R. David Murray added the comment:
In case it wasn't clear: I seriously doubt that iPython is hanging due to the
warning, I think something else must be happening and the warning is a red
herring.
--
___
Python tracker <rep...@bugs.python.
R. David Murray added the comment:
If a warning is causing iPython to hang, there is something seriously wrong
with iPython. A warning is just a message written to stderr, it doesn't affect
the execution of the program.
The json module does not natively support datetime, so whatever
R. David Murray added the comment:
Numpy is not part of the python standard library. You should report this issue
to the numpy bug tracker, which appears to be here:
https://github.com/numpy/numpy/issues. Or perhaps first ask the numpy
community if this is really a bug: the new version
R. David Murray added the comment:
Sorry, but I no longer have access to that application (I'm a consultant, and
the owner is no longer a client).
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
I've worked on an application (proprietary, unfortunately) that created a lot
of empty dictionaries that only sometimes got populated. It involved
sqlalchemy, but I don't remember if the dicts came from sqlalchemy itself or
from the code that used
R. David Murray added the comment:
Well, ETL is semi-standardized. Try dealing with csv files exported from excel
spreadsheets written by non-programmers :)
"e"X is not a quoting the csv module will produce, but I don't think it is a
csv error. insofar as csv has a standard, it is
R. David Murray added the comment:
In my experience CSV files with fields with embedded newlines are pretty
common. I don't really think we want to support invalid CSV files.
--
nosy: +r.david.murray
___
Python tracker <rep...@bugs.python.
R. David Murray added the comment:
There is, however, an issue that if you pass a message with the default policy
to the generator and specify SMTP as the policy, it doesn't *recode* the line
endings. I thought there was an open issue for that, but I can't find it.
One solution would
R. David Murray added the comment:
Huh. I ran something like that test and thought I saw the reverse. I guess I
misread my terminal. Looking at the code, set_content does take care to fix
the line ending according to the policy before doing the encoding.
--
resolution: ->
R. David Murray added the comment:
The API exists in python3.5 and python3.4 as well, it was just provisional.
Very few things changed between the provisional version and the final version
in 3.6.
--
___
Python tracker <rep...@bugs.python.
Changes by R. David Murray <rdmur...@bitdance.com>:
--
components: +email
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue30032>
___
Changes by R. David Murray <rdmur...@bitdance.com>:
--
versions: -Python 2.7
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python
R. David Murray added the comment:
That is true for text/ types, yes. The policy is named after the target
wire protocol, and if you are transmitting an email message over SMTP, that
implies MIME. What to do if you are not sending it over SMTP, though, is a
tougher question. One could
R. David Murray added the comment:
Actually, I think the fix would go in the generator, not in the contentmanager,
but it's been long enough since I've worked on the code that I'm not sure.
--
___
Python tracker <rep...@bugs.python.org>
R. David Murray added the comment:
This appears to be a problem in the new API as well. I don't think we can
change the legacy API because its been that way forever and applications might
be depending on it (that is, the library preserves exactly what it is handed,
and an application might
R. David Murray added the comment:
Yeah, I was wondering if part of the demo was to show something that can be
done with no library support...but that probably isn't the case.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.p
R. David Murray added the comment:
Yes, this sub-optimal, but it's the way it works in the legacy API, and we
aren't going to change the legacy (compat32) API at this point. The new
policies and the new API in python3 handle this sensibly.
--
resolution: -> out of date
st
R. David Murray added the comment:
I'm not sure about using argparse. Currently the demo uses no imports. I'm
not strongly against, though.
Did you mean __init__ instead of __new__? Also, its value could be made True
and False instead of 0 and 1. (Which tells you how old this demo
R. David Murray added the comment:
This is a clever idea, but I vote -1 for this proposal. I think it makes
attrgetter more complex for little purpose. The fact that only some attribute
names work and the others get mangled makes the API very ugly and not, IMO,
desirable.
Finally, if you
R. David Murray added the comment:
I suspect you just need to add pickle support to your class. When I subclassed
str in the email package, I found I needed to do that. I'd have to go through
the docs again to remember how the code works, but you can take a look at the
BaseHeader class
New submission from David E. Franco G.:
wandering for the internet I fount some unicode character in a random comment,
and just for curiosity I wanted to use python (3.6.1) to see their value, so I
copy those characters and paste them in IDLE, and in doing so it just close
without warning
R. David Murray added the comment:
I don't think this is really a bug, I think it's a consequence of the different
byte/string models of python2 and python3 coupled with the different
binary/text models of posix and windows.
