Changes by Antoine Pitrou : ---------- stage: -> patch review versions: +Python 2.7, Python 3.2, Python 3.3 _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Serhiy Storchaka added the comment:

It seems to be the latter: Žiga Seilnacht

Then Misc/ACKS should be corrected too. ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Martin v. Löwis added the comment: There was a long-standing opposition by Guido to use UTF-8 in that file, and also complaints about legibility. Not sure what the current status is. It doesn't matter much to me, even though the spelling of my name is affected. ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Marc-Andre Lemburg added the comment: Thank you for taking the initiative. Regarding use of UTF-8 for text files: I think we ought to acknowledge that UTF-8 has become the defacto standard for non-ASCII text files by now and with Python 3 being all Unicode, it feels silly not make use of it in Python source files. Regarding my name: I have no issue with the apostrophe missing on the e. I've long given up using it in source code or emails :-) ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Ezio Melotti added the comment: Make sure that it's still possible to generate the pdf of the docs (with `make latex` and then `make all-pdf` in build/latex/). Latin1 should be fine, but IIRC non-latin1 will break (sorry Žiga). ---------- nosy: +ezio.melotti _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Ezio Melotti added the comment:

Indeed, PDF rendering is important. OTOH, Latex T1 does support Ž.

I tried to put a Ž directly in an rst source and indeed the pdf has been created correctly. ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Terry J. Reedy added the comment: I basically agree with Marc-André. Indeed, I think Acks, at least, should contain names in native spelling plus, if that is not Latin-based, a romanization. Three reasons: 1. I want English-speaking Python programmers to feel welcome to contribute and be acknowledged with what is visibly their native name. 2. I want people to be able to looks at the contributors list and see that Python is an international project. 3. Transliterations are ofter lossy, and in both directions. Romanizations do not necessarily tell contributors' compatriots how to address them in their mutual native language. If stdlib .py files are still restricted to romanizations, those could at least be looked up in Acks. But Python does come with an utf-8 unicode editor, Idle, that handles at least the BMP just fine (because tcl/tk does). ---------- nosy: +terry.reedy _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Chris Jerdonek added the comment:

While working on issue 15437, it occurred to me that storing the names in a structured form might come in handy.

In a separate discussion, Ezio