Change by Jason R. Coombs :
--
pull_requests: +30349
pull_request: https://github.com/python/cpython/pull/32288
___
Python tracker
<https://bugs.python.org/issue46
Jason R. Coombs added the comment:
Correction. In msg416618, link should have been:
[docs for
zipfile](https://docs.python.org/3/library/zipfile.html#zipinfo-objects)
--
___
Python tracker
<https://bugs.python.org/issue47
Jason R. Coombs added the comment:
> Are the methods expected to raise specific exception types (e.g. if a
> resource is missing, or you call iterdir on a “file”)?
The short answer is no. The protocol does not stipulate specific exceptions.
Perhaps it should be documented that any exce
Jason R. Coombs added the comment:
> Can the argument to `joinpath` contain path separators?
Good question. Currently, I'm aware of three concrete implementations of
Traversable.joinpath:
- pathlib.Path.joinpath
- zipfile.Path.joinpath
- importlib.resources.simple.ResourceContainer.joinp
Change by Jason R. Coombs :
--
assignee: docs@python -> jaraco
___
Python tracker
<https://bugs.python.org/issue47142>
___
___
Python-bugs-list mailing list
Un
Jason R. Coombs added the comment:
My goal with this issue is to simply unblock the use of docstrings in Python
tests. I believe with the work above and the proposed published post, I've
accomplished that goal. I've already spent more time than I'd hoped on this
issue and would like
Change by Jason R. Coombs :
--
pull_requests: +30208
pull_request: https://github.com/python/cpython/pull/32128
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
resolution: -> works for me
___
Python tracker
<https://bugs.python.org/issue46126>
___
___
Python-bugs-list mailing list
Un
Jason R. Coombs added the comment:
I've confirmed that prior to the patch in PR 30194, the location of the failure
was indeed reported. It was just not reported on the same line:
```
cpython main $ git log -1
commit 3e93af0b06cada874c4a16868b6f863b599919f2 (HEAD -> main)
Author: Jaso
Jason R. Coombs added the comment:
It's possible my bad experience may have been avoided through issue35905.
--
___
Python tracker
<https://bugs.python.org/issue46
Jason R. Coombs added the comment:
Aha. If I configure/make without LDFLAGS or CPPFLAGS set, compilation works.
Then I noticed for `LDFLAGS`, some users were using `-L`. I thought it was
slightly odd that my recipe was using `-I` for both flags. How is it that [this
mistake](https
Jason R. Coombs added the comment:
As suggested, I uninstalled and reinstalled everything in homebrew:
$ brew list | xargs brew remove
$ brew install python@3.10 python-launcher python@3.9 python@3.8 gh git
Even after following those steps and setting LDFLAGS and CPPFLAGS to point to
`brew
Jason R. Coombs added the comment:
I did [this
search](https://www.google.com/search?q=gittext+homebrew+"ld%3A+symbol(s)+not+found+for+architecture+arm64"),
which surfaced a few related results.
[This
article](https://lifesaver.codes/answer/pyenv-arm64-builds-confused-by-x86-
Jason R. Coombs added the comment:
SG. Thanks for the advice. I'll dive in and experiment and report back when I
have progress.
--
status: closed -> open
___
Python tracker
<https://bugs.python.org/issu
Jason R. Coombs added the comment:
Well, I'm using a mac with gettext installed as part of `brew install git` with
homebrew installed using the standard procedure.
Only after running `brew remove --ignore-dependencies gettext` and re-running
configure/make did the command build, but doing
Jason R. Coombs added the comment:
And indeed, after removing the `grep ERROR` part of the repro, even the repro
seems to be invalid:
```
cpython main $ ./python.exe -m test.test_importlib -v -k
test_entry_points_unique
test_entry_points_unique_packages
Jason R. Coombs added the comment:
I ran into this same issue. I notice that gettext is a dependency of git, so a
common dependency when developing. Is there perhaps a way that CPython could be
made to ignore gettext on macos Silicon or to detect an incompatible platform
and thus ignore
Jason R. Coombs added the comment:
Inada, you're right. Good catch. I think I missed that behavior from before
because I used `grep ERROR` and when I ran it this time, I just assumed the
output would be the same, but it's clear that even _with descriptions enabled_,
the location of the test
Jason R. Coombs added the comment:
In an attempt to replicate Terry's usage, I added the following patch:
```
diff --git a/Lib/idlelib/idle_test/test_text.py
b/Lib/idlelib/idle_test/test_text.py
index 0f31179e04..be977bbfff 100644
--- a/Lib/idlelib/idle_test/test_text.py
+++ b/Lib/idlelib
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Jason R. Coombs added the comment:
I'm going to close this issue again, as the implementation is now present in at
least a couple of releases. May I suggest that if there are ongoing concerns or
issues to open up a new issue and reference this one and loop me into the
conversation? I'm also
Jason R. Coombs added the comment:
Closing without prejudice. Happy to revisit if there's more information on how
importlib could/should behave differently.