--
nosy: +r.david.murray
R. David Murray added the comment:
Actually, looking at the issue related to the patch, we conferred at the time,
Barry, and decided on no backports. It was applied only to default. Sijian:
the reason we put the issue number in the commit message is because the issue
often contains relevant
R. David Murray added the comment:
I consciously decided not to backport this to 2.7 at the time, though I'm not
sure I said that out loud. I think it is too much of a behavior change for 2.7.
--
resolution: -> rejected
stage: -> resolved
status: open -&g
R. David Murray added the comment:
I'm not surprised that trying to render a message parsed with 'headersonly'
fails. headersonly treats the entire message body as a single string payload.
I'm not sure what the correct behavior should be for the email package, but the
fact that this doesn't
R. David Murray added the comment:
The documentation is technically correct, as far as I can see. Issue 8743 is
not about disallowing certain comparison operations, but rather incorrectly
implementing the earlier sentence in that same doc section: "If the comparison
is undefined, it
R. David Murray added the comment:
Agreed. Time to close this.
--
resolution: -> not a bug
stage: needs patch -> resolved
status: open -> closed
___
Python tracker <rep...@bugs.python.org>
<http://bugs.py
R. David Murray added the comment:
If all connections fail (not file :)
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue29980>
___
__
New submission from R. David Murray:
create_connection will try multiple times to connect if there are multiple
addresses returned by getaddrinfo. If all connections file it inspects the
exceptions, and raises the first one if they are all equal. But since the
addresses are often different
R. David Murray added the comment:
I don't see any omission, myself. Keep in mind that the language reference is
as much or more of a specification as it is a reference, so we tend to try to
use the minimum language that precisely describes the expected behavior. Which
is why I suggested
R. David Murray added the comment:
I think you meant "the language reference" rather than "the devguide". The
sentence about the comment is redundant with the preceding line that says that
the thing that results in a join is a physical line that ends with a backslash
R. David Murray added the comment:
I also have no idea what your comment about stripping white space is in
reference to ;)
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
Yes, that's exactly right. 'time' is a low-level os-function wrapper, and
inherits many of the deficiencies of the platform. datetime attempts to be a
more comprehensive, portable solution. (But even it has its quirks...timezones
are *hard
R. David Murray added the comment:
Yes, the difficulty in renaming the parameter was why I suggested a doc change
only. I'm not sure it it is worth it to go through a deprecation cycle for
socketserver to change the name, though it certainly would be nice. Martin and
I could make
R. David Murray added the comment:
Given how old socket server is, and that it doesn't actually require that API,
we should probably just reword the documentation so that it is clear that
RequestHandlerClass is the default Handler implementation (and that you can
work with setup/handle/finish
R. David Murray added the comment:
Well, we could document it as a factory argument, and explain that the argument
name is an historical artifact.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
I think I'm missing something here. What prevents one from passing a factory
function as the RequestHandlerClass argument? In Python, a class name *is* a
factory function for class instances.
--
nosy: +r.david.murray
R. David Murray added the comment:
If you want to be completely unambiguous, you could say "keyword argument
names". "keyword argument" appears to mean different things in different
contexts; sometimes it means the name and the value together, sometimes one or
the other
R. David Murray added the comment:
Agreed with Brett. What I was trying to say was that we aren't going to get
people to change to a different term, nor are the other 'x-string'
abbreviations interesting, so we should document just f-string. But I wasn't
exactly clear that that was my point
R. David Murray added the comment:
Oops. I merged the patch without coming back here first :(. Still getting
used to the new workflow.
