Martin Panter added the comment:
I think this patch is okay
--
nosy: +martin.panter
___
Python tracker
<http://bugs.python.org/issue27109>
___
___
Python-bug
Martin Panter added the comment:
For this particular case, there is only one existing test class called
CgiTests, without any special setup or handling. It seems too arbitrary to have
a second class called MiscTestCase, so I would prefer the new test be added to
CgiTests
Changes by Martin Panter :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
I don’t see any violation or anything needing fixing in the documentation or
code. Can you elaborate?
The fcntl() documentation says “The values used for ‘cmd’ [F_DUPFD, etc] are
operating system dependent, and are available as constants in the ‘fcntl’
module
Martin Panter added the comment:
It looks like your commit to the “default” branch already included the relevant
changes, so I did the merge while keeping the files the same. If I missed
something, it may need an extra update. But the tests seem to be working (is
there a turtledemo test
Martin Panter added the comment:
Thanks for the report Phoenix
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
It seems like this code does get used, so I agree to add it.
I think the table in the RST documentation should also be updated, possibly
with a versionadded/versionchanged notice.
--
nosy: +martin.panter
___
Python
Martin Panter added the comment:
Indeed it is duplicated! I must have read this in my head many times before
pushing it, and never picked it up.
--
___
Python tracker
<http://bugs.python.org/issue27
Martin Panter added the comment:
This recently hung AMD64 FreeBSD 9.x 3.5. The stack trace was different, and
there is only one thread:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.x%203.5/builds/828/steps/test/logs/stdio
[398/398] test_io
Timeout (0:15:00)!
Thread
Martin Panter added the comment:
I found a relatively recent case of this failing on a buildbot:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.x%203.x/builds/4339/steps/test/logs/stdio
== CPython 3.6.0a1+ (default:eaee5aed6fbc, May 27 2016, 05:52:21) [GCC 4.2.1
20070831 patched
Martin Panter added the comment:
Thanks, that is a very good point about threads. I remember that testing
tkinter leaves an internal TK or TCL thread running in the background. And I
understand Free BSD delivers process signals to arbitrary threads (not
necessarily the main thread).
But the
Martin Panter added the comment:
Found the same hang on the 3.6 branch, which was definitely running separate
processes:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.x%203.x/builds/4210/steps/test/logs/stdio
== CPython 3.6.0a0 (default:ed6345cb08ab, Apr 29 2016, 18:35:36) [GCC
Martin Panter added the comment:
Here is a patch for just the wsgiref module, which I plan to commit in time for
3.5.2. I also updated the test case to avoid hanging due to a (theoretical)
race with how signal handling works.
--
Added file: http://bugs.python.org/file43066/wsgiref
Martin Panter added the comment:
Attached patch adds a possible test, and fixes the truncation problem I
mentioned above.
I tried testing set_completer_delims() with a UTF-8 locale, but I suspect Gnu
Readline does not support it. I called set_completer_delims("\xF6"), which
encode
Changes by Martin Panter :
Added file: http://bugs.python.org/file43068/readline_locale.v2.patch
___
Python tracker
<http://bugs.python.org/issue16182>
___
___
Python-bug
Changes by Martin Panter :
Removed file: http://bugs.python.org/file43067/readline_locale.v2.patch
___
Python tracker
<http://bugs.python.org/issue16182>
___
___
Pytho
Martin Panter added the comment:
For Python 3, the bytes.splitlines() and bytearray.splitlines() documentation
has been moved to a separate section out (Issue 21777). I don’t think it is
good to add much detail of bytes.splitlines() in the str.splitlines()
documentation.
For Python 2
New submission from Martin Panter:
This fixes various spelling mistakes in documentation, code comments, and test
function names.
There is also a correction to an exception message in
Modules/_ctypes/callproc.c, which I propose to fix only in 3.6.
--
files: typos.py3.patch
keywords
Changes by Martin Panter :
Added file: http://bugs.python.org/file43077/typos.py2.patch
___
Python tracker
<http://bugs.python.org/issue27171>
___
___
Python-bugs-list m
Martin Panter added the comment:
Thanks. The py2 version is partly a back port, but I have searched for the same
kind of spelling mistakes and fixed them in different places, e.g. “overriden”
in Lib/bsddb/test/test_all.py, which does not exist in Py 3
Martin Panter added the comment:
Reopening to change the Python 2 documentation. A starting point may be
Matthew’s patch and/or Alexander’s patch in Issue 22232.