--
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.p
Change by Jason R. Coombs :
--
assignee: -> jaraco
___
Python tracker
<https://bugs.python.org/issue47060>
___
___
Python-bugs-list mailing list
Unsubscrib
Jason R. Coombs added the comment:
Thanks for the report.
Yes, the issues are related, where .version and .name returning None are
specific manifestations of the metadata not having that key and the behavior
being ill-defined.
I haven't yet decided if metadata items being undefined should
Jason R. Coombs added the comment:
I'd advocate for not supporting bytes paths and instead updating the
documentation to require strings.
--
nosy: +jaraco
___
Python tracker
<https://bugs.python.org/issue47
Jason R. Coombs added the comment:
See https://github.com/pypa/setuptools/issues/3007#issuecomment-1068621865
where I did a brief analysis and probable explanation.
What it boils down to: Setuptools needs to supply its own copy of distutils
aggressively. It provides an opt out
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Jason R. Coombs added the comment:
New changeset 177be52517da9a876a3f9e670f88c4731b906986 by Jason R. Coombs in
branch '3.9':
[3.9] bpo-47004: Sync with importlib_metadata 4.11.3. (GH-31854). (GH-31859)
https://github.com/python/cpython/commit/177be52517da9a876a3f9e670f88c4731b906986
Jason R. Coombs added the comment:
New changeset d929aa70e2a324ea48fed221c3257f929be05115 by Jason R. Coombs in
branch '3.10':
[3.10] bpo-47004: Sync with importlib_metadata 4.11.3. (GH-31854). (GH-31857)
https://github.com/python/cpython/commit/d929aa70e2a324ea48fed221c3257f929be05115
Change by Jason R. Coombs :
--
pull_requests: +29957
pull_request: https://github.com/python/cpython/pull/31859
___
Python tracker
<https://bugs.python.org/issue47
Change by Jason R. Coombs :
--
pull_requests: +29955
pull_request: https://github.com/python/cpython/pull/31857
___
Python tracker
<https://bugs.python.org/issue47
Jason R. Coombs added the comment:
New changeset b1e286860742e7ba6fadc75e3ddb6c2899a56919 by Jason R. Coombs in
branch 'main':
bpo-47004: Sync with importlib_metadata 4.11.3. (#31854)
https://github.com/python/cpython/commit/b1e286860742e7ba6fadc75e3ddb6c2899a56919
Change by Jason R. Coombs :
--
keywords: +patch
pull_requests: +29952
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/31854
___
Python tracker
<https://bugs.python.org/issu
New submission from Jason R. Coombs :
Importlib_metadata 4.11.1-3 introduced a few bug fixes. Importantly, 4.11.2
fixed a [serious
defect](https://github.com/python/importlib_metadata/issues/369). Let's
incorporate those fixes into CPython.
--
messages: 415075
nosy: jaraco
priority
Change by Jason R. Coombs :
--
assignee: -> jaraco
___
Python tracker
<https://bugs.python.org/issue47004>
___
___
Python-bugs-list mailing list
Unsubscrib
Jason R. Coombs added the comment:
Aha. I learned how to run commands in the poetry environment... and how to
locate files in that environment. With that, I figured out where the
environment is and where the package metadata is coming from:
```
$ docker run -it @$(docker build -q .) bash -c
Jason R. Coombs added the comment:
The behavior you describe is intentional _and_ deterministic. The library
discovers distributions in the order found based on the search path provided,
with the search path defaulting to sys.path.
The expectation is therefore that the metadata should
Jason R. Coombs added the comment:
I'm in support of adding Path support for sys.path, but I also agree with Eric,
there are innumerable consumers of sys.path beyond importlib. and since
pathlib.Path isn't a str, it would likely introduce incompatibility. On the
other hand, users
R. David Murray added the comment:
The policy is named 'default' because it was intended to become the default two
feature releases after the new email code became non-provisional (first:
deprecate not specifying an explicit policy, next release make default the
default policy and make
Jason R. Coombs added the comment:
I'm pretty sure both EntryPoints and DeprecatedList were introduced in Python
3.10, so 3.9 and 3.8 aren't relevant.
