bpo-35066: Make trailing percent test more portable. (GH-15907)
Different libc implementations have different behavior when presented with trailing % in strftime strings. To make test_strftime_trailing_percent more portable, compare the output of datetime.strftime directly to that of time.strftime rather than hardcoding.
(cherry picked from commit f2173ae38fa49235c3cdc28ae2ca2e19a375a596)
Co-authored-by: Benjamin Peterson <benjamin@python.org>
bpo-37424: Avoid a hang in subprocess.run timeout output capture (GH-14490)
Fixes a possible hang when using a timeout on subprocess.run() while
capturing output. If the child process spawned its own children or otherwise
connected its stdout or stderr handles with another process, we could hang
after the timeout was reached and our child was killed when attempting to read
final output from the pipes.
(cherry picked from commit 580d2782f70f8e0bed7ec20abb03d740cb83b5da)
Co-authored-by: Gregory P. Smith <greg@krypto.org>
* Write a message when killing a worker process
* Put a timeout on the second popen.communicate() call
(after killing the process)
* Put a timeout on popen.wait() call
* Catch popen.kill() and popen.wait() exceptions
Fix calling order of PyEval_InitThreads. (GH-5833)
As described in Doc/c-api/init.rst, PyEval_InitThreads() cannot be called
before Py_Initialize() function.
(cherry picked from commit 9e61066355b4b55c873d56f5f106a22463b56862)
Steve Dower [Tue, 10 Sep 2019 08:46:40 +0000 (01:46 -0700)]
bpo-35941: Fix performance regression in SSL certificate code (GH-12610)
Accumulate certificates in a set instead of doing a costly list contain
operation. A Windows cert store can easily contain over hundred
certificates. The old code would result in way over 5,000 comparison
operations
Signed-off-by: Christian Heimes <christian@python.org>
bpo-38077: IDLE no longer adds 'argv' to the user namespace (GH-15818)
This only happened when initializing the subprocess to run a module.
This recent bug only affected 3.7.4 and 3.8.0b2 to 3.8.0b4.
(cherry picked from commit c59295a1ca304f37ca136dd7efca9e560db27d28)
Co-authored-by: Terry Jan Reedy <tjreedy@udel.edu>
Victor Stinner [Mon, 9 Sep 2019 22:31:20 +0000 (00:31 +0200)]
bpo-38006: Avoid closure in weakref.WeakValueDictionary (GH-15641) (GH-15789)
weakref.WeakValueDictionary defines a local remove() function used as
callback for weak references. This function was created with a
closure. Modify the implementation to avoid the closure.
bpo-36502: Update link to UAX GH-44, the Unicode doc on the UCD. (GH-15301)
The link we have points to the version from Unicode 6.0.0, dated 2010.
There have been numerous updates to it since then:
https://www.unicode.org/reports/tr44/GH-Modifications
bpo-37702: Fix SSL's certificate-store leak on Windows (GH-15632)
ssl_collect_certificates function in _ssl.c has a memory leak.
Calling CertOpenStore() and CertAddStoreToCollection(), a store's refcnt gets incremented by 2.
But CertCloseStore() is called only once and the refcnt leaves 1.
(cherry picked from commit ed70129e15ea028469145111044a4349960a4e6f)
T. Wouters [Mon, 9 Sep 2019 13:38:37 +0000 (06:38 -0700)]
[3.7] bpo-37589: Add a few missing dependencies on .h files in the Makefile. (GH-15757) (GH-15769)
The missing dependencies prevented incremental builds from working when you touched any
of these files. Based on GH-14758 by @vemakereporter..
(cherry picked from commit b4612f5d54aced5bc37f1b85bf50b4cafa2480f0)
bpo-37936: Avoid ignoring files that we actually do track. (GH-15451) (GH-15748)
There were about 14 files that are actually in the repo but that are
covered by the rules in .gitignore.
Git itself takes no notice of what .gitignore says about files that
it's already tracking... but the discrepancy can be confusing to a
human that adds a new file unexpectedly covered by these rules, as
well as to non-Git software that looks at .gitignore but doesn't
implement this wrinkle in its semantics. (E.g., `rg`.)
Several of these are from rules that apply more broadly than
intended: for example, `Makefile` applies to `Doc/Makefile` and
`Tools/freeze/test/Makefile`, whereas `/Makefile` means only the
`Makefile` at the repo's root.
bpo-37705: Improve the implementation of winerror_to_errno() (GH-15623)
winerror_to_errno() is no longer automatically generated.