--
status: closed -> open
___
Python tracker
<http://bugs.python.org/issu
Changes by Martin Panter :
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
Here is an updated patch for Python 2, based on Benjamin’s commit, Matthew’s
earlier py27 patch, and Alexander’s backport of related changes from Python 3.
Let me know what you think.
--
nosy: +Alexander Schrijver
stage: resolved -> patch review
Ad
Martin Panter added the comment:
You should be able to use a compression (or decompression) object as a
workaround. But calling zlib.compress() multiple times would generate multiple
separate deflated streams, which is different.
I think it is reasonable for Python to handle larger data sizes
Martin Panter added the comment:
I don’t know much about the conventions for stdout etc encoding on Windows. But
in general, the patch does not seem robust. Does it work if sys.stdout is a
pipe or file (not a console)? I doubt it will work when sys.stdout has been
replaced by e.g. StringIO
Martin Panter added the comment:
Thanks for the patches. Maybe Gregory knows more about this than I do (I never
used zdict). But I think I have figured it out today, so I left some review
comments :)
I’m not sure whether this really is a bug fix, although it would have little if
any impact
Martin Panter added the comment:
I think Issue 6135 has a bit of discussion on adding encoding and error
parameters to subprocess.Popen etc.
--
___
Python tracker
<http://bugs.python.org/issue27
Martin Panter added the comment:
Patch v4 resolves conflicts against the current code. I also used AC_CHECK_DECL
invocations instead of AC_COMPILE_IFELSE in the configure script.
FTR I think my problems with the prompt and the terminal settings are caused by
a line of code that is commented
Martin Panter added the comment:
V3 finishes what I started in v2:
* Changed unchecked PyBytes_AsString() → PyBytes_AS_STRING()
* Testing more functions for non-ASCII characters
I tried to test it with Editline on Linux (using my patch for Issue 13501).
There seem to be many quirks with my
Changes by Martin Panter :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
Patch B changes _Py_device_encoding() to accept a file descriptor of 3, which
seems wrong to me.
Patch A is like the earlier patch, but calls os.device_encoding(1) instead of
relying on sys.stdout, etc. I think this will still fail when the Python
parent’s
Martin Panter added the comment:
Klamann, thanks for crash report. I think your decompress crash is explained by
the bug expanding past UINT_MAX I identified above. The key is that length = 0
in zlib_Decompress_decompress_impl(), as if wrapped around, and the return
value will have been
Martin Panter added the comment:
I prefer your first solution because it seems to fit in better with how things
were intended.
I can add in handling of partial writes with a deprecation warning when I get a
chance. I guess the documentation would be something like “Since 3.5.2, partial
Martin Panter added the comment:
Don’t forget to fix the RFC number (see review) :)
--
___
Python tracker
<http://bugs.python.org/issue26589>
___
___
Python-bug
Martin Panter added the comment:
FWIW I doubt Git is any better at this than Mercurial:
<https://github.com/python/cpython/blame/master/Lib/test/test_string.py#L190>
Git can automatically pick up file renames and copies when analysing the
history, but has no special metadata for t
Martin Panter added the comment:
This broke test_socketserver:
http://buildbot.python.org/all/builders/AMD64%20Windows7%20SP1%203.x/builds/7709/steps/test/logs/stdio
0:08:57 [193/400/1] test_socketserver failed
test test_socketserver crashed -- Traceback (most recent call last):
File
&q
New submission from Martin Panter:
$ gdb --args ./python -c '
import io
class R(io.RawIOBase):
def writable(self): return True
def write(self, b):
print("About to evaluate {!r}.format".format(b))
b.format
print("Never reached")
Martin Panter added the comment:
I agree with avoiding the term “bytes-like object” in 2.7. It is easy to be
specific when talking about concrete classes like BufferedWriter and BytesIO.
But the harder question is what to specify for the abstract base classes.
In patch v8, I propose to change
Martin Panter added the comment:
I left some minor review suggestions. But a more serious problem is that the
change to Doc/reference/datamodel.rst breaks the formatting for me. Instead of
a HTML , I get a followed by a . Unfortunately I don’t know
of a way to get index entries to point to
Martin Panter added the comment:
IMO I wouldn’t bother. David has already done the change a different way, which
is simpler to understand commit-by-commit, even if it messes with the Mercurial
annotate history.
