Marco Barisione added the comment:
Actually, sorry I realise I can pass `include_extras` to `get_type_hints`.
Still, it would be nicer not to have to do that.
--
___
Python tracker
<https://bugs.python.org/issue39
Marco Barisione added the comment:
This is particularly annoying if you are using `Annotated` with a dataclass.
For instance:
```
from __future__ import annotations
import dataclasses
from typing import Annotated, get_type_hints
@dataclasses.dataclass
class C:
v: Annotated[int, &quo
Marco Pagliaricci added the comment:
Andrew,
many thanks for your time, solving this issue.
I think your solution is the best to fix this little problem and I agree
with you on backporting.
My Best Regards,
and thanks again.
Marco
On Thu, Feb 17, 2022 at 10:29 AM Andrew Svetlov
wrote
Change by Marco Cognetta :
--
keywords: +patch
nosy: +mcognetta
nosy_count: 6.0 -> 7.0
pull_requests: +27293
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/29020
___
Python tracker
<https://bugs.python.org/i
Marco Pagliaricci added the comment:
Chris,
ok, I have modified the snippet of code to better show what I mean.
Still here, the message of the CancelledError exception is lost, but if I
comment line 10, and uncomment line 11, so I throw a ValueError("TEST"),
that "TEST" st
Marco Pagliaricci added the comment:
Chris,
I'm attaching to this e-mail the code I'm referring to.
As you can see, in line 10, I re-raise the asyncio.CancelledError exception
with a message "TEST".
That message is lost, due to the reasons we've talked about.
My point
Marco Pagliaricci added the comment:
OK, I see your point.
But I still can't understand one thing and I think it's very confusing:
1) if you see my example, inside the job() coroutine, I get correctly
cancelled with an `asyncio.CancelledError` exception containing my message.
2) No
New submission from Marco Pagliaricci :
I've spotted a little bug in how asyncio.CancelledError() exception is
propagated inside an asyncio.Task.
Since python 3.9 the asyncio.Task.cancel() method has a new 'msg' parameter,
that will create an asyncio.CancelledError
Marco Sulla added the comment:
Since probably Monica are taking her holidays, I try to decipher her answer.
Probably, the more problematic function spotted by Monica is update_one_slot. I
re-quote her sentence:
update_one_slot looks for the parent implementation by trying to find the
Marco Sulla added the comment:
I not finished my phrase. I'm sure that if there's a way to turn lemons
into lemonade, she is **MUCH** more skilled than me to find one.
--
___
Python tracker
<https://bugs.python.o
Marco Sulla added the comment:
Since my knowledge of this is very poor, I informed Monica about the issue. I'm
quite sure that if there's a way to turn lemons into lemonade :)
--
___
Python tracker
<https://bugs.python.o
New submission from Marco Sulla :
I asked on SO why subclassing dict makes the subclass much slower in some
operations. This is the answer by Monica
(https://stackoverflow.com/a/59914459/1763602):
Indexing and in are slower in dict subclasses because of a bad interaction
between a dict
Marco Sulla added the comment:
Close it, I have no time now :-(
--
resolution: -> later
stage: -> resolved
status: pending -> closed
___
Python tracker
<https://bugs.python.or
New submission from Marco E. :
The CSV library does not correctly interpret files in the following format
(test.csv):
"A" ,"B" ,"C"
"aa","bbb",""
"a" ,"bb" ,"ccc"
"aaa
Marco Franzo added the comment:
So, I use Ubuntu 20.10 and the terminal
is the one distributed with the system.
I think this problem born in my code here:
def generate_input():
while True:
str = input().strip()
yield helloworld_pb2.Operazione(operazione = str)
I think
New submission from Marco Franzo :
It would be better to write at the end of the program this:
os.system('stty sane')
because when you import readline, at the end of program, the console remains
unusable
--
assignee: docs@python
components: Documentation
messages: 384379
Marco Atzeri added the comment:
The Analysis is correct.