Do not rely on the old _dosmapperr() function.
Add ERROR_NO_UNICODE_TRANSLATION (1113) -> EILSEQ.
(cherry picked from commit 19052a11314e7be7ba003fd6cdbb5400a5d77d96)
bpo-37445: Include FORMAT_MESSAGE_IGNORE_INSERTS in FormatMessageW() calls (GH-14462)
If FormatMessageW() is passed the FORMAT_MESSAGE_FROM_SYSTEM flag without FORMAT_MESSAGE_IGNORE_INSERTS, it will fail if there are insert sequences in the message definition.
(cherry picked from commit a6563650c835d50f7302971a5b145e94f9d0dc68)
Restart lines now always start with '=' and never end with ' ' and fill the width of the window unless that would require ending with ' ', which could be wrapped by itself and possible confusing the user.
(cherry picked from commit 38da805d563422cf1bb9cd9be24c73806840fe30)
Co-authored-by: Terry Jan Reedy <tjreedy@udel.edu>
bpo-37380: subprocess: don't use _active on win (GH-14360) (GH-15706)
As noted by @eryksun in [1] and [2], using _cleanup and _active(in
__del__) is not necessary on Windows, since:
> Unlike Unix, a process in Windows doesn't have to be waited on by
> its parent to avoid a zombie. Keeping the handle open will actually
> create a zombie until the next _cleanup() call, which may be never
> if Popen() isn't called again.
This patch simply defines `subprocess._active` as `None`, for which we already
have the proper logic in place in `subprocess.Popen.__del__`, that prevents it
from trying to append the process to the `_active`. This patch also defines
`subprocess._cleanup` as a noop for Windows.
bpo-37902: IDLE: Add scrolling for IDLE browsers. (GH-15368)
Modify the wheel event handler so it can also be used for module, path, and stack browsers.
Patch by George Zhang.
(cherry picked from commit 2cd902585815582eb059e3b40e014ebe4e7fdee7)
bpo-38022: IDLE: upgrade help.html to sphinx 2.x HTML5 output (GH-15664)
The HTML5 output from Sphinx 2.x adds '<p>' tags within list elements. Using a new prevtag attribute, ignore these instead of emitting unwanted '\n\n'.
Also stop looking for 'first' classes on tags (no longer present) and fix the bug of double-spacing instead of single spacing after <pre> blocks.
(cherry picked from commit 580bdb0ece681537eadb360f0c796123ead7a559)
Co-authored-by: Tal Einat <taleinat+github@gmail.com>
Fixes a case in which email._header_value_parser.get_unstructured hangs the system for some invalid headers. This covers the cases in which the header contains either:
- a case without trailing whitespace
- an invalid encoded word
Extending the hover delay in test_tooltip should avoid spurious test_idle failures.
One longer delay instead of two shorter delays results in a net speedup.
(cherry picked from commit 132acaba5a7f01373ca624b1a5975b190fe866f5)
Co-authored-by: Tal Einat <taleinat+github@gmail.com>
bpo-37936: Remove some .gitignore rules that were intended locally. (GH-15542)
These appeared in commit c5ae169e1. The comment on them, as well as
the presence among them of a rule for the .gitignore file itself,
indicate that the author intended these lines to remain only in their
own local working tree -- not to get committed even to their own repo,
let alone merged upstream.
They did nevertheless get committed, because it turns out that Git
takes no notice of what .gitignore says about files that it's already
tracking... for example, this .gitignore file itself.
Give effect to these lines' original intention, by deleting them. :-)
Git tip, for reference: the `.git/info/exclude` file is a handy way
to do exactly what these lines were originally intended to do. A
related handy file is `~/.config/git/ignore`. See gitignore(5),
aka `git help ignore`, for details.
These were caused by keeping around a reference to the Squeezer
instance and calling it's load_font() upon config changes, which
sometimes happened even if the shell window no longer existed.
This change completely removes that mechanism, instead having the
editor window properly update its width attribute, which can then
be used by Squeezer.
(cherry picked from commit d4b4c00b57d24f6ee2cf3a96213406bb09953df3)
Co-authored-by: Tal Einat <taleinat+github@gmail.com>
_PyTime_t type is defined as int64_t, and so min/max are INT64_MIN/INT64_MAX,
not PY_LLONG_MIN/PY_LLONG_MAX.
