[issue8668] Packaging: add a 'develop' command

Alexis Metaireau added the comment: Carl, I believe that's this one: http://wiki.python.org/moin/UsecasesOfDevelop -- ___ Python tracker <http://bugs.python.org/i

[issue12697] timeiot documention still refers to the timeit.py script

New submission from Alexis Metaireau : The example section of the timeit documentation still refers to "timeit.py", and isn't using the "python -m timeit" syntax used above. http://docs.python.org/library/timeit.html#examples I'm not sure when and if the timeit.p

[issue12697] timeiot documention still refers to the timeit.py script

Changes by Alexis Metaireau : -- nosy: +Boris.FELD ___ Python tracker <http://bugs.python.org/issue12697> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue12697] timeit documention still refers to the timeit.py script

Changes by Alexis Metaireau : -- title: timeiot documention still refers to the timeit.py script -> timeit documention still refers to the timeit.py script ___ Python tracker <http://bugs.python.org/issu

[issue6203] 3.x locale does not default to C, contrary to the documentation and to 2.x behavior

Alexis Metaireau added the comment: Maybe could it be useful to specify in the documentation that getlocale() is not intended to be used to get information about what is the locale of the system? It's not explained currently and thus it's a bit weird to have getlocale returning (

[issue6203] 3.x locale does not default to C, contrary to the documentation and to 2.x behavior

Alexis Metaireau added the comment: I see two different things here: 1) the fact that getlocale() doesn't return (None, None) on some python versions 2) the fact that having it returning (None, None) by default is a bit misleading as users may think that getlocale() is tied to enviro

[issue12726] explain why locale.getlocale() does not read system's locales

New submission from Alexis Metaireau : The documentation about locale.getlocale() doesn't talk about the fact that the locale isn't read from the system locale. Thus, it seemed strange to have locale.getlocale() returning (None, None). As it seems to be the expected behaviour, it se

[issue8668] Packaging: add a 'develop' command

Alexis Metaireau added the comment: _run_setuptools_install is only intended to support setuptools setup.py, converting .egg-info to .dist-info, internally. IMO, you should not care about the differences between setuptools/distutils1/setuptools at this level, as it should be taken care at

[issue8668] Packaging: add a 'develop' command

Alexis Metaireau added the comment: IOW, in my opinion, support for setuptools develop command is not needed in packaging core, and still be taken care directly be the users wanting to run python setup.py develop: I don't see any reason to make it avaible on the s

[issue8668] Packaging: add a 'develop' command

Alexis Metaireau added the comment: Yep, packaging is not keeping the .egginfo directories, or at least does not plan to keep them (It should be the case currently but I haven't checked recently) in the upcoming release, so I would go on removing support for setuptools' develop co

[issue8668] Packaging: add a 'develop' command

Alexis Metaireau added the comment: On 08/18/2011 05:54 PM, higery wrote: > Then do you also mean support that for setuptools install is also not > necessary in packaging core? setuptools install is only supported in packaging because it's a widely used thing, and many python di

[issue12703] Improve error reporting for packaging.util.resolve_name

Alexis Metaireau added the comment: Thanks Rémy, About testing, I would go for modules with errors in it and check that when imported trough this function it does what it is supposed to do. IOW: 1. Create a test python module with errors in their definition (Throw an exception would do

[issue8190] Add a PyPI XML-RPC client module

Alexis Metaireau added the comment: Please, can you be a bit more precise about the missing features ? I'll be glad to cover them. -- ___ Python tracker <http://bugs.python.org/i

[issue8190] Add a PyPI XML-RPC client module

Alexis Metaireau added the comment: Yes, all the functions are covered. By the way, I've covered all the functions based on the pypi source code, not on the wiki. I'll update the wiki page with some examples soon :) -- ___ Python trac

[issue11357] Add support for PEP 381 -- Mirror Authenticity

Alexis Metaireau added the comment: Some nitpicks: In mirrors.get_server_key, the documentation is not up to date with your last changes (raises an error if there is a problem instead of returning None) You do use the name 'package' while talking about distributions or projects.