FWIW I guess you could do a similar merge with an old version to restore the
Martin Panter added the comment:
Actually this bug report was opened because an implementation that lacked
memoryview support was broken by 2.7. The point is to document that (a subset
of) memoryview objects may be passed to custom implementations.
Does this need a “changed in 2.7” notice? I
Martin Panter added the comment:
I don’t see the point of mentioning array() objects at all. It’s hard to
support array in a Python 2 implementation, as demonstrated by readinto(). And
the special support for array('b') won’t help if you pass in array('B') with
values 12
Martin Panter added the comment:
I thought :class:`str` is better documented in Python 2, but I can write bytes
if you prefer. I guess it is more consistent with the rest of io.rst.
--
___
Python tracker
<http://bugs.python.org/issue20
Martin Panter added the comment:
In wsgiref-only.v2.patch I added deprecated support for continuing when a
partial write is detected. I don’t think it needs documenting though.
--
Added file: http://bugs.python.org/file43143/wsgiref-only.v2.patch
Martin Panter added the comment:
Here is my suggested change to use setitimer(). I also closed the pipe, which
means there is no need to time out the read, speeding the test up a bit more.
--
keywords: +patch
stage: -> patch review
versions: +Python 2.7
Added file: h
Martin Panter added the comment:
Here is a patch with my suggestion to send the signal only when read() returns,
and also closing the read end of the pipe before the write end.
--
keywords: +patch
stage: -> patch review
Added file: http://bugs.python.org/file43147/int-write.pa
Martin Panter added the comment:
I haven’t tested, but maybe this will do what you want:
.. index:: object; code, code object
or
.. index::
single: object; code
single: code object
--
___
Python tracker
<http://bugs.python.org/issue15
Martin Panter added the comment:
Not sure why it doesn’t come up in the code review, but in
diff --git a/Doc/reference/datamodel.rst b/Doc/reference/datamodel.rst
index c0b4930..f34ba9f 100644
--- a/Doc/reference/datamodel.rst
+++ b/Doc/reference/datamodel.rst
@@ -847,8 +847,10 @@ Internal
Martin Panter added the comment:
Thanks for the patch Tommy. One minor quirk that bugs me: *mode*'s gets marked
up with an opening quote rather than an apostrophe. Perhaps we can change it to
a Unicode apostrophe (*mode*’s). Another alternative would be: the octal
representation of
Martin Panter added the comment:
What failures are seeing, and what version of Sphinx are you using? Perhaps
this is related to Issue 26638?
--
nosy: +martin.panter
___
Python tracker
<http://bugs.python.org/issue27
Changes by Martin Panter :
--
Removed message: http://bugs.python.org/msg267213
___
Python tracker
<http://bugs.python.org/issue27202>
___
___
Python-bugs-list m
Martin Panter added the comment:
Ignore my message; I was thinking of something different :)
I presume you are running something like the following, which seems to
smoke-test the examples in the documentation:
make -C Doc doctest
--
___
Python
Martin Panter added the comment:
I suggest to drop the RESET_CONTENT exception. Content-Length: 0 is explicitly
allowed by <https://tools.ietf.org/html/rfc7231#section-6.3.6> option a), and
is not very different to the general case IMO. Content-Length was added here to
help with buggy c
Martin Panter added the comment:
Ah, forget that bit about RESET_CONTENT. If we include Content-Length, it will
probably be non-zero, which will be wrong. So your patch is better in that
regard.
--
___
Python tracker
<http://bugs.python.
Martin Panter added the comment:
make -C Doc/ suspicious fails:
WARNING: [faq/programming:1177] ".. testsetup:" found in ".. testsetup::"
Suspicious check complete; look for any errors in the above output or in
build/suspicious/suspicious.csv. If all issues are false pos
Martin Panter added the comment:
Since I stuck my neck out, I will say that this seems reasonable to me. We
could probably still run doctest in the actual 2.7 branch to detect any errors
there.
