New submission from Gerrit Holl :
In Python 3.9.4, in a function call, when assigning a comprehension to a
keyword argument, and the comprehension contains a spurious colon (if a list or
set comprehension) or a missing value (when a dict comprehension), the syntax
error message gives a hint
New submission from Gerrit Holl :
When there is a syntax error in a multi-line f-string, the arrow in the
reported syntax error points to the wrong line:
$ cat mwe.py
s = ("apricot "
"pineapple"
f"shallot{")
$ python mwe.py
File &quo
New submission from Gerrit Holl :
Pythons commandline warning filter cannot currently handle custom warning
categories. See https://bugs.python.org/issue22543
I tried a workaround as suggested on Stack Overflow
<https://stackoverflow.com/a/53538033/974555> to filter on a warning ca
New submission from Gerrit Holl :
I am unable to use `pdb` to debug a problem I have with the `python-hdf4`
installer. The exception in the program to be debugged is printed twice,
followed by an exception in pdb itself, ending with `ValueError: underlying
buffer has been detached`. See
Gerrit Holl added the comment:
There is a thread on StackOverflow related to this problem:
https://stackoverflow.com/q/42495641/974555
The (currently only and accepted) answer proposes as a workaround to set the
PYTHONPATH variable, as the contents of those apparently *are* already
New submission from Gerrit Holl :
In the [`enum`
module](https://docs.python.org/3/library/enum.html#supported-sunder-names)
documentation, some of the `_sunder_` names are on `EnumMeta`, whereas others
are on the produced `Enum` class:
> 8.13.15.3.2. Supported `_sunder_` names
> * `
Changes by Gerrit Holl <gerrit.h...@gmail.com>:
--
title: `catch_warnings` context manager should reset warning registry to
previous state upon exiting, to prevent warnings from being reprinted ->
`catch_warnings` context manager causes all warnings to be printed every ti
Gerrit Holl added the comment:
I get bitten by this frequently, and find myself patching many libraries just
to avoid them from calling .catch_warnings.
Does anyone know whether to fix this it would suffice to edit warnings.py, or
would I need to dig into _warnings.c as well
New submission from Gerrit Holl:
The OO API allows to create empty enums, i.e. without any values. However, the
functional API does not, as this will result in an IndexError as shown below:
In [1]: import enum
In [2]: class X(enum.IntFlag): pass
In [3]: Y = enum.IntFlag(&q
Gerrit Holl added the comment:
I believe this fix causes this bug: http://bugs.python.org/issue29672
--
nosy: +Gerrit.Holl
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/
Changes by Gerrit Holl <gerrit.h...@gmail.com>:
--
pull_requests: +304
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue29677>
___
_
Gerrit Holl added the comment:
>>> round(21345, -2)
21300
Desired and useful to me.
--
nosy: +Gerrit.Holl
___
Python tracker <rep...@bugs.python.org>
<http://bugs.pyt
Gerrit Holl added the comment:
To resolve this, the `catch_warnings` context manager upon exiting should not
only reset `filters`, but also `_filters_version`, and perhaps an associated
dictionary? Not sure if I'm understanding the code in `warnings.c` correctly,
and whether, for this change
Changes by Gerrit Holl <gerrit.h...@gmail.com>:
--
title: `catch_warnings` context manager causes warnings to be reprinted ->
`catch_warnings` context manager should reset warning registry to previous
state upon exiting, to prevent warnings from being
Gerrit Holl added the comment:
To clarify, this behaviour crosses module boundaries. So even if some piece of
code many layers deep buried in a module uses this context manager, it has
global consequences.
--
___
Python tracker <
Gerrit Holl added the comment:
I suppose this is a consequence of the change in:
http://bugs.python.org/issue4180
and although I can see why the warnings filter would be reset when entering the
context manager, I don't think it is desirable to have side effects for modules
that are entirely
New submission from Gerrit Holl:
Entering the `catch_warnings` context manager is causing warnings to be printed
over and over again, rather than just once as it should. Without such a
context manager, the behaviour is as expected:
$ cat ./mwe.py
#!/usr/bin/env python3.5
New submission from Gerrit Holl:
I believe this page is not supposed to exist:
https://docs.python.org/3.6/whatsnew/3.7.html
Nor this one:
https://docs.python.org/3.5/whatsnew/3.6.html
Neither are maintained, nor would I expect them to be.