[issue11357] Add support for PEP 381 -- Mirror Authenticity

Alexis Metaireau added the comment: Antoine Pitrou on #python-dev made interesting remarks about the validation: 16:19 < __ap__> hmm the way the patch does validation is bogus 16:22 < __ap__> because it opens the URL a first time, validates it, then opens it a second time with url

[issue11921] distutils2 should be able to compile an Extension based on the Python implementation version

Alexis Metaireau added the comment: This raises a concern about python specific python implementations dependencies. We probably could extend PEP 345 in order to support things such as 'platform.python_implementation == "cpython"'. -- __

[issue11921] distutils2 should be able to compile an Extension based on the Python implementation version

Alexis Metaireau added the comment: On 29/04/2011 18:20, Daniel Holth wrote: > New in version 2.6. Yep that's it. We would need to backport it in the python2 port of packaging (namely distutils2), but it would do the trick. I just started a discussion about that on the fellow

[issue5302] Allow package_data globs match directories

Alexis Metaireau added the comment: The MANIFEST.in is definitely gone in distutils2. Can we close that? (don't have the rights to do so ‑ it can be handy on distutils2 bugs) -- nosy: +alexis ___ Python tracker <http://bugs.python.org/i

[issue5243] Missing dependency in distutils build

Alexis Metaireau added the comment: Has the patch been applied on distutils(1/2) ? -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue5243> ___ ___

[issue901727] extra_path kwarg to setup() undocumented

Alexis Metaireau added the comment: I've applied the patch on distutils2. This can now be closed. -- nosy: +alexis ___ Python tracker <http://bugs.python.org/iss

[issue6087] distutils.sysconfig.get_python_lib gives surprising result when used with a Python build

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue6087> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue2200] find_executable fails to find .bat files on win32

Alexis Metaireau added the comment: Have the patch been applied ? (the state is still open since last message) -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue2

[issue7546] msvc9compiler.py: add .asm extension

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue7546> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue901727] extra_path kwarg to setup() undocumented

Alexis Metaireau added the comment: Still, the extra_path argument exists and can be used, it's worth to have it documented somewhere, especially if someone have done it. That's also true that apiref.rst is outdated in d2, and will need a comple

[issue1083299] Distutils doesn't pick up all the files it should.

Changes by Alexis Metaireau : -- components: -Distutils2 nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1083299> ___ ___ Python-bugs-list mailin

[issue1597850] Cross compiling patches for MINGW

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1597850> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue1294032] Distutils writes keywords comma-separated

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1294032> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue1092365] Distutils needs a way *not* to install files

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1092365> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue1635363] Add command line help to windows unistall binary

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1635363> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue1011113] Make “install” find the build_base directory

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue103> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue11213] distutils2 test issue

Alexis Metaireau added the comment: yeah, thanks for that ! -- ___ Python tracker <http://bugs.python.org/issue11213> ___ ___ Python-bugs-list mailing list Unsub

[issue11219] Produce a warning when the license is specified in both the License and Classifier metadata fields

Alexis Metaireau added the comment: Hi, I was more thinking about something like: if the license is specified in the "License" metadata, then check that it's not a well known license (which can and must be provided by the classifiers instead). At the end, the code you

[issue11219] Produce a warning when the license is specified in both the License and Classifier metadata fields

Alexis Metaireau added the comment: That's almost what the PEP says, or at least what I understand of it. The platform and license fields should be used only if no matching classifier exists for them. -- ___ Python tracker <http://bugs.py

[issue14317] index.simple module lacking in distutils2

Changes by Alexis Metaireau : -- resolution: -> wont fix status: open -> closed ___ Python tracker <http://bugs.python.org/issue14317> ___ ___ Python-bugs-

[issue14317] index.simple module lacking in distutils2

Alexis Metaireau added the comment: The APIs of distutils2 have changed. the "index" module is now named "pypi". So, doing something like:: >>> from distutils2.pypi.simple import Crawler should work. --