Removing the test for CYGWIN and always include the
solved the problem building all python (3.6,3.7,3.8) packages
https://sourceware.org/pipermail/cygwin-apps/2020-December/040845.html
https://sourceware.org/pipermail/cygwin
Marco Sulla added the comment:
The PR will probably be rejected... you can do something like this:
1. in the venv on our machine, do `pip freeze`. This gives you the whole list
of installed dependencies
2. download all the packages using `pip download`
3. copy all the packages on the cloud
Marco Sulla added the comment:
I did PGO+LTO... --enable-optimizations --with-lto
--
___
Python tracker
<https://bugs.python.org/issue41835>
___
___
Python-bug
Marco Sulla added the comment:
Well, actually Serhiy is right, it does not seem that the macro benchs did show
something significant. Maybe the code can be used in other parts of CPython,
for example in _pickle, where dicts are loaded. But it needs also to expose,
maybe internally only
Marco Sulla added the comment:
Well, following your example, since split dicts seems to be no more supported,
I decided to be more drastic. If you see the last push in PR 22346, I do not
check anymore but always resize, so the dict is always combined. This seems to
be especially good for
Marco Castelluccio added the comment:
I've opened https://github.com/python/cpython/pull/22751 to fix this, I know
there was already a PR, but it seems to have been abandoned.
--
___
Python tracker
<https://bugs.python.org/is
Change by Marco Castelluccio :
--
nosy: +marco-c
nosy_count: 6.0 -> 7.0
pull_requests: +21928
pull_request: https://github.com/python/cpython/pull/22751
___
Python tracker
<https://bugs.python.org/issu
Marco Sulla added the comment:
Well, after a second thought I think you're right, there's no significant
advantage and too much duplicated code.
--
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.py
Marco Sulla added the comment:
The fact is that, IMHO, PGO will "false" the results, since it's quite
improbable that in the test battery there's a test of creation of a dict from
another dict with an hole. It seems to me that the comparison between the
normal builds
Marco Sulla added the comment:
Note that this time I've no slowdown in the macro bench, since I used normal
builds, not optimized ones. I suppose an optimized build will show slowdown
because the new functions are not in the test ba
Marco Sulla added the comment:
I'm quite sure I not invented the wheel :) but I think it's a good improvement:
| pathlib | 35.8 ms | 35.1 ms| 1.02x
faster | Significant (t=13.21) |
| scimark_monte_carlo | 176 ms | 172 ms
New submission from Marco Sulla :
The PR #22948 is an augmented version of #22346. It speeds up also the creation
of:
1. dicts from other dicts that are not "perfect" (combined and without holes)
2. fromkeys
3. copies of dicts with many holes
4. dict from keywords, as in #22346
Marco Sulla added the comment:
I commented out sqlalchemy in the requirements.txt in the pyperformance source
code, and it worked. I had also to skip tornado:
pyperformance run -r
-b,-sqlalchemy_declarative,-sqlalchemy_imperative,-tornado_http -o
../perf_master.json
This is my result
Marco Sulla added the comment:
@Mark.Shannon I tried to run pyperformance, but wheel does not work for Python
3.10. I get the error:
AssertionError: would build wheel with unsupported tag ('cp310', 'cp310',
'linux_x86_64')
--
Marco Sulla added the comment:
@methane: well, to be honest, I don't see much difference between the two
pulls. The major difference is that you merged insertdict_init in
dict_merge_init.
But I kept insertdict_init separate on purpose, because this function can be
used in other f
Marco Sulla added the comment:
Another bench:
python -m pyperf timeit --rigorous "dict(ihinvdono='doononon',
gowwondwon='nwog', bdjbodbob='nidnnpn', nwonwno='vndononon',
dooodbob='iohiwipwgpw', doidonooq='ndwnnpnpnp', fn
Change by Marco Castelluccio :
--
keywords: +patch
pull_requests: +21713
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/22751
___
Python tracker
<https://bugs.python.org/issu
New submission from Marco Castelluccio :
Shelve is currently defaulting to Pickle protocol 3, instead of using Pickle's
default protocol for the Python version in use.