It turns out that format has a parameter, monetary, that isn't supported by
format_string. So what we did was add that parameter to format_string
R. David Murray added the comment:
New changeset 1cf93a76c2cf307f2e1e514a8944864f746337ea by R. David Murray
(Garvit Khatri) in branch 'master':
bpo-10379: add 'monetary' to format_string, deprecate format
https://github.com/python/cpython/commit/1cf93a76c2cf307f2e1e514a8944864f746337ea
R. David Murray added the comment:
Thanks,
--
stage: patch review -> backport needed
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.or
Changes by R. David Murray <rdmur...@bitdance.com>:
--
stage: backport needed -> needs patch
___
Python tracker <rep...@bugs.python.org>
<http://bugs.pyt
Changes by R. David Murray <rdmur...@bitdance.com>:
--
stage: needs patch -> backport needed
___
Python tracker <rep...@bugs.python.org>
<http://bugs.pyt
R. David Murray added the comment:
New changeset 0ae7c8bd614d3aa1fcaf2d71a10ff1148c80d9b5 by R. David Murray (Amit
Kumar) in branch 'master':
bpo-16011 clarify that 'in' always returns a boolean value
https://github.com/python/cpython/commit/0ae7c8bd614d3aa1fcaf2d71a10ff1148c80d9b5
R. David Murray added the comment:
"raw" and "byte" are one syllable names, and thus as easy and more meaningful
to say than "r-string" or "b-string". "unicode string" is more descriptive and
not much longer, but "u-string" does occa
R. David Murray added the comment:
His problem is that the file has already been deleted by the time the
subprocess returns, so the unlink is going to raise an exception.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.p
R. David Murray added the comment:
He's still going to get an error using your code, Christian. But if he knows
that the file being gone is OK, he can catch and ignore the error. Having exit
do the unlink wouldn't help him; in that case he'd have to wrap the whole
'with' clause in a try
R. David Murray added the comment:
This is a specific example of the general problem reported in issue 25538.
--
nosy: +r.david.murray
resolution: -> duplicate
stage: -> resolved
status: open -> closed
superseder: -> Traceback from __exit__ method i
R. David Murray added the comment:
See also issue 29922.
--
nosy: +Tadhg McDonald-Jensen, yselivanov
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
Looking over the PR, and especially in the context of Serhiy's point about this
being about 'deep copy' and not 'deepcopy', I think this would be clearer if it
were shortened even further, to just:
"Because deep copy copies everything, it may copy
R. David Murray added the comment:
Let's retitle this, then.
--
nosy: +r.david.murray
title: string.Template: Add PHP-style variable expansion example ->
string.Template: Rewrite docs to emphasize i18n use case
___
Python tracker &
R. David Murray added the comment:
Oops, I didn't meant to close this.
--
resolution: not a bug ->
stage: resolved ->
status: closed -> open
___
Python tracker <rep...@bugs.python.org>
<http://bugs.pyt
R. David Murray added the comment:
Yes, that looks wrong to me. IMO it should be returning a new function object,
not updating the __code__ of the existing object. I couldn't figure when that
is actually triggered, though.
There are also some other oddnesses, given the definition
R. David Murray added the comment:
Agreed with Terry. The general policy in Python is that we let errors bubble
up unless there is a good reason to do something else with them. And errors
that bubble up are not, in general, documented. (In short, Python is not Java
:)
--
nosy
R. David Murray added the comment:
On the other hand, the current mixed case sending found a bug in your code, so
it has some value. I'm neither in favor nor in objection to the change, at
this point.
--
___
Python tracker <rep...@bugs.python.
R. David Murray added the comment:
It is interesting that in all the years smtplib has been in use, this is the
first time (as far as I know) this has been reported as a problem. I don't see
any reason to object to changing it to send the commands in upper case, but the
server you
R. David Murray added the comment:
Decorators are called with the decorated *function* objection when the class is
compiled. There can be no instance involved by their very nature, since the
instance doesn't exist yet. So no, you can't have a decorator that affects
instance attributes
R. David Murray added the comment:
A warning is not appropriate (we reserve those for things that are security
related, pretty much). A sentence might be, though. For example, we could
change the initial discussion of the run function to say:
This does not capture stdout or stderr
R. David Murray added the comment:
Easier, but if we do it in the tool, then it is done for everyone and they
don't *each* have to spend that "less time" writing their own script. And
--indent and --compact are both useful for debugging/hand testing, since it
allows you to generate
R. David Murray added the comment:
The cross check test itself doesn't depend on a regeneration, but it does
depend on the information in token.h. Meanwhile the validity of *that* is
checked by regeneration in your test_token tests. This is exactly what I had
in mind :).