[issue11880] add a {dist-info} category to distutils2

Alexis Metaireau added the comment: Hey, is there any news on this bug? Berker? -- ___ Python tracker <http://bugs.python.org/issue11880> ___ ___ Python-bug

[issue14294] Requirements are not properly copied into metatdata of dist-info

Alexis Metaireau added the comment: Thanks for reporting this Preston, Could you be a bit more precise on this one? What is the input you are giving (in the dist-info) and what is not being done? (what's the expected output and what's what you have as a result

[issue5302] Allow package_data specs/globs to match directories

Alexis Metaireau added the comment: Agreed on this one. That would avoid to have a new syntax for this case, which seems to be very common. -- ___ Python tracker <http://bugs.python.org/issue5

[issue14319] cleanup index switching mechanism on packaging.pypi

New submission from Alexis Metaireau : I need to do some general cleanup on packaging/d2, and especially on this bit. Currently, the implementation is way too complicated for what it does. -- assignee: alexis components: Distutils2 messages: 155903 nosy: alexis, tarek priority: normal

[issue14279] packaging.pypi should support flat directories of distributions

Alexis Metaireau added the comment: It is supposed to work already, but I'm not sure this is tested or complete. I'll have a closer look on this. +1 on the general idea. -- assignee: eric.araujo -> alexis ___ Python tracker <http:/

[issue14280] packaging.pypi should not require checksums

Alexis Metaireau added the comment: If no MD5 checksum is present on the crawled simple index, then we don't have to check them. This means we introduce a potential security hole here (md5 checksums were added for a reason). What could be done is to explicitely don't check them i

[issue14275] pysetup create doesn't handle install requirements

Changes by Alexis Metaireau : -- dependencies: +Requirements are not properly copied into metatdata of dist-info ___ Python tracker <http://bugs.python.org/issue14

[issue14275] pysetup create doesn't handle install requirements

Changes by Alexis Metaireau : -- dependencies: -Requirements are not properly copied into metatdata of dist-info ___ Python tracker <http://bugs.python.org/issue14

[issue14275] pysetup create doesn't handle install requirements

Alexis Metaireau added the comment: Thanks for reporting this, Jan-Jaap, I've marked this issue as easy, it's mostly a matter of looking at the passed parameter and outputing the list in the generated file, respecting the PEP 345 format (http://www.python.org/dev/peps/pep-0345

[issue14280] packaging.pypi should not require checksums

Alexis Metaireau added the comment: Right, I'll go for this then. -- ___ Python tracker <http://bugs.python.org/issue14280> ___ ___ Python-bugs-list m

[issue14279] packaging.pypi should support flat directories of distributions

Alexis Metaireau added the comment: Oh, thanks for clarifying this. I'll have a look at what the blockers are on this. I'm wondering if "local files" can be considered a simple index or not. If not, we can add another PyPI reader, which wo

[issue14356] Distutils2 ignores site-local configuration

Alexis Metaireau added the comment: The supposed way to work, for OS packagers, is to ship this sysconfig.cfg thing. I'm not sure we should rely on a customized site-local configuration, without defining any standard way of doing this (IOW: what are we looking for in the site-local c

[issue14915] pysetup may leave a package in a half-installed state

Alexis Metaireau added the comment: Oh, a potential way to avoid this would be to check that the metadata have the python 3 trove classifier in it. We could also add a way to force the installation even if the right classifier is not present with a special flag passed to the command line

[issue14946] packaging’s pysetup script should be named differently than distutils2’s pysetup

Alexis Metaireau added the comment: Good catch, One solution would be to have the pysetup script name suffixed by the version of python is comes with. For instance, for python 3.3 it would be "pysetup3.3" (that's how it is atm). This means that "pysetup" would poin

[issue14948] setup.cfg - rename home_page to homepage

Alexis Metaireau added the comment: While I kind of agree with you, this field is proposed by the PEP 345 and had been approved as it is. It seems probably better to stick with what's defined there mainly because any change to the PEP is a heavy process and we won't have time to go

[issue14949] Documentation should state clearly the differences between "pysetup run install_dist" and "pysetup install"

New submission from Alexis Metaireau : The packaging tutorial currently talks about "pysetup run install_dist" and "pysetup install" which aren't doing the same thing: one fetches the dependencies while the other doesn't. This should be stated clearly. --

[issue14950] packaging.install does not have a clear API.