This way, Shelve's users don't benefit from improvements introduced in newer
Pickle protocols, unless the
Marco Sulla added the comment:
I closed it for this reason:
https://github.com/python/cpython/pull/22438#issuecomment-702794261
--
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.org/i
Marco Sulla added the comment:
I do not remember the problem I had, but when I experimented with frozendict I
get one of these errors. I failed to understand the problem so I added the
additional info.
Maybe adding an assert in debug mode? It will be visible only to devs
New submission from Marco Sulla :
All pickle error messages in typeobject.c was a generic "cannot pickle 'type'
object". Added some explaining for every individual error.
--
components: Interpreter Core
messages: 377747
nosy: Marco Sulla
priority: normal
pull_requ
Marco Sulla added the comment:
> `dict(**o)` is not common use case. Could you provide some other benchmarks?
You can do
python -m timeit -n 200 "dict(key1=1, key2=2, key3=3, key4=4, key5=5,
key6=6, key7=7, key8=8, key9=9, key10=10)"
or with pyperf. In this case, sinc
New submission from Marco Sulla :
I've done a PR that speeds up the vectorcall creation of a dict using keyword
arguments. The PR in practice creates a insertdict_init(), a specialized
version of insertdict. I quote the comment to the function:
Same to insertdict but specialize
Marco Paolini added the comment:
I was thinking to just clarify a bit the error message that results from
Py_NumberAdd. This won't make it slower in the "hot" path
doing something like (not compile tested, sorry)
--- a/Python/bltinmodule.c
+++ b/Python/bltinmodule.c
@@ -
Marco Paolini added the comment:
also worth noting, the start argument is type checked instead. Maybe we could
apply the same checks to the items of the iterable?
python3 -c "print(sum(('a', 'b', 'c'), start='d'))"
Traceback (most recent
Marco Paolini added the comment:
This happens because the default value for the start argument is zero , hence
the first operation is `0 + 'a'`
--
nosy: +mpaolini
___
Python tracker
<https://bugs.python.o
Change by Marco Trevisan :
--
keywords: +patch
pull_requests: +20875
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/21731
___
Python tracker
<https://bugs.python.org/issu
New submission from Marco Trevisan :
Webbrowser uses env variables such as GNOME_DESKTOP_SESSION_ID that have been
dropped by GNOME in recent releases
--
components: Library (Lib)
messages: 374806
nosy: Trevinho
priority: normal
severity: normal
status: open
title: webbrowser uses
Marco Paolini added the comment:
hello Thomas,
do you need any help fixing the conflicts in your PR?
even if Lib/warnings.py changed a little in the last 2 years, your PR is still
good!
--
nosy: +mpaolini
___
Python tracker
<ht
New submission from Marco Barisione :
The generation of pickle files in load_grammar in lib2to3/pgen2/driver.py is
racy as other processes may end up reading a half-written pickle file.
This is reproducible with the command line tool, but it's easier to reproduce
by importing lib2to3
New submission from Marco Sulla :
This is a little PR with some micro-optimizations to the PySequence_Tuple()
function. Mainly, it simply add a support variable new_n_tmp_1 instead of
reassigning newn multiple times.
--
components: Interpreter Core
messages: 363974
nosy: Marco Sulla
Marco Sulla added the comment:
@Eric V. Smith: that you for your effort, but I'll never use an API marked as
private, that is furthermore undocumented.
--
___
Python tracker
<https://bugs.python.org/is
Marco Sulla added the comment:
> What would "{} {}".partial_format({}) return?
`str.partial_format()` was proposed exactly to avoid such tricks.
> It is not possible to implement a "safe" variant of str.format(),
> because in difference to Template it can call ar
Change by Marco Sulla :
--
resolution: -> duplicate
stage: -> resolved
status: open -> closed
type: -> behavior
___
Python tracker
<https://bugs.python
Marco Sulla added the comment:
This is IMHO broken.
1. _ensure_list() allows strings, because, documentation says, they are split
in finalize_options(). But finalize_options() does only split keywords and
platforms. It does _not_ split classifiers.
2. there's no need that key
Marco Sulla added the comment:
> Do you have some concrete use case for this?
Yes, for EWA:
https://marco-sulla.github.io/ewa/
Since it's a code generator, it uses templates a lot, and much times I feel the
need for a partial substitution. In the end I solved with some ugl
New submission from Marco Sulla :
I got this warning. I suppose that `distutils` can use any iterable.
--
components: Distutils
messages: 363354
nosy: Marco Sulla, dstufft, eric.araujo
priority: normal
severity: normal
status: open
title: Warning: 'classifiers' should be a
Marco Sulla added the comment:
IMHO such a feature is useful for sysops that does not have a graphical
interface, as Debian without an X. That's why vi is (unluckily) very popular
also in 2020. IDLE can't be used in this cases.
Windows users can't remotely login withou
New submission from Marco Sulla :
In `string` module, there's a very little known class `Template`. It implements
a very simple template, but it has an interesting method: `safe_substitute()`.
`safe_substitute()` permits you to not fill the entire Template at one time. On
the contrar
Marco Sulla added the comment:
Excuse me, but my original "holistic" proposal was rejected and it was
suggested to me to propose only relevant changes, and one for issue. Now you
say exactly the contrary. I feel a bit confused.
PS: yes, I can, and I use, IPython. But IMHO IPytho
Marco Sulla added the comment:
I agree with Pablo Galindo Salgado: https://bugs.python.org/issue35912#msg334942
The "quick and dirty" solution is to change MAINCC to CC, for _testembed.c AND
python.c (g++ fails with both).
After that, _testembed.c and python.c should be changed s
Marco Sulla added the comment:
Please read the message of Terry J. Reed:
https://bugs.python.org/issue38747#msg356345
I quote the relevant part below
> Skipping the rest of your post, I will just restate why I closed this
> issue.
>
> 1. It introduces too many features
Marco Sulla added the comment:
> Is this even possible in a plain text console?
Yes. See Jupyter Console (aka IPython).
--
___
Python tracker
<https://bugs.python.org/issu
New submission from Marco Sulla :
I suggest to add an implementation of bracketed paste mode in the REPL.
Currently if you, for example, copy & paste a piece of Python code to see if it
works, if the code have a blank line without indentation and the previous and
next line are indented,
Marco Sulla added the comment:
Furthermore, I have not understood a think: if I understood well,
--with-cxx-main is used on _some_ platforms that have problems with C++
extensions. What platforms? Is there somewhere a unit test for testing if
Python compiled on one of these platforms with
Marco Sulla added the comment:
Okay... if I have understood well, the problem is with C++ Extensions.
Some questions:
1. does this problem exists yet?
2. if yes, maybe Python have to wrap the python.c and _testembed.c so they can
also be compiled with a C++ compiler?
3. --with-cxx-main is
Marco Sulla added the comment:
OS: Lubuntu 18.04.4
Steps to reproduce:
sudo apt-get install git libbz2-dev liblzma-dev uuid-dev libffi-dev
libsqlite3-dev libreadline-dev libssl-dev libgdbm-dev libgdbm-compat-dev tk-dev
libncurses5-dev
git clone https://github.com/python/cpython.git
cd
Marco Sulla added the comment:
Mmmm... wait a moment. It seems the behavior is intended:
https://bugs.python.org/issue1324762
I quote:
The patch contains the following changes:
[...]
2) The compiler used to translate python's main() function is
stored in the configure / Mak
Marco Sulla added the comment:
https://github.com/python/cpython/pull/18721
--
___
Python tracker
<https://bugs.python.org/issue39697>
___
___
Python-bugs-list m
Change by Marco Sulla :
--
keywords: +patch
pull_requests: +18079
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/18721
___
Python tracker
<https://bugs.python.org/issu
Marco Sulla added the comment:
The problem is here:
Programs/_testembed.o: $(srcdir)/Programs/_testembed.c
$(MAINCC) -c $(PY_CORE_CFLAGS) -o $@ $(srcdir)/Programs/_testembed.c
`MAINCC` in my Makefile is `g++-9`. Probably, MAINCC is set to the value of
``--with-cxx-main`, if
New submission from Marco Sulla :
During `make test`, I get the error in the title.