So, I haven't done
David Mertz added the comment:
Raymond wrote:
> The idea is that the method would return a new counter instance
> and leave the existing instance untouched.
Your own first example suggested:
c /= sum(c.values())
That would suggest an inplace modification. But even i
New submission from R. David Murray:
http://bugs.python.org/issue24622 made reminded me that a while back we added
tests for the keyword module that includes a test that if you run it, you get
the result that is checked in. The same thing could be done for the token.py
module. And then we
David Mertz added the comment:
I definitely wouldn't want a mutator that "normalized" counts for the reason
Antoine mentions. It would be a common error to normalize then continue
meaningless counting.
One could write a `Frequency` subclass easily enough. The essential feature i
R. David Murray added the comment:
I don't see how adding a constant increases the complexity of the API.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
DAVID ALEJANDRO Pineda added the comment:
Hello Again.
The problem can be replicated in the same structure when call the
'functools.partial' feature
I wrote a simple example. It uses the asyncio and multiprocessing structure.
https://github.com/dpineiden/async_multiprocessing
David MacIver added the comment:
So it does. My apologies. I'll close this.
--
stage: -> resolved
status: open -> closed
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python
New submission from David MacIver:
When run under Python 3.6.0 or 3.5.1 (and presumably other versions of Python
3) the attached code fails with "Fatal Python error: Cannot recover from stack
overflow." then aborts with a core dump and an error code indicating it got a
SIGABRT.
On
R. David Murray added the comment:
OK. This looks good to me. I haven't figured out the new commit process,
though (like how to do misc news and backports), so I'm not going to be the one
to merge it, I'm afraid. At least not until I do find time to learn
R. David Murray added the comment:
So what happens when you do that same operation in 3.5/6 with your change in
place? Does the behavior change? (I haven't looked back at the code to see if
I think it will :)
--
___
Python tracker <
R. David Murray added the comment:
Thanks for the PR. However, rereading this: since compat32 is providing
backward compatibility with the behavior of the python 3.2 email package, we
need to check what it would do in this situation before changing the behavior.
What we may need instead
R. David Murray added the comment:
If you use smtplib.send_message in python3, it will do what you want (including
stripping BCC headers before sending the message).
If someone wants to create a PR to add an example of what Eric is talking about
(specifying additional senders
R. David Murray added the comment:
See msg253287. Someone should check the RFC. It is not obvious that just
encoding using utf8 is correct; fundamentally passwords are binary data. But
the auth methods don't currently accept binary data. UTF8 is a reasonable
default these days, I think
david added the comment:
I'm sorry I rushed my comment. Same thing happens on line 604
return encode_base64(s.encode('ascii'), eol='')
changing both from 'ascii' to 'utf-8' works for me.
--
___
Python tracker <rep...@bugs.python.org>
New submission from david:
Trying to use unicode passwords on smtplib fails miserably on python3.
My particular issue arises on line 643 of said library:
(code, resp) = self.docmd(encode_base64(password.encode('ascii'), eol=''))
which obviously dies when trying to handle unicode chars
R. David Murray added the comment:
This has already been discussed and rejected (issue 14403). In practice the
distinction between a failure and an error is not useful, and a comment in the
test is IMO clearer than a no-op context manager: you can use a positive
sentence instead
R. David Murray added the comment:
Thanks for the report and PR, but your fix is not obviously correct. In
general exceptions are the way in Python that errors are reported, and asking
for extra_info on a closed stream is an error. The exception raised is not
clear, so perhaps the asyncio
R. David Murray added the comment:
If Raymond is on the side of skipping the deprecation than I'm good with it.
Like I said, this is a marginal case.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
I disagree, I think a __json__ protocol is sensible. But this is why it needs
to be discussed on python-dev or python-ideas first :) In the meantime adding
deque support like we added enum support is reasonable, but IMO we shouldn't go
to crazy adding
David Robins added the comment:
I saw a similar error with Python 3.6 on a MIPS (32-bit -
mipsisa32r2el-axis-linux-gnu) platform, but during interpreter startup, not
install (perhaps because it was cross-compiled so install on the host doesn't
run the target Python). It was due to building
David E. Franco G. added the comment:
I found the problem, it was a test.pyc file that was in my personal folder...
once deleted it work perfect
the build-in test module should have some other name...
--
___
Python tracker <rep...@bugs.python.
R. David Murray added the comment:
Have you tried '-' plus any other character? argparse treats '-' and '--'
specially, and this is a known issue.
--
nosy: +r.david.murray
___
Python tracker <rep...@bugs.python.org>
<http://bugs.p
David E. Franco G. added the comment:
Ok, Done.
that work in both 3.5 and 3.6
I also did the
python -m idlelib.ColorDelegator
and
python -m idlelib.colorizer
and they turn ok
I also open those modules in their respective idle and run them, ColorDelegator
work ok but colorizer
R. David Murray added the comment:
It works fine for me. If I write the data to a file (using print) and look at
it with vi, I see your expected string with <81> on the end. It also works
fine in my console (which otherwise produces mostly unknown character glyphs;
I'm using a utf8
David E. Franco G. added the comment:
>I presume you would like 'async' and 'await' highlighted as keywords.
Yes.
Some others tools like IPython and the Spider IDE that come with the Anaconda
package already highlighted them as keywords, even if you can use them as
normal variab
R. David Murray added the comment:
Correct. Tests are good, it's the fix I was rejecting.
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
New submission from David E. Franco G.:
Well, this is pretty self explanatory, when playing with this new features of
async and await (https://docs.python.org/3.5/whatsnew/3.5.html#new-features) I
found to me surprise that there is no syntax highlighting for it in the IDLE
for py3.5 and also
R. David Murray added the comment:
Thanks for the response, but I do not believe that this is a bug.
The python3 email package will decode the headers automatically if you use the
new policies, so if you iterate through the headers, you'll get the decoded
versions, with access to the parms
R. David Murray added the comment:
So these will become positional only? In that case I'd say "Using 'x' as a
keyword argument is deprecated; specify the value as a positional argument
instead"
--
___
Python tracker <rep...@bugs.pyth
R. David Murray added the comment:
I don't think we should do it without deprecation, since it could break working
code. But it certainly sounds like a marginal case: I doubt there is *much*
code that uses the existing keyword names.
--
nosy: +r.david.murray
R. David Murray added the comment:
This almost qualifies as a FAQ :)
In 2.7 (and in 3.x, though we've added some additional platform independent
stuff there), strptime supports what the platform supports, and this is
documented: "The following is a list of all the format codes that
R. David Murray added the comment:
We don't make breaking changes unless there is strong motivation, which is
lacking here.
--
nosy: +r.david.murray
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
R. David Murray added the comment:
Agreed with Christoph. If the wxPython team finds there really is a bug in
cpython itself triggering this, they (or you) can open a new issue with
specifics.
--
nosy: +r.david.murray
resolution: -> third party
stage: -> resolved
status
R. David Murray added the comment:
Ah, that should all be implementable from your python shell, then, no need for
support in the os module.
For reference, when I said 'cd man page', this is what I was referring to:
http://www.unix.com/man-page/posix/1posix/cd/
I'm going to close this issue
R. David Murray added the comment:
Since yours is the itch, I'm afraid you are going to have to be the one to
figure out how this could be implemented :)
--
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/i
Changes by R. David Murray <rdmur...@bitdance.com>:
--
type: behavior -> enhancement
___
Python tracker <rep...@bugs.python.org>
<http://bugs.pyt
R. David Murray added the comment:
The function is chdir, not cd. The 'cd' man page says that 'cd -P' should
"perform actions equivalent to the chdir() function".
So, you are asking for a different function, which should *not* be named
'os.chdir'. You'll have to make a case fo
R. David Murray added the comment:
The obvious thing to do would be to make the Action subclasses have public
names. I personally would continue to use the strings, though, as they are
easier to type. (The same would be true if an enum were introduced (that I'd
continue to use the stings
New submission from David Ellis:
Related:
https://github.com/python/cpython/pull/242
https://bugs.python.org/issue29623
In the discussion over my PR for bpo-29623 it became clear that bytestring
paths were also unsupported and it was suggested that that should be a separate
issue (currently
Changes by David Ellis <ducks...@gmail.com>:
--
pull_requests: +205
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue29623>
___
_
1601 - 1700 of 12771 matches
Mail list logo