--
assignee: docs@python
components
New submission from Gerrit Holl:
I am building and testing Python 3.6 on the JASMIN Analysis Platform
<http://www.jasmin.ac.uk/services/jasmin-analysis-platform/>, which runs Red
Hat Enterprise Linux Server release 6.8 on a machine with 48 × Intel(R) Xeon(R)
CPU E7-4860 v2 @ 2.60GHz,
Changes by Gerrit Holl <gerrit.h...@gmail.com>:
--
versions: +Python 3.6
___
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue29155>
___
Gerrit Holl added the comment:
I get the same failure cause in test_pdb and test_zipimport_support, also in
situations with tests based on doctest, again getting an unexpected `***
ModuleNotFoundError: No module named 'IPython'`
--
___
Python
New submission from Gerrit Holl:
I am building and testing Python 3.6 on the JASMIN Analysis Platform
<http://www.jasmin.ac.uk/services/jasmin-analysis-platform/>, which runs Red
Hat Enterprise Linux Server release 6.8 on a machine with 48 × Intel(R) Xeon(R)
CPU E7-4860 v2 @ 2.60GHz,
Gerrit Holl added the comment:
I experience this problem when trying to build/test Python 3.6 on the [JASMIN
Analysis Platform](http://www.jasmin.ac.uk/services/jasmin-analysis-platform/)
which runs Red Hat Enterprise Linux Server release 6.8 on a machine with 48 ×
Intel(R) Xeon(R) CPU E7
Gerrit Holl added the comment:
When the documentation says there is no public constructor, I expected it would
be impossible to create instances, as in:
TypeError: cannot create 'builtin_function_or_method' instances
Perhaps I misunderstand the documentation
Changes by Gerrit Holl topjakl...@gmail.com:
--
type: - crash
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue24538
___
___
Python-bugs-list
New submission from Gerrit Holl:
`shutil.copystat` fails on [panfs](https://en.wikipedia.org/wiki/Panasas#PanFS)
if the source file lacks u+w, because setting extended attributes results in a
`PermissionError`. This leads to higher end functions such as
`shutil.copytree` to fail. More
Gerrit Holl added the comment:
Perhaps the solution would be some kind of flag, at least for copytree and
possibly others, on what to do when attributes cannot be completely copied — a
bit like numpys options to raise, warn, or ignore
New submission from Gerrit Holl:
When a decorater contains a `lambda` declaration, using the pdb command
`longlist` will show only the definition of the decorator. The definition of
the function itself is not shown:
cat mini.py
#!/usr/bin/python3.4
def foo(x, y=None):
return x
@foo
New submission from Gerrit Holl:
It would be great if we could add either str or Path to Path objects.
Currently, we can join paths only by adding a directory, or by replacing the
suffix. But sometimes we want to build up a path more directly. With strings
we can do that simply
New submission from Gerrit Holl:
When I initialise a class that doesn't define its own __init__, but I still
pass arguments, the error message is confusing:
class A: pass
...
A(42)
Traceback (most recent call last):
File stdin, line 1, in module
TypeError: object() takes no parameters
New submission from Gerrit Holl:
The standard library documentation “6.1. string — Common string operations”
describes string formatting through the `.format` method and the corresponding
mini-language, and the `Template` class. In the part describing the `Template`
class (6.1.4) is the text
New submission from Gerrit Holl topjakl...@gmail.com:
When accidentally passing a string to warnings.warn where one should pass a
Warning-class, the error message is rather confusing:
$ ./python
Python 2.7.1+ (release27-maint:88766, Mar 8 2011, 16:51:59)
[GCC 4.4.5] on linux2
Type help
New submission from Gerrit Holl topjakl...@gmail.com:
When the time passed to time.ctime is large, it adds a newline:
import time
time.ctime(236)
'Wed Apr 8 17:04:32 6325'
time.ctime(237)
'Wed Jul 14 08:09:04 10680\n'
--
components: Library (Lib)
messages: 122665
nosy: Gerrit.Holl
Gerrit Holl topjakl...@gmail.com added the comment:
I'm on a 64bit system (kubuntu lucid lynx)
$ uname -a
Linux sjisjka 2.6.32-25-generic #45-Ubuntu SMP Sat Oct 16 19:52:42 UTC 2010
x86_64 GNU/Linux
--
___
Python tracker rep...@bugs.python.org
http
Gerrit Holl topjakl...@gmail.com added the comment:
Yes, this solves the issue:
$ ./python
Python 3.1.3 (r313:86834, Nov 28 2010, 17:34:23)
[GCC 4.4.3] on linux2
Type help, copyright, credits or license for more information.
import time
time.ctime(240)
'Fri Apr 10 03:12:32 71654
35 matches
Mail list logo