(venv_3_9) marco@buzz:~/sources/cpython_test$ ll /dev/tty
crw-rw-rw- 1 root tty 5, 0 Mar 1 15:24 /dev/tty
--
components: Tests
messages: 363063
nosy: Marco Sulla
priority: normal
severity: normal
Marco Sulla added the comment:
> >>> int(1e100)
> 1159028911097599180468360808563945281389781327557747838772170381060813469985856815104
.
Oh my God... I'm just more convinced than before :-D
> Ya, this change will never be made - give up gracef
Marco Sulla added the comment:
All the examples you mentioned seems to me to fix code, instead of breaking it.
About 1e300**1, it's not a bug at all. No one can stop you to full your RAM
in many other ways :-D
About conventions, it does not seems to me that Python cares about
Marco Sulla added the comment:
Sorry, but I can't figure out what code can break this change. Integers are
implicitly converted to floats in operations with floats. How can this change
break old code?
> if you are worried about the performance
No, I'm worried about the ex
New submission from Marco Sulla :
(venv_3_9) marco@buzz:~/sources/python-frozendict$ python
Python 3.9.0a0 (heads/master-dirty:d8ca2354ed, Oct 30 2019, 20:25:01)
[GCC 9.2.1 20190909] on linux
Type "help", "copyright", "credits" or "license" for more inf
New submission from Marco Sulla :
I think a tuple comprehension could be very useful.
Currently, the only way to efficiently create a tuple from a comprehension is
to create a list comprehension (generator comprehensions are more slow) and
convert it with `tuple()`.
A tuple comprehension
Marco Sulla added the comment:
> I also distinctly remember seeing code (and writing such code myself) that
> performs computation on timeouts and does not care if the end value goes
> below 0.
This is not a good statistics. Frankly we can't measure the impact of the
cha
Change by Marco Sulla :
--
resolution: not a bug -> rejected
___
Python tracker
<https://bugs.python.org/issue39698>
___
___
Python-bugs-list mailing list
Un
Marco Sulla added the comment:
I asked why on StackOverflow, and an user seemed to find the reason. The
problem for him/her is in `update_one_slot()`.
`dict` implements directly `__contains__()` and `__getitem__()`. Usually,
`sq_contains` and `mp_subscript` are wrapped to implement
Change by Marco Sulla :
--
resolution: -> duplicate
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.org/issue39754>
___
___
New submission from Marco Sulla :
I noticed that `__contains__()` and `__getitem__()` of subclasses of `dict` are
much slower. I asked why on StackOverflow, and an user seemed to find the
reason.
The problem for him/her is that `dict` implements directly `__contains__()` and
`__getitem__
Marco Sulla added the comment:
I think in this case the error is more trivial: simply `Programs/_testembed.c`
is compiled with g++ but it should be compiled with gcc.
Indeed, there are much gcc-only options in the compilation of
`Programs/_testembed.c`, and g++ complains about them
Marco Sulla added the comment:
I see that many breaking changes was done in recent releases. I get only the
ones for `asyncio` in Python 3.8:
https://bugs.python.org/issue36921
https://bugs.python.org/issue36373
https://bugs.python.org/issue34790
https://bugs.python.org/issue32528
https
Marco added the comment:
uhm, no.
I can no more reproduce this. I was wrong. Sorry for the noise.
--
resolution: -> not a bug
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Marco Sulla added the comment:
Ah, well, this is not possible. I was banned from the mailing list. I wrote my
"defense" to conduct...@python.org in date 2019-12-29, and I'm still waiting
for a response...
--
___
Python
Marco Sulla added the comment:
Well, the fact is, basically, for the other libraries you have not to re-run
`configure`. You have to install only the missing C libraries and redo `make`.
This works, for example, for zlib, lzma, ctypes, sqlite3, readline, bzip2.