New submission from Alexis Metaireau : The "packaging.install" module doesn't have a clear API defined and it's doing a lot of indirections between all the functions to get metadata, fetch the dependencies and install what's need to be installed. We might be a

[issue12355] Crawler doesn't follow redirection

Alexis Metaireau added the comment: Did someone changed anything in the codebase regarding this (or did it solved itself magically?) -- ___ Python tracker <http://bugs.python.org/issue12

[issue14949] Documentation should state clearly the differences between "pysetup run install_dist" and "pysetup install"

Alexis Metaireau added the comment: That would be helpful yep. I can add this in the docs if you've not done it already. -- ___ Python tracker <http://bugs.python.org/is

[issue13160] Rename install_dist to install

Alexis Metaireau added the comment: I'm +1 on this. Having an "install" command, on a distribution, makes more sense than having an "install_dist" one, because of course the object is a distribution! -- ___ Python tracker

[issue13160] Rename install_dist to install

Changes by Alexis Metaireau : -- stage: -> needs patch ___ Python tracker <http://bugs.python.org/issue13160> ___ ___ Python-bugs-list mailing list Unsubscri

[issue12779] Update packaging documentation

Alexis Metaireau added the comment: I think it could be helpful to have this work at least somewhere so we can work on top of it. For instance if someone wants to do changes in the documentation, then only thing it would do now would be to cause potential merge conflicts. I'm okay to

[issue14899] Naming conventions and guidelines for packages and namespace packages

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue14899> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue14974] rename packaging.pypi to packaging.index

New submission from Alexis Metaireau : PyPI is the name of a particular index, whereas "index" is a generic term. So ISTM that it would be better to use the latter, semantically-wise. -- assignee: alexis components: Distutils2 messages: 162027 nosy: alexis, tarek priori

[issue14974] rename packaging.pypi to packaging.index

Alexis Metaireau added the comment: We are all calling these indexes quite everywhere. We are talking about a Python Package Index and even the mirroring infrastructure is talking about indexes. Because that's under the "packaging" namespace, it's kind of obvious that thi

[issue14974] rename packaging.pypi to packaging.index

Alexis Metaireau added the comment: The problem is that PyPI isn't a good name since it's the name of ONE index. Westley, do you have any other idea? Otherwise, I think we should stick to "index", which is a good name for something that's named "index" q

[issue12354] packaging.pypi.simple docs use both client and crawler name, which might be confusing

Alexis Metaireau added the comment: > Alexis, you introduced the client/crawler naming after reading some book; > what do you think now? Making the distinction between a crawler and a client is probably not a good idea since it introduces two concepts instead of one simple one. W

[issue14950] Make packaging.install API less confusing and more extensible

Alexis Metaireau added the comment: Attaching a work in progress file which intend to replace the current install.py file. The implementation isn't finished yet but the overall design is here. It comes with four classes: - Installer which manages the overall installation procedur

[issue3754] cross-compilation support for python build

Changes by Alexis Metaireau : -- nosy: -alexis ___ Python tracker <http://bugs.python.org/issue3754> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue1596321] KeyError at exit after 'import threading' in other thread

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue1596321> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue8190] Add a PyPI XML-RPC client module

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue8190> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue4908] Implement PEP 376

Changes by Alexis Metaireau : -- nosy: +alexis ___ Python tracker <http://bugs.python.org/issue4908> ___ ___ Python-bugs-list mailing list Unsubscribe: