Mailman 3 November 2017 - docs

[issue31950] Default event loop policy doc lacks precision
by Antoine Pitrou 07 Nov '17

07 Nov '17

New submission from Antoine Pitrou <pitrou(a)free.fr>: The doc for the default event loop policy states that """the default policy defines context as the current thread, and manages an event loop per thread that interacts with asyncio""". What it doesn't mention, though, is that you get a RuntimeError if you call get_event_loop() in a non-main thread -- you first have to call set_event_loop(new_event_loop()) in that thread. This wasn't expected by me. As a side note, customizing the event loop policy isn't trivial either: it seems it's best to subclass DefaultEventLoopPolicy and make adjustments by overriding the API methods. Writing your own wouldn't work, as e.g. the Unix default policy has custom initialization code for child watchers. ---------- assignee: docs@python components: Documentation, asyncio messages: 305607 nosy: docs@python, giampaolo.rodola, gvanrossum, haypo, pitrou, yselivanov priority: normal severity: normal status: open title: Default event loop policy doc lacks precision type: behavior versions: Python 3.6, Python 3.7 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31950> _______________________________________

3 6

[issue31953] Dedicated place for security announcements?
by Jean-Philippe Ouellet 06 Nov '17

06 Nov '17

New submission from Jean-Philippe Ouellet <jpo(a)vt.edu>: Hello, My apologies if this is not the right place to discus this. I would like to ensure that I stay informed of any potential future security issues in python (specifically at least the cpython runtime and standard library, although select very-popular 3rd party libraries wouldn't hurt). I cannot find a single place where such announcements are guaranteed to land. Good examples of the type of thing I am looking for are the openssl-announce list [1][2] and the golang-announce list [3], where the projects pre-announce "Hey, we're going to have a release on <date> which addresses a <rough severity> security issue in <vague area of impact>." and then announces again when patches are available such that responsible maintainers (such as I am trying to be) can ensure that updates are available to our users ASAP. The python-announce-list [4] does not serve this purpose because it has lots of noise from initial release announcements about random 3rd party stuff, and the "security news" page [5] is really just a "how to disclose vulns" page. Note that I'm *not* advocating for the creation of a pre-disclosure list! Python is such a ubiquitous piece of software that I don't think it's reasonable to expect that such a list could contain all affected parties without also leaking details to those who would cause harm. I'm only asking for something public that I can subscribe to in order to be sure I'll have a heads up of when patching is imminently required. Regards, Jean-Philippe (a contributor to the Qubes OS project [6] whose security relies mostly on Python's and Xen's - and is on Xen's pre-disclosure list) [1]: https://mta.openssl.org/pipermail/openssl-announce/2017-October/thread.html [2]: https://mta.openssl.org/pipermail/openssl-announce/2017-November/thread.html [3]: https://groups.google.com/forum/#!forum/golang-announce [4]: https://mail.python.org/mailman/listinfo/python-announce-list [5]: https://www.python.org/news/security/ [6]: https://www.qubes-os.org/ ---------- assignee: docs@python components: Documentation, email messages: 305614 nosy: barry, docs@python, jpo, r.david.murray priority: normal severity: normal status: open title: Dedicated place for security announcements? type: security _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31953> _______________________________________

2 2

Font quality in docs degraded in 3.6.3 in comparison to 3.6.2 -- please use type 1 fonts only
by n sigrist 05 Nov '17

05 Nov '17

Example: tutorial.pdf => Documents are compiled with pdflatex -- can you ensure to use type 1 fonts and avoid the old type 3 fonts. The latex settings just needs a tuning and you get higher quality pdf docs. All type 3 fonts can be replaced with type 1 fonts available in any TeX distribution. It makes the documents more readable and enjoyable. -- Norbert

1 0

[issue31923] Misspelled "loading" in Doc/includes/sqlite3/load_extension.py
by Davy Wybiral 04 Nov '17

04 Nov '17

New submission from Davy Wybiral <davy.wybiral(a)gmail.com>: Line #14 of Doc/includes/sqlite3/load_extension.py reads: # disable extension laoding again ---------- assignee: docs@python components: Documentation messages: 305423 nosy: davywtf, docs@python priority: normal severity: normal status: open title: Misspelled "loading" in Doc/includes/sqlite3/load_extension.py versions: Python 3.6 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31923> _______________________________________

3 8

[issue31668] "fixFirefoxAnchorBug" function in doctools.js causes navigating problem in Py3 doc in Chrome
by fireattack 04 Nov '17

04 Nov '17

New submission from fireattack <human.peng(a)gmail.com>: Problem This is a regression bug/flaw in Sphinx's doctools.js, a JS file used in its base template, and therefore got inherited to Python 3's documentation website. Python 2's documentation website is not affected because it's based on an older version of Sphinx. There is a function in doctools.js: /** * workaround a firefox stupidity * see: https://bugzilla.mozilla.org/show_bug.cgi?id=645075 */ fixFirefoxAnchorBug : function() { if (document.location.hash) window.setTimeout(function() { document.location.href += ''; }, 10); } This function was supposed to fix an anchor bug in Firefox (see comment). It *used to* have a condition of $.browser outside, so it will only be applied to Firefox; but it was removed in JQuery so it was removed here as well. Therefore, this function now applies to all the browsers. Unfortunately, it causes navigating problem in Chrome now, when you use back and forward. The problem STR (Chrome only): 1. Open a link with hash (anchor), e.g. https://docs.python.org/3/library/stdtypes.html#text-sequence-type-str 2. Scroll away from the anchor position. 3. Click any external link (means the link that is not an anchor of current page). 4. Hit "back" button in the browser. What happened: When you navigating back, it doesn't go to your previous position. Instead, it goes to the anchor's location. What it should do: It should go to your previous position. Ironically, it won't cause this problem in Firefox, despite it's supposed to be a fix for (a different) anchor bug in Firefox. Comments I reported it to Sphinx as well: https://github.com/sphinx-doc/sphinx/issues/3549 but didn't get any response so far. Please keep in mind the Firefox anchor bug mentioned above will only happen if the anchor is assigned with <section id>. From my observation we don't really use <section> in HTML, so I don't think this workaround function is really necessary here to begin with. ---------- assignee: docs@python components: Documentation messages: 303576 nosy: docs@python, fireattack priority: normal severity: normal status: open title: "fixFirefoxAnchorBug" function in doctools.js causes navigating problem in Py3 doc in Chrome type: behavior versions: Python 3.4, Python 3.5, Python 3.6, Python 3.7, Python 3.8 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31668> _______________________________________

3 3

[issue31678] Incorrect C Function name for timedelta
by Михаил Миронов 04 Nov '17

04 Nov '17

New submission from Михаил Миронов <phobosmir(a)gmail.com>: Documentation: PyDateTime_DELTA_GET_MICROSECOND https://docs.python.org/3/c-api/datetime.html#c.PyDateTime_DELTA_GET_MICROS… Should be: PyDateTime_DELTA_GET_MICROSECONDS ---------- assignee: docs@python components: Documentation messages: 303616 nosy: docs@python, Михаил Миронов priority: normal severity: normal status: open title: Incorrect C Function name for timedelta versions: Python 3.6 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31678> _______________________________________

4 5

[issue18669] curses.chgat() moves cursor, documentation says it shouldn't
by productivememberofsociety666 04 Nov '17

04 Nov '17

New submission from productivememberofsociety666: According to docs.python.org/2/library/curses.html#chgat (or any other version), curses.chgat() is not supposed to move the cursor at all. This is true if you don't give it x,y coordinates, but if you do, the cursor does in fact move to those coordinates, since it then uses C curses's mvchgat() internally and doesn't correct the position afterwards. See included file for demonstration: Expected output is: one two Actual output is: two Fix the documentation or fix the code :-) ---------- assignee: docs@python components: Documentation, Extension Modules files: chgat-bug.py messages: 194552 nosy: docs@python, productivememberofsociety666 priority: normal severity: normal status: open title: curses.chgat() moves cursor, documentation says it shouldn't type: behavior versions: Python 2.6, Python 2.7, Python 3.1, Python 3.2, Python 3.3, Python 3.4, Python 3.5 Added file: http://bugs.python.org/file31175/chgat-bug.py _______________________________________ Python tracker <report(a)bugs.python.org> <http://bugs.python.org/issue18669> _______________________________________

6 10

[issue31936] "5. The import system" grammatical error
by Daniel U. Thibault 03 Nov '17

03 Nov '17

New submission from Daniel U. Thibault <d.u.thibault(a)gmail.com>: https://docs.python.org/3/reference/import.html#importsystem "Other mechanisms for invoking the import system (such as importlib.import_module()) may choose to subvert __import__() and use its own solution to implement import semantics." should be "Other mechanisms for invoking the import system (such as importlib.import_module()) may choose to subvert __import__() and use their own solution to implement import semantics." ---------- assignee: docs@python components: Documentation messages: 305489 nosy: Daniel.U..Thibault, docs@python priority: normal severity: normal status: open title: "5. The import system" grammatical error type: enhancement versions: Python 3.8 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31936> _______________________________________

2 5

[issue31914] Document Pool.(star)map return type
by Дилян Палаузов 03 Nov '17

03 Nov '17

New submission from Дилян Палаузов <dilyan.palauzov(a)aegee.org>: https://docs.python.org/3/library/multiprocessing.html#multiprocessing.pool… says: starmap(func, iterable[, chunksize]) Like map() except that the elements of the iterable are expected to be iterables that are unpacked as arguments. Hence an iterable of [(1,2), (3, 4)] results in [func(1,2), func(3,4)]. If it was like map() then it would have returned an iterator. Please clarify, that Pool.map and Pool.starmap return list. ---------- assignee: docs@python components: Documentation messages: 305337 nosy: dilyan.palauzov, docs@python priority: normal severity: normal status: open title: Document Pool.(star)map return type type: enhancement versions: Python 3.6 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31914> _______________________________________

1 2

[issue31928] DOC: Show sys.version_info as a named tuple
by Cheryl Sabella 03 Nov '17

03 Nov '17

New submission from Cheryl Sabella <chekat2(a)gmail.com>: Invoking sys.version_info displays results as a named tuple, but the documentation shows the results as a tuple. ---------- assignee: docs@python components: Documentation messages: 305446 nosy: csabella, docs@python priority: normal severity: normal status: open title: DOC: Show sys.version_info as a named tuple versions: Python 3.7 _______________________________________ Python tracker <report(a)bugs.python.org> <https://bugs.python.org/issue31928> _______________________________________

1 2