Antoine Pitrou added the comment:
It's likely that your global variable gets caught in the traceback attached to
the SystemExit, and that either never gets deallocated, or gets deallocated too
late.
--
nosy: +pitrou
___
Python tracker
New submission from STINNER Victor:
When storing an exception in an asyncio Future object, there is a high risk of
creating a reference cycle. In Python 3, exception objects store a traceback
object which store frame objects. The problem is that a frame can also have a
reference to the
Changes by Barry A. Warsaw ba...@python.org:
--
assignee: - barry
status: open - closed
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue15014
___
Guido van Rossum added the comment:
Doesn't the cycle-detecting GC handle these?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24598
___
___
Changes by Martin Panter vadmium...@gmail.com:
--
nosy: +vadmium
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24598
___
___
Python-bugs-list
STINNER Victor added the comment:
Hm. If the problem is most prominent with 3.3, why mark the issue as 3.6?
Well, I plan to implement this feature in asyncio, so for 3.3-3.6 in fact.
Do you have an implementation already?
Nope, it's more a TODO task for myself :-)
Maybe it can be a 3rd
Ethan Furman added the comment:
Is it too late to have the default for that option be to not allow the
replacement? That would be the safer course.
--
nosy: +ethan.furman
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24597
Guido van Rossum added the comment:
OK, no problem. (Side comment: Future is being subclassed a lot, so
parametrizing its construction may not be so easy.)
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24598
Antoine Pitrou added the comment:
I don't know. I'm assuming some people actually want to redefine existing
specializations.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24597
___
STINNER Victor added the comment:
Doesn't the cycle-detecting GC handle these?
Maybe you are lucky and the GC is able to break the cycle. Maybe you are
unlucky and all objects part of the cycle will never be deleted. Python 3.4 is
better to handle these cases, but Python 3.3 is worse to
Antoine Pitrou added the comment:
Actually, the problem is in PyRun_SimpleFileExFlags(). The executed module is
decref'ed after calling PyErr_Print(), but the latter never returns when the
exception is a SystemExit.
--
nosy: +ncoghlan
___
Python
Roundup Robot added the comment:
New changeset 97a29b86a2dc by Barry Warsaw in branch '3.5':
- Issue #15014: SMTP.auth() and SMTP.login() now support RFC 4954's optional
https://hg.python.org/cpython/rev/97a29b86a2dc
New changeset 2d9003d44694 by Barry Warsaw in branch 'default':
- Issue
Meador Inge added the comment:
I will review this today.
--
assignee: - meador.inge
stage: - patch review
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue23319
___
Ethan Furman added the comment:
Sure. I just saying that
@f.register(int, replace=True)
requires opt-in to replacing, whilst
@f.register(int, replace=False) # don't replace if one already exists
is still prone to bugs.
--
___
Python tracker
On 9-7-2015 11:28, Benj wrote:
Hello,
could you please recommend some python news sites: sites that post news, code
samples, good practices, new libraries... things like that, to follow daily.
Thanks,
Benj
There's a lot of stuff appearing on the Python subReddit;
R. David Murray added the comment:
Does it not work to create an unbuffered makefile object from the socket and
assign that to stdout? Regardless, if you need to set it and restore it around
runcode you can use super() to call the base class method.
I think providing an easier API to change
R. David Murray added the comment:
Yes it is too late. You'd have to do a couple of deprecation cycles to change
the default.
--
nosy: +r.david.murray
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24597
Stefano Mazzucco added the comment:
I have run the minimal example provided on both Python2 and Python3 with the
same results. Sorry if that was not clear.
I did look at issue 1424152 but it seemed to me that I was experiencing a
different problem.
When I try and open the page, I get a squid
Antoine Pitrou added the comment:
Ah, but I wasn't suggesting to add an argument to the .register() call, but to
the singledispatch() call; i.e. it would be a function-wide parameter.
--
___
Python tracker rep...@bugs.python.org
New submission from Stefano Mazzucco:
Hello,
at work, I am behind a proxy (squid) that is only available over http. So, I
have to configure both the http_proxy and https_proxy environment variables to
be something like http://proxy.corp.com:8181;
Now, when I try and use urllib to open an
Ethan Furman added the comment:
Ah, I see.