(cherry picked from commit 8e76c456226438f2e4931ce7baf05ac8faae34a1)
bpo-37482: Fix email address name with encoded words and special chars (GH-14561)
Special characters in email address header display names are normally
put within double quotes. However, encoded words (=?charset?x?...?=) are
not allowed withing double quotes. When the header contains a word with
special characters and another word that must be encoded, the first one
must also be encoded.
In the next example, the display name in the From header is quoted and
therefore the comma is allowed; in the To header, the comma is not
within quotes and not encoded, which is not allowed and therefore
rejected by some mail servers.
From: "Foo Bar, France" <foo@example.com>
To: Foo Bar, =?utf-8?q?Espa=C3=B1a?= <foo@example.com>
Added back mention that ensure_future actually scheduled obj. This documentation just mentions what ensure_future returns, so I did not realize that ensure_future also schedules obj.
(cherry picked from commit 092911d5c0d8f6db8a0cb02fecd73dbb650f9e2e)
Co-authored-by: Roger Iyengar <ri@rogeriyengar.com>
bpo-35518: Skip test that relies on a deceased network service. (GH-15349)
If this service had thoroughly vanished, we could just ignore the
test until someone gets around to either recreating such a service
or redesigning the test to somehow work locally. The
`support.transient_internet` mechanism catches the failure to
resolve the domain name, and skips the test.
But in fact the domain snakebite.net does still exist, as do its
nameservers -- and they can be quite slow to reply. As a result
this test can easily take 20-30s before it gets auto-skipped.
bpo-36502: Correct documentation of str.isspace() (GH-15019) (GH-15296)
The documented definition was much broader than the real one:
there are tons of characters with general category "Other",
and we don't (and shouldn't) treat most of them as whitespace.
Rewrite the definition to agree with the comment on
_PyUnicode_IsWhitespace, and with the logic in makeunicodedata.py,
which is what generates that function and so ultimately governs.
Add suitable breadcrumbs so that a reader who wants to pin down
exactly what this definition means (what's a "bidirectional class"
of "B"?) can do so. The `unicodedata` module documentation is an
appropriate central place for our references to Unicode's own copious
documentation, so point there.
Also add to the isspace() test a thorough check that the
implementation agrees with the intended definition.
(cherry picked from commit 8c1c426a631ba02357112657193f82c58d3e08b4)
fix link to time function from time_ns doc (GH-15285)
Because mod, func, class, etc all share one namespace, :func:time creates a link to the time module doc page rather than the time.time function.
(cherry picked from commit 1b1d0514adbcdd859817c63d1410455c64660d78)
Paul Ganssle [Thu, 15 Aug 2019 19:09:37 +0000 (15:09 -0400)]
bpo-37642: Update acceptable offsets in timezone (GH-14878) (#15226)
This fixes an inconsistency between the Python and C implementations of
the datetime module. The pure python version of the code was not
accepting offsets greater than 23:59 but less than 24:00. This is an
accidental legacy of the original implementation, which was put in place
before tzinfo allowed sub-minute time zone offsets.
Replace usage of the obscure PEM_read_bio_X509_AUX with the more standard PEM_read_bio_X509 (GH-15303)
X509_AUX is an odd, note widely used, OpenSSL extension to the X509 file format. This function doesn't actually use any of the extra metadata that it parses, so just use the standard API.
faulthandler now allocates a dedicated stack of SIGSTKSZ*2 bytes,
instead of just SIGSTKSZ bytes. Calling the previous signal handler
in faulthandler signal handler uses more than SIGSTKSZ bytes of stack
memory on some platforms.
(cherry picked from commit ac827edc493d3ac3f5b9b0cc353df1d4b418a9aa)
Co-authored-by: Victor Stinner <vstinner@redhat.com>
bpo-37811: FreeBSD, OSX: fix poll(2) usage in sockets module (GH-15202)
FreeBSD implementation of poll(2) restricts the timeout argument to be
either zero, or positive, or equal to INFTIM (-1).
Unless otherwise overridden, socket timeout defaults to -1. This value
is then converted to milliseconds (-1000) and used as argument to the
poll syscall. poll returns EINVAL (22), and the connection fails.
This bug was discovered during the EINTR handling testing, and the
reproduction code can be found in
https://bugs.python.org/issue23618 (see connect_eintr.py,
attached). On GNU/Linux, the example runs as expected.
projects / thirdparty / Python / cpython.git / log