--
versions: -Python 3.8, Python 3.9
___
Python tracker
<https://bugs.python.org/issue46
Jason R. Coombs added the comment:
The tempora library implements a [portable
strftime](https://tempora.readthedocs.io/en/latest/index.html#tempora.strftime).
--
___
Python tracker
<https://bugs.python.org/issue13
New submission from Jason R. Coombs :
Attempting to define a lazy-loaded property for a module, I found [this
guidance](https://stackoverflow.com/a/52018676/70170) referencing [module
attribute
access](https://docs.python.org/3/reference/datamodel.html#customizing-module-attribute-access
Jason R. Coombs added the comment:
Thanks Nikita for the report. I agree, it's inelegant that this property leaks.
Because this code is synced with importlib_metadata, I'm not sure it's worth
the effort of changing it here. Your change here implies a cherry-pick to the
backport and updating
R. David Murray added the comment:
Yeah, I think we need a complete example here.
Note that in the general case there is no such thing as an RFC-valid email in
unicode (which is what python strings are), though with utf8=True and an email
involving only text you might get away with it. I
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Jason R. Coombs added the comment:
New changeset 1514d1252f96e6a83eb65c439522a6b5443f6a1a by Jason R. Coombs in
branch '3.9':
[3.9] bpo-46474: Avoid REDoS in EntryPoint.pattern (sync with
importlib_metadata 4.10.1) (GH-30803). (GH-30828)
https://github.com/python/cpython/commit
Jason R. Coombs added the comment:
New changeset a7a4ca4f06c8c31d7f403113702ad2e80bfc326b by Jason R. Coombs in
branch '3.10':
[3.10] bpo-46474: Avoid REDoS in EntryPoint.pattern (sync with
importlib_metadata 4.10.1) (GH-30803) (GH-30827)
https://github.com/python/cpython/commit
Change by Jason R. Coombs :
--
pull_requests: +29016
pull_request: https://github.com/python/cpython/pull/30829
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
pull_requests: +29015
pull_request: https://github.com/python/cpython/pull/30828
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
pull_requests: +29014
pull_request: https://github.com/python/cpython/pull/30827
___
Python tracker
<https://bugs.python.org/issue46
Jason R. Coombs added the comment:
New changeset 51c3e28c8a163e58dc753765e3cc51d5a717e70d by Jason R. Coombs in
branch 'main':
bpo-46474: Avoid REDoS in EntryPoint.pattern (sync with importlib_metadata
4.10.1) (GH-30803)
https://github.com/python/cpython/commit
Jason R. Coombs added the comment:
New changeset 443dec6c9a104386ee90165d32fb28d0c5d29043 by Jason R. Coombs in
branch 'main':
bpo-46474: Apply changes from importlib_metadata 4.10.0 (GH-30802)
https://github.com/python/cpython/commit/443dec6c9a104386ee90165d32fb28d0c5d29043
Jason R. Coombs added the comment:
New changeset d888ff5381594641126065e78dc9210dae4436a4 by Jason R. Coombs in
branch 'main':
bpo-46425: Partially revert "bpo-46425: fix direct invocation of
`test_importlib` (GH-30682)" (GH-30799)
https://github.com/python/cpyt
Change by Jason R. Coombs :
--
stage: patch review -> needs patch
___
Python tracker
<https://bugs.python.org/issue41682>
___
___
Python-bugs-list mai
Change by Jason R. Coombs :
--
pull_requests: +28989
pull_request: https://github.com/python/cpython/pull/30803
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
pull_requests: +28990
pull_request: https://github.com/python/cpython/pull/30803
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
pull_requests: +28988
pull_request: https://github.com/python/cpython/pull/30802
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
keywords: +patch
pull_requests: +28987
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/30802
___
Python tracker
<https://bugs.python.org/issu
Jason R. Coombs added the comment:
Because I want this security issue to be back-portable to older Pythons, I'll
first apply importlib_metadata 4.10.0 and then apply the change from 4.10.1
separately.
--
___
Python tracker
<ht
New submission from Jason R. Coombs :
Originally reported to the Python Security Response Team, the
EntryPoint.pattern demonstrates a potential
[ReDoS](https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_).