--
stage: -> patch review
versions: +Python
Martin Panter added the comment:
Today I discovered that Christian’s defusedxml project already does the same
sort of thing. The difference is he calls the parameter forbid_entities. So I
have updated my patch and changed the name from reject_entities to
forbid_entities for compatibility
Martin Panter added the comment:
I think it would be good to move the sentence about FileExistsError away from
your new paragraph about *mode*. Also, I would try to make the hyperlink a bit
more explicit and descriptive, maybe
The *mode* parameter is passed to :func:`mkdir`; see :ref:`the
Martin Panter added the comment:
It seems the test is broken on Windows:
http://buildbot.python.org/all/builders/x86%20Windows7%203.5/builds/915/steps/test/logs/stdio
==
ERROR: test_mixed_case_module_names_are_lower_cased
Martin Panter added the comment:
Above two commits were actuall for Issue 26372
--
nosy: +martin.panter
___
Python tracker
<http://bugs.python.org/issue26
Martin Panter added the comment:
Maybe use os.devnull instead of "/dev/null"?
http://buildbot.python.org/all/builders/AMD64%20Windows7%20SP1%203.x/builds/7728/steps/test/logs/stdio
==
Martin Panter added the comment:
Okay here are some more thoughts about your latest patch:
## Automatic RTLD_MEMBER ##
I was still uneasy about the automatic setting of RTLD_MEMBER. But I looked for
how others handle this, and I found Libtool’s LTDL library, and Apache Portable
Runtime (APR
Martin Panter added the comment:
For a HEAD request, I think we should continue to send Content-Length (except
in combination with one of the special responses). HEAD is slightly different
to 304 Not Modified. With HEAD vs GET, the response code and other header
values do not change, but the
Martin Panter added the comment:
My guess is Michael was doing something like
>>> class Monkey(http.client.HTTPSConnection):
... pass
...
>>> http.client.HTTPSConnection = Monkey
>>> http.client.HTTPSConnection("bugs.python.org")
Traceback (most rec
Martin Panter added the comment:
I think it would be unusual to have different error handling for GET vs HEAD,
given HEAD is supposed to be exactly like GET but without a body. On the other
hand, see SimpleHTTPRequestHandler.send_head() for an example of calling
send_error(NOT_FOUND) for both
Martin Panter added the comment:
Sorry to be a pain, but I think the new logic for HEAD is wrong. See review.
--
___
Python tracker
<http://bugs.python.org/issue25
Martin Panter added the comment:
Here is a patch which I propose to use that factors out a set_inflate_zdict()
function. I think this will help with maintainence, e.g. if someone changes the
UINT_MAX checking, but forgets that there are two places than need changing.
Let me know if that is
Martin Panter added the comment:
Here is an updated patch with the remaining fixes for other servers.
I considered changing the behaviour of shutil.copyfileobj() to handle partial
writes better. But I decided against it, because: (a) nobody has complained
yet, and (b) it could potentially
Martin Panter added the comment:
This test is already decorated with @skip_if_broken_ubuntu_ssl. I’m not sure
Python should go too far out of its way to handle downstream patches, but it
seems there is a precedent here.
--
nosy: +martin.panter
Martin Panter added the comment:
I was a little concerned that Alex’s patch used “test A -ef B”, which is not
specified by Posix. But it is apparently widely supported (and consider we
would only need it for cross compilation). Anyway, I was going to suggest test
"$(srcdir)" == &
Martin Panter added the comment:
Part of this patch is superceded by revision c7adad17f663 (Issue 22579). Here
is a patch of the remaining significant changes. But I don’t know enough about
Modules/Setup.config.in and Modules/Setup.dist to give it a proper review.
--
nosy
Changes by Martin Panter :
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
Minor thing: the patch has tabbed intentation in places rather than spaces.
As I understand it, if there is no entropy initialized, this patch will fall
back to reading /dev/urandom, which will return predictable data (opposite of
“random” data!). But since we
Martin Panter added the comment:
Maybe an alternative would be to add a special PYTHONHASHSEED=best-effort (or
whatever) value that says if there is no entropy available, use a predictable
hash seed. That would force whoever starts the Python process to be aware of
the problem
Changes by Martin Panter :
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Changes by Martin Panter :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Changes by Martin Panter :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
Thanks for the patches Jacek. This one I committed after moving the test into
CgiTests.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bug
Martin Panter added the comment:
FTR the closest thing I could find discussing reap_children() is Issue 16968.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
I haven’t really been following the FS-path stuff, but I guess this also needs
.. versionadded:: 3.6
--
nosy: +martin.panter
stage: -> patch review
___
Python tracker
<http://bugs.python.org/issu
Martin Panter added the comment:
Thanks Susumu, I think the logic is right this time. I left some new
suggestions for the tests, if you want to take them on or not, that’s up to you.
--
___
Python tracker
<http://bugs.python.org/issue25
Martin Panter added the comment:
The documentation says __file__ can be missing. It is missing for builtin
modules (e.g. sys) and frozen modules (e.g. __hello__).
--
nosy: +martin.panter
___
Python tracker
<http://bugs.python.org/issue27
Martin Panter added the comment:
Thanks Alex & Xavier.
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.pytho
Martin Panter added the comment:
BytesIO is supposed to implement IOBase. I would treat this as a bug in
existing versions too.
--
nosy: +martin.panter
stage: -> needs patch
versions: +Python 2.7, Python 3.5
___
Python tracker
&l
Martin Panter added the comment:
Patch 07 looks fine. I presume you still want to do the porting to 3.5 and 2.7.
--
___
Python tracker
<http://bugs.python.org/issue25
Martin Panter added the comment:
Thanks for the patch. I left a couple review comments.
Also, this new paragraph should be outside the “Note” box, which is
specifically about libedit aka Editline, not Gnu Readline.
--
stage: needs patch -> patch rev
Martin Panter added the comment:
Emanuel, have you seen
<https://docs.python.org/devguide/documenting.html#building-the-documentation>?
From my experience, install Sphinx, and then “make -C Doc/ html”.
About the patch, I think the “.. note” boxes should be intended und
Martin Panter added the comment:
Yes it seems you even fixed the problem with the definition list, which I
thought was not fixable :)
--
stage: needs patch -> commit review
___
Python tracker
<http://bugs.python.org/issu
Martin Panter added the comment:
New patch that I plan to commit:
* Use bytes instead of str
* bytes, bytearray or memoryview for readinto() methods
* bytes or memoryview for write() methods
* Added single “Changed in version 2.7” notice under the IOBase class (rather
than every version of
Martin Panter added the comment:
Thanks for helping with this Jelle.
The documentation of unpacking sequences vs iterables was adjusted in 3.6 as
part of Issue 23275. I guess part of revision 8a0754fed986 should be extracted
to 3.5 as well.
Looking at the function call syntax, positional and
Martin Panter added the comment:
Bug 1 is still present in 3.5. In 3.6 it is no longer relevant because the
documentation was changed for Issue 9232 (allowing trailing commas in function
definitions).
Bug 2 is present in 2.7, but is not relevant to 3.5+. Trailing commas are now
allowed in
Martin Panter added the comment:
Closing now that platform._sys_version() can tolerate the truncated version
info.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Changes by Martin Panter :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Martin Panter added the comment:
I think the box about NotImplementedError needs indenting too. See the box for
<https://docs.python.org/3.5/library/functions.html#locals> for a better
example.
--
___
Python tracker
<http://bugs.p
Martin Pitt added the comment:
> you could give some kind of command-line flag
That already exists -- set PYTHONHASHSEED=0.
> But I'll let someone else have the joys of negotiating with Lennart, and I
> won't blame the Python devs if using GRND_NONBLOCK unconditionally is l
Martin Panter added the comment:
If the seed is to be based on time.time(), why not use something based on
hash(time.time()) to avoid the 1/256 s resolution?
If there is a practical way to get pseudo-random data from the platform, it
would be better to use that, rather than cooking up our own
Martin Panter added the comment:
In Python 3.5 (since Issue 15944), you can now cast normal C-contiguous
memoryviews (including ones from ctypes) to bytes:
>>> a = (ctypes.c_double * 3)(1,2,3)
>>> m = memoryview(a)
>>> m.format
'>> byteview = m.cast("
Martin Panter added the comment:
I dislike the “block” term for a different reason: it suggests raising EAGAIN
(= BlockingIOError). But the proposal here is actually to generate data with
low entropy.
In the long term, it sounds like two separate functions is the way to go. I
would prefer
Martin Panter added the comment:
According to <http://man7.org/linux/man-pages/man2/getrandom.2.html>,
getrandom() returns no more than 32 MiB as an int on Linux. Doesn’t that mean
you can rely on syscall()’s long return value fitting in an int? Maybe just
cast n = (int)sycall(...)
Martin Panter added the comment:
Make that INT_MAX. Or change n from an int to a Py_ssize_t. Both Linux and
Solaris versions or getrandom() are documented as accepting size_t buflen.
--
___
Python tracker
<http://bugs.python.org/issue27
1501 - 1600 of 9832 matches
Mail list logo