Furthermore, it happened to
Marco Sulla added the comment:
> I recall very many cases in third-party libraries and commercial applications
Source?
--
___
Python tracker
<https://bugs.python.org/issu
New submission from Marco Sulla :
Python 3.9.0a3+ (heads/master-dirty:f2ee21d858, Feb 19 2020, 23:19:22)
[GCC 9.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import time
>>> time.sleep(-1
New submission from Marco Sulla :
I tried to compile Python 3.9 with:
CC=gcc-9.2.0 ./configure --enable-optimizations --with-lto
--with-cxx-main=g++-9.2.0
make -j 2
I got this error:
g++-9.2.0 -c -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall
-flto -fuse-linker-plugin
New submission from Marco Sulla :
Similarly to enhancement request #39695, I missed to install the debian package
with the include files for SSL, before compiling Python 3.9.
After installed it, `make` continued to not find the libraries and skipped the
creation of module _ssl.
Searching on
New submission from Marco Sulla :
When I first done `make` to compile Python 3.9, I did not installed some debian
development packages, like `uuid-dev`. So `_uuid` module was not built.
After installed the debian package I re-run `make`, but it failed to build
`_uuid` module. I had to edit
New submission from Marco :
Hello,
if I write
```
msg = email.message_from_bytes(...)
for part in msg.walk():
content_type = part.get_content_type()
if not part.get_content_maintype() == 'multipart':
filename = part.get_filename(None)
attachment = part.get_payload(d
Marco Sulla added the comment:
> this is the sort of thing that is usually best suited to be reported by
> linters, not the Python runtime.
TL;DR: if you write something like `a -- b`, it's quite extraordinary that you
really wanted to write this. You probably wanted to write
Marco Sulla added the comment:
> `++` isn't special
Indeed the problem is that no error or warning is raised if two operators are
consecutive, without a space between. All the cases you listed are terribly
unreadable and hardly intelligible.
Anyway I do not agree `++` is not
Marco Sulla added the comment:
> This is not a bug
No one said it's a bug. It's a defect.
> This has been part of Python since version 1
There are many things that was part of Python 1 that was removed.
> `++` should never be an operator in the future, precisely because
New submission from Marco Sulla :
Python 3.9.0a0 (heads/master-dirty:d8ca2354ed, Oct 30 2019, 20:25:01)
[GCC 9.2.1 20190909] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> 1 ++ 2
3
This is probably because
Marco Sulla added the comment:
marco@buzz:~$ python3.9
Python 3.9.0a0 (heads/master-dirty:d8ca2354ed, Oct 30 2019, 20:25:01)
[GCC 9.2.1 20190909] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from decimal
Marco Sulla added the comment:
Excuse me, ignore my previous post.
--
___
Python tracker
<https://bugs.python.org/issue36095>
___
___
Python-bugs-list mailin
Marco Sulla added the comment:
marco@buzz:~$ python3.9
Python 3.9.0a0 (heads/master-dirty:d8ca2354ed, Oct 30 2019, 20:25:01)
[GCC 9.2.1 20190909] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from decimal
Marco Sulla added the comment:
Excuse me, I had an epiphany.
NaN returns False for every comparison.
So in teory any element of the iterable should result minor that NaN.
So NaN should treated as the highest element, and should be at the end of the
sorted result!
Indeed this is the
Marco Sulla added the comment:
> No idea what "are minor that another object" could possibly mean.
Oh my god... a < b?
> I don't know what purpose would be served by checking ">=" too
Well, it's very simple. Since the sorting algorithm checks if
Marco Sulla added the comment:
Anyway, Java by default puts NaNs at the end of the iterable:
https://onlinegdb.com/SJjuiXE0S
--
___
Python tracker
<https://bugs.python.org/issue36
Marco Sulla added the comment:
Excuse me, but have you, Dickinson and Peters, read how I propose to check if
the object is orderable or not?
I explained it in a very detailed way, and this does not change the float
comparison. And does not need to check first if the iterable it totally
1 - 100 of 402 matches
Mail list logo