So you say up-front if you are willing to have redefinition occur later.
That doesn't feel like a consenting-adults attitude, and could also make
testing harder.
I prefer adding an option to the register method, and move towards making the
default
Serhiy Storchaka added the comment:
Perhaps the incref/decref pair ought to be moved into
PyObject_RichCompareBool().
This wouldn't help because key can be used after PyObject_RichCompareBool().
--
___
Python tracker rep...@bugs.python.org
R. David Murray added the comment:
It is not clear from your description if you actually tested it with python3.
In python2, I believe urllib does not support this (see issue 1424152) while
urllib2 does.
Assuming you have, I wonder if the not implemented error is your squid saying
it
Sven R. Kunze added the comment:
... this sounds like it is encouraging staying ignorant.
True. However, I being ignorant about the complexity eventually led to the
development of high-level languages like Python. Each time, a next generation
simply asks the question: 'does it really need to
Hi,
I use pip to install modules and setuptools to install dependencies, and
generate a console_script using the entry_point parameter of setup.
Here is the issue :
my current sources depend on modules, let's say A=1.0, B=1.0, C=2.0. And C
depends on B=1.1
I have no problem with using pip to
Stefan Behnel added the comment:
It's generally worth running the benchmark suite for this kind of optimisation.
Being mostly Python code, it should benefit quite clearly from dictionary
improvements, but it should also give an idea of how much of an improvement
actual Python code (and not
On 07/09/2015 08:05 PM, Chris Angelico wrote:
I thought so at first add well. It looks like the Mailman system handles the
X-No-Archive and/or Archive headers. I couldn't find his name in this
month's archive.
Even if it respects that, there's no way that Mailman can know to
respect his
On 07/09/2015 02:42 PM, Marko Rauhamaa wrote:
Skip Montanaro skip.montan...@gmail.com:
It makes perfect sense to me that TAB and Ctrl-TAB would generate the
same keycode, as TAB is itself a control character (Ctrl-I). As the
Ctrl modifier bit is effectively already set, I don't think you can
Steven D'Aprano added the comment:
On Fri, Jul 10, 2015 at 02:18:33AM +, Gregory P. Smith wrote:
for bytes, \v (0x0b) is not considered a line break. for unicode, it is.
[...]
I think these should be consistent.
I'm not sure that they should. Unicode includes other line breaks which
JohnLeitch added the comment:
Attaching repro.
--
Added file: http://bugs.python.org/file39889/SRE_SEARCH_Integer_Underflow.py
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24602
___
Joshua Harlow added the comment:
Out of curiosity what reference cycles can't be broken in various python
versions? Is it documented/explained anywhere?
--
nosy: +Joshua.Harlow
___
Python tracker rep...@bugs.python.org
On 7/9/2015 8:10 PM, candide wrote:
The official doc explains that :
Python evaluates expressions from left to right.
cf. https://docs.python.org/3.3/reference/expressions.html#evaluation-order
But consider the following snippet :
t=[2020, 42, 2015]
t*(1+int(bool(t.sort(
[42, 2015,
JohnLeitch added the comment:
Attaching proposed patch for unit tests to cover this issue.
--
Added file: http://bugs.python.org/file39888/test_re.py.patch
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24602
On Fri, Jul 10, 2015 at 9:27 AM, Skip Montanaro
skip.montan...@gmail.com wrote:
Also, he should not post here, because all articles posted here are
automatically put on a mailing list and that list's web-accessible
archive.
I thought so at first add well. It looks like the Mailman system
On Fri, Jul 10, 2015 at 10:10 AM, candide c.cand...@laposte.net wrote:
The official doc explains that :
Python evaluates expressions from left to right.
cf. https://docs.python.org/3.3/reference/expressions.html#evaluation-order
But consider the following snippet :
t=[2020, 42, 2015]
New submission from Gregory P. Smith:
for bytes, \v (0x0b) is not considered a line break. for unicode, it is.
this traces back to the Objects/stringlib/ code where unicode defers to the
decision made by Objects/unicodeobject.c's ascii_linebreak table which contains
7 line breaks in the
New submission from JohnLeitch:
The Python 2.7 regular expression module suffers from an integer underflow in
the SRE_SEARCH function of _sre.c, which leads to a buffer over-read condition.