The issue has been patched and fix released
Change by Jason R. Coombs :
--
pull_requests: +28984
pull_request: https://github.com/python/cpython/pull/30799
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
assignee: -> jaraco
___
Python tracker
<https://bugs.python.org/issue46126>
___
___
Python-bugs-list mailing list
Unsubscrib
Jason R. Coombs added the comment:
I've merged the fix for regrtest and I'll explore Terry's concerns and see what
I can devise for those concerns as well.
--
___
Python tracker
<https://bugs.python.org/issue46
Change by Jason R. Coombs :
--
nosy: +jaraco
___
Python tracker
<https://bugs.python.org/issue46425>
___
___
Python-bugs-list mailing list
Unsubscribe:
Jason R. Coombs added the comment:
Closing, presumed fixed. Please re-open if not.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
R. David Murray added the comment:
The general idea is that the string version of the header should contain all of
the original information, but the parsed elements (the things returned by
special header attributes) will contain the valid data, if any. So if the
string version
R. David Murray added the comment:
Note that the parser does attempt to accept obsolete syntax (registering
defects for it), so if there is a bug in the implementation of the obsolete
syntax handling it should be fixed. And yes, there have been other bugs with
whitespace handling
R. David Murray added the comment:
Note also that datetime.now() gives you a naive datetime. From an API
consistency standpoint I think it makes sense that datetime.utcnow() gives a
naive datetime. It would actually be confusing (IMO) for it to return an aware
datetime. I can see why you
Jason R. Coombs added the comment:
Nice reference. Indeed, the
[rationale](https://www.python.org/dev/peps/pep-0533/#id15) of that pep gives a
similar example and the
[background](https://www.python.org/dev/peps/pep-0533/#id3) describes the
problem with the solution I've drafted above
Jason R. Coombs added the comment:
Hi Eric. I did mention that option in my report, but that option requires
loading the whole file into memory. I'd like something equivalent to the
iterator that `open()` provides, which yields lines lazily.
Serihy, thanks for the feedback. I do indeed
New submission from Jason R. Coombs :
I'd like to be able to do something pretty fundamental: lazily load lines from
a file in a single expression.
Best I can tell, that's not possible in the language without triggering
warnings.
One can use 'open' but that triggers a ResourceWarning
Jason R. Coombs added the comment:
I don't have a good answer, but given the title of this issue (which is
specifically scoped to site install schemes), I'm tempted to say we should deal
with prefixes in a separate, perhaps broader issue, and there address the
reported issue (that a user's
Jason R. Coombs added the comment:
Today I learned something. Thanks Arie. Yes, I agree that's a mistake. Perhaps
the test suite should also have a test to capture the missed expectation (that
__dict__ should not be present or setting attributes on EntryPoints instances
should fail
Jason R. Coombs added the comment:
Perhaps it is a mistake. The `__slots__` were added as a (possible premature)
optimization in [this
conversation](https://github.com/python/importlib_metadata/pull/278/files#r565475347).
It's not obvious to me what the danger is in defining __slots__
Jason R. Coombs added the comment:
I'd recommend not to block on issue24132. It's not obvious to me that
subclassing would be valuable. It depends on how it's implemented, but in my
experience, zipfile.Path doesn't and cannot implement the full interface of
pathlib.Path. Instead
Jason R. Coombs added the comment:
> > I presume I don't need to explain why docstrings are nice and preferable
> > over comments.
> Actually, can you?
I searched around and didn't find any good treatise or thorough overview of
reasons _why_ docstrings should be preferred,
Change by Jason R. Coombs :
--
status: open -> closed
___
Python tracker
<https://bugs.python.org/issue46118>
___
___
Python-bugs-list mailing list
Unsubscrib
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Jason R. Coombs added the comment:
New changeset 35628e4cde71e54afe12aea50c74069afe2c3389 by Jason R. Coombs in
branch 'main':
bpo-46118: Make sure importlib.resources is included. (GH-30311)
https://github.com/python/cpython/commit/35628e4cde71e54afe12aea50c74069afe2c3389
Jason R. Coombs added the comment:
I encountered this issue again today in issue46118.
I started looking into creating the patchcheck, but I realize it may be a
little tricky to detect the introduction of a new folder, because `git diff`
doesn't actually report new folders, and my initial
Jason R. Coombs added the comment:
And the underlying cause is reported in issue37043.