The issue is caused by unchecked subtraction performed while handling
SR_OP_INFO blocks:
if
Changes by Serhiy Storchaka storch...@gmail.com:
--
assignee: - serhiy.storchaka
nosy: +pitrou, serhiy.storchaka
stage: - patch review
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24602
Changes by Raymond Hettinger raymond.hettin...@gmail.com:
Added file: http://bugs.python.org/file39890/intermediary.diff
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24583
___
Stefan Krah added the comment:
Qt had a similar initiative regarding -msse2 -mfpmath:
http://lists.qt-project.org/pipermail/development/2013-December/014530.html
They say that also Visual Studio 2012 has switched to sse2 by default.
The only problem are the Linux distributions that are stuck
David Beazley added the comment:
Wanted to add: I see this as being about the same as having a broken window
pane on the front of Python 3. Maybe there are awesome things inside, but it
makes a bad first impression on anyone who dares to use the interactive console.
--
Benjamin Schubert added the comment:
Ok, sorry for the delay.
I see your point and understand the difficulty of having done right.
Should I close the issue, or propose something else ?
Thanks
--
___
Python tracker rep...@bugs.python.org
Neil Girdhar added the comment:
Copied from closed issue 24240:
Since Grammar/Grammar relies on semantic postprocessing in ast.c,
it would be nice to have an update of the (human readable) Grammar
in the language reference docs.
--
___
Python
Changes by Mark Lawrence breamore...@yahoo.co.uk:
--
components: +Windows
nosy: +paul.moore, steve.dower, tim.golden, zach.ware
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24594
___
Ah - looking at the response headers, they include Transfer-Encoding
chunked - I don't think urlopen handles chunked responses by default,
though I could be wrong, I don't have time to check the docs right now.
The requests library (https://pypi.python.org/pypi/requests) seems to handle
Nick Coghlan added the comment:
The problems with using concurrent.futures directly for running synchronous
tasks in the background are:
1. You have to manage the lifecycle of the executor yourself, rather than
letting asyncio do it for you
2. There's no easy process wide way to modify the
On Thu, Jul 9, 2015 at 5:28 AM, Benj webko...@gmail.com wrote:
Hello,
could you please recommend some python news sites: sites that post news, code
samples, good practices, new libraries... things like that, to follow daily.
Thanks,
Benj
--
Guido van Rossum added the comment:
1. You have to manage the lifecycle of the executor yourself, rather than
letting asyncio do it for you
2. There's no easy process wide way to modify the size of the background task
thread pool (or switch to using processes instead)
But if that's what
We just released devpi-server-2.2.2, devpi-web-2.4.0 and devpi-client-2.3.0,
core parts of the private pypi package management and testing system.
Among the highlights are support for distributed testing with devpi test
--detox,
new status pages at /+status for replica and master sites and
dieter wrote:
It looks strange that you can set s.encoding after you have
called s.get - but, as you apparently get an error related to
the gbk encoding, it seems to work.
Ooo! Sorry, typo - that was outside the function but before the call.
Unfortunately whilst improving my function for
Hello,
could you please recommend some python news sites: sites that post news, code
samples, good practices, new libraries... things like that, to follow daily.
Thanks,
Benj
--
https://mail.python.org/mailman/listinfo/python-list
David Beazley added the comment:
Frivolity aside, I really wish this issue would get more traction and a fix.
Indentation is an important part of the Python language (obviously). A pretty
standard way to indent is to hit tab in whatever environment you're using to
edit Python code.
Yet, at
Mark Lawrence wrote:
If it's provided why have you snipped it this time around? May I most
humbly suggest that the next time you ask, please ensure that you tell
us what you've googled for prior to putting your question.
umm.. I can't determine what you mean by 'snipped it'.
1. I posted a
New submission from Anton Astafiev:
I have a use-case when I need to forward InteractiveConsole through Unix/TCP
socket. Expected implementation:
class InteractiveSocket(InteractiveConsole):
def __init__(self, socket):
self._socket = socket
...
def raw_input(...):
# read from
Nick Coghlan added the comment:
I'll at least write a new python-ideas post, as I realised my original idea
*is* wrong (and you're right not to like it). The focus needs to be on Sven's
original question (How do you kick off a coroutine from otherwise synchronous
code, and then later wait for
Martin Richard added the comment:
Hi,
I would like to update this patch so it can finally land in cpython, hopefully
3.6.
tl;dr of the thread:
In a nutshell, the latest patch from Kristján Valur Jónsson updates
SSLContext.load_cert_chain(certfile, keyfile=None, password=None) and
Antoine Pitrou added the comment:
I thing adding keydata and certdata makes things more complicated, on the
contrary. You start having an API with many optional arguments but some of them
are exclusive with each other (because you can only specify a single key and
cert chain).
The cafile,
New submission from Petr Viktorin:
When this program is invoked as a script (`python reproducer.py`), the __del__
is never called:
---
class ClassWithDel:
def __del__(self):
print('__del__ called')
a = ClassWithDel()
a.link = a
raise SystemExit(0)
---
Raising a different
Martin Richard added the comment:
You are right.
And if certfile and keyfile (args of load_cert_chain()) accept file-like
objects, we agree that cafile (load_verify_location()) should accept them too?
--
___
Python tracker rep...@bugs.python.org
Antoine Pitrou added the comment:
Le 09/07/2015 15:40, Martin Richard a écrit :
And if certfile and keyfile (args of load_cert_chain()) accept
file-like objects, we agree that cafile (load_verify_location()) should
accept them too?
It could, but that's a separate issue. Let's stay focused on
New submission from Antoine Pitrou:
singledispatch currently doesn't defend against unwanted redefinition of an
existing specialization, e.g.:
def f(x): return default
...
f = functools.singledispatch(f)
@f.register(int)
... def _(x): return 1
...
@f.register(int)
... def _(x): return 2
STINNER Victor added the comment:
Sorry, I didn't take time to read the whole discussion. For me, it's a good
idea to accept a filename or a file object in the same parameter. Having two
exclusive parameters for the same thing (ex: CA) doesn't smell like a great API.
--
Christian Heimes added the comment:
I'd rather introduce new types and have the function accept either a string
(for path to fiel) or a X509 object and a PKey object. It's more flexible and
secure. With a private key type we can properly support crypto ENGINEs and wipe
memory when the object
Guido van Rossum added the comment:
Yeah, we should strongly consider writing more documentation before adding more
convenience APIs. Esp. tutorial-style docs, which neither Victor nor I can
supply because we've already moved beyond wizard level ourselves so it's hard
for us to imagine the
Julian Taylor added the comment:
Large objects are just if size 512: return malloc(size) there is no reason it
should be slower.
Also for large objects allocation speed does not matter as much.
--
___
Python tracker rep...@bugs.python.org
Cześć Polska Społeczności Pythona!
Jakiś czas temu uruchomiono PyCon PL 2015 - Call for Proposals, czyli nabór na
propozycje prelekcji, warsztatów, paneli dyskusyjnych oraz innych aktywności
konferencyjnych. Zaakceptowani prowadzący, którzy spełnią warunki Call for
Proposals, otrzymają darmowy
Python Hackers,
PyCon PL 2015 is pleased to announce that its Call for Proposals will be closed
soon!
We encourage you all to come and share experience with a varied audience of
ethusiastic Pythonistas that the conference attracts each year. Talks and
workshops from all Python-related areas
On Fri, Jul 10, 2015 at 6:36 AM, Marko Rauhamaa ma...@pacujo.net wrote:
Chris Angelico ros...@gmail.com:
How do you expect the end result to work? Will it be that your code
imports one version of a module, but other code imports another? You
would have to rename one of them or something.
At
Chris Angelico ros...@gmail.com:
In general, I would expect that B 1.1 is backward-compatible with B
1.0, unless otherwise stated. Why must it be declared in any way other
than the version number?