--
___
Python tracker
<https://bugs.python.org/issue46118>
___
___
Pytho
Jason R. Coombs added the comment:
Looks like I encountered the same foot gun when originally introducing the
importlib.metadata package (issue34632).
--
stage: patch review -> resolved
___
Python tracker
<https://bugs.python.org/issu
Change by Jason R. Coombs :
--
pull_requests: +28525
stage: resolved -> patch review
pull_request: https://github.com/python/cpython/pull/30311
___
Python tracker
<https://bugs.python.org/issu
Jason R. Coombs added the comment:
Installed buildbots are failing (like this one:
https://buildbot.python.org/all/#/builders/350/builds/1164). I'm going to see
if it's a simple fix like updating the makefile or if a rollback is called for.
--
status: closed -> o
Jason R. Coombs added the comment:
New changeset 99945c6b5cf280bd90075cffae942af44941abcc by Jason R. Coombs in
branch 'main':
bpo-46109: Separate out files relating to importlib.resources (GH-30160)
https://github.com/python/cpython/commit/99945c6b5cf280bd90075cffae942af44941abcc
Change by Jason R. Coombs :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Jason R. Coombs added the comment:
New changeset e712a5b277866a71c195f38c1b5d87d9126dba3e by Jason R. Coombs in
branch 'main':
bpo-46118: Move importlib.resources to its own package. (#30176)
https://github.com/python/cpython/commit/e712a5b277866a71c195f38c1b5d87d9126dba3e
Jason R. Coombs added the comment:
Normalize_path from legacy implementation:
https://github.com/python/importlib_resources/blob/3beb2fd5831e65f7b45033e1ec276c4a6b4ca973/importlib_resources/_legacy.py#L30-L40
--
___
Python tracker
<ht
Jason R. Coombs added the comment:
> Does `joinpath` have less validation?
Yes. Previously, resources.* would perform some validation on the path to
ensure that it didn't contain path separators (to avoid users attempting to get
resources in subdirectories or perhaps manipulating the p
Jason R. Coombs added the comment:
I realize, I can limit the scope of this issue to address 'resources' alone,
leaving ABCs where they are except for resource-related ones, allowing us to
defer a larger refactoring to a separate issue/effort.
--
title: Separate resources and abc
Jason R. Coombs added the comment:
I recently noticed that some docs already refer to `importlib.resources.abc`
even though it doesn't currently exist
(https://github.com/python/cpython/blame/main/Doc/library/importlib.rst#L948).
That finding leads me to adjust my expectation that perhaps
Jason R. Coombs added the comment:
As I'm exploring this issue, I notice that currently, the docs refer to
:class:`importlib.resources.abc.Traversable` (even though that doesn't
currently exist)
(https://github.com/python/cpython/blame/main/Doc/library/importlib.rst#L948
Jason R. Coombs added the comment:
> a ToC at the top of the main page which identifies the submodules (and
> provides a brief summary for each) would be extra helpful.
I've added this in the latest commit (1adefaf552).
--
___
Python t
Jason R. Coombs added the comment:
The reason I wanted to consider issue46118 first was because I wanted to
explore whether the ABCs from `importlib_resources.abc` should be _documented_
as `importlib.resources.abc` or `importlib.abc` (as they are now).
After filing the issue and exploring
Change by Jason R. Coombs :
--
assignee: jaraco -> p-ganssle
___
Python tracker
<https://bugs.python.org/issue46124>
___
___
Python-bugs-list mailing list
Un
Change by Jason R. Coombs :
--
keywords: +patch
pull_requests: +28415
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/30194
___
Python tracker
<https://bugs.python.org/issu
Jason R. Coombs added the comment:
After some investigation (and tracing calls through a dozen or more layers), I
found that Python's own regression tests can disable this behavior thus:
```
diff --git a/Lib/test/support/testresult.py b/Lib/test/support/testresult.py
index 2cd1366cd8
Jason R. Coombs added the comment:
It looks like that 'descriptions' attribute is passed through from
TextTestRunner, which sets it to True by default
(https://github.com/python/cpython/blob/9b52920173735ac609664c6a3a3021d24a95a092/Lib/unittest/runner.py#L163).
It seems that behavior isn't
Jason R. Coombs added the comment:
Looks like the code, I believe here's where the reporting happens:
https://github.com/python/cpython/blob/9b52920173735ac609664c6a3a3021d24a95a092/Lib/unittest/runner.py#L48-L51
I see a "description" property there that may already provide t
1 - 100 of 12043 matches
Mail list logo