To make it explicit. The generic component system shouldn't impose
(m)any assumptions on version
On Fri, Jul 10, 2015 at 7:11 AM, Marko Rauhamaa ma...@pacujo.net wrote:
Chris Angelico ros...@gmail.com:
In general, I would expect that B 1.1 is backward-compatible with B
1.0, unless otherwise stated. Why must it be declared in any way other
than the version number?
To make it explicit.
Mark Shannon added the comment:
I still think that this is a good idea, but I would like to see a small speed
test for large objects. Just to be sure that it is no slower.
--
___
Python tracker rep...@bugs.python.org
Chris Angelico ros...@gmail.com:
How do you expect the end result to work? Will it be that your code
imports one version of a module, but other code imports another? You
would have to rename one of them or something.
At work, we have created an analogous component system that has solved
this
On Fri, Jul 10, 2015 at 5:08 AM, Cyril Scetbon cyril.scet...@free.fr wrote:
Forcing my-module to use B=1.1 fixes the issue. However it's just a sample
and my code is using a lot of modules that use other shared modules too. Is
there a way to let dependencies use their own version of the
Guido van Rossum added the comment:
Please move the philosophical discussion to python-ideas.
Regarding the phrasing about the two Future classes being almost compatible,
that is unfortunate wording. Two things can have a similar API (merely having
the same methods etc.) without being
Julian Taylor added the comment:
right at best its probably too insignificant to really be worthwhile, closing.
--
status: open - closed
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue21148
New submission from Grégory Starck:
Consider following:
Python 3.4.0 (default, Jun 19 2015, 14:20:21)
[GCC 4.8.2] on linux
def f(**kw): pass
f(a=1 , )
# ok
f(**{'a': 1} )
# ok
# but :
f(**{'a': 1} , )
SyntaxError: invalid syntax
shouldn't the last form be also allowed as is the
It's just a sample. I'd like to get a general answer. So think about the worst
case.
On Jul 9, 2015, at 21:50, Chris Angelico ros...@gmail.com wrote:
On Fri, Jul 10, 2015 at 5:08 AM, Cyril Scetbon cyril.scet...@free.fr wrote:
Forcing my-module to use B=1.1 fixes the issue. However it's just
On Fri, Jul 10, 2015 at 5:55 AM, Cyril Scetbon cyril.scet...@free.fr wrote:
It's just a sample. I'd like to get a general answer. So think about the
worst case.
(Please don't top-post on this list.)
The worst case is virtually impossible to handle. Somewhere along the
way, you need to say
Julian Taylor added the comment:
any comments on the doc changes?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue23530
___
___
Python-bugs-list
I thought this thread had died down. I guess not quite...
It makes perfect sense to me that TAB and Ctrl-TAB would generate the same
keycode, as TAB is itself a control character (Ctrl-I). As the Ctrl
modifier bit is effectively already set, I don't think you can really set
it a second time and
Skip Montanaro skip.montan...@gmail.com:
It makes perfect sense to me that TAB and Ctrl-TAB would generate the
same keycode, as TAB is itself a control character (Ctrl-I). As the
Ctrl modifier bit is effectively already set, I don't think you can
really set it a second time and be able to
Mark Shannon added the comment:
Indeed there is no *obvious* reason why they should be slower.
But when it comes to optimisation, *never* trust your (or anyone else's)
intuition.
Running a simple check is always worth the effort.
--
___
Python
On Thu, Jul 9, 2015, at 15:36, Tony the Tiger wrote:
On Sun, 05 Jul 2015 20:29:11 +, Stefan Ram wrote:
X-Copyright: (C) Copyright 2015 Stefan Ram. All rights reserved.
Distribution through any means other than regular usenet channels is
forbidden. It is forbidden to publish this
Antoine Pitrou added the comment:
I suppose the simplest fix would be to replace relevant instances of
int(random() * N)
with
min(int(random() * N), N-1)
That sounds simple and generic. It skews the distribution a tiny little bit,
but it doesn't sound significant (perhaps Mark would
Also, he should not post here, because all articles posted here are
automatically put on a mailing list and that list's web-accessible
archive.
I thought so at first add well. It looks like the Mailman system handles
the X-No-Archive and/or Archive headers. I couldn't find his name in this
Guido van Rossum added the comment:
Hm. If the problem is most prominent with 3.3, why mark the issue as 3.6? Do
you have an implementation already? Maybe it can be a 3rd party package rather
than integrated in asyncio debug mode?
--
___
Python
Eric Snow added the comment:
I agree with Antoine.
--
nosy: +eric.snow
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24597
___
___
Chris Angelico ros...@gmail.com:
On Fri, Jul 10, 2015 at 7:11 AM, Marko Rauhamaa ma...@pacujo.net wrote:
Whoever creates B-1.1 ought to make it backward-compatible, but he
should also say so. The majority of developers are careless about
backward-compatibility; having the component system make
On Fri, Jul 10, 2015 at 7:33 AM, Marko Rauhamaa ma...@pacujo.net wrote:
And just how compatible does it have to be to get a tick?
It must be a safe binary replacement of the earlier version. Bug fixes
and new features are ok, but none of the old functionality can be
obsoleted.
Your
Tim Peters added the comment:
It skews the distribution a tiny little bit, ...
But it doesn't - that's the point ;-)
If double-rounding doesn't occur at all (which appears to be the case on most
platforms), absolutely nothing changes (because min(int(random() * N), N-1) ==
int(random() * N)
STINNER Victor added the comment:
Again, why not using only integers?
Pseudo-code to only use integers:
def randint(a, b):
num = b - a
if not num:
return a
nbits = (num + 1).bit_length()
while True:
x = random.getrandbits(nbits)
if x = num:
Martin Panter added the comment:
In the meantime, Issue 24599 has been opened about URLopener(), which I
understand is related to the Python 2 “urllib” half of this bug. Since this
issue has been closed, perhaps it is best to continue discussion of
issue1424152-py27-urllib.diff there instead.
Martin Panter added the comment:
David: the original patch made in Issue 1424152 fixed Python 2’s
urllib.request.urlopen() and Python 2’s urllib2.urlopen(). But Stefano is using
URLopener, which I understand comes from Python 2’s older “urllib” module.
When I run the demonstration, the
Tim Peters added the comment:
Victor, if people want to use getrandbits(), we should backport the Python3
code, not reinvent it from scratch.
Note too Mark's comment: There are several places in the source where
something of the form `int(i * random.random())` is used. The `min()` trick
is
Chris Angelico ros...@gmail.com:
Your descriptions conflict. A safe binary replacement usually cannot
even add new features, in case this breaks something.
Linus Torvalds is adamant about maintaining ABI compatibility across
Linux versions. That hasn't prevented him from accepting numerous new
STINNER Victor added the comment:
Victor, if people want to use getrandbits(), we should backport the Python3
code, not reinvent it from scratch.
Sorry, I don't understand your comment. Backport what? getrandbits() is
available on Python 2 and Python 3, for Random and SystemRandom.
I
The official doc explains that :
Python evaluates expressions from left to right.
cf. https://docs.python.org/3.3/reference/expressions.html#evaluation-order
But consider the following snippet :
t=[2020, 42, 2015]
t*(1+int(bool(t.sort(
[42, 2015, 2020]
Is there not some
Hi all,
I'm part of Plotly, and we've just finished a few releases I thought I'd pass
along.
These tools make it easy to craft interactive graphs and dashboards with D3.js
using Python. We're especially drawn towards matplotlib, pandas, and IPython.
We're still early in building, so any and
Tim Peters added the comment:
Victor, don't ask me, look at the code: the random.choice() implementations in
Python 2 and Python 3 have approximately nothing in common, and the bug here
should already be impossible in Python 3 (but I can't check that, because I
don't have a platform that
On 10/07/2015 01:04, Marko Rauhamaa wrote:
Chris Angelico ros...@gmail.com:
Your descriptions conflict. A safe binary replacement usually cannot
even add new features, in case this breaks something.
New functions in C libraries do not cause runtime breakage.
It's good to know that there's
100 matches
Mail list logo
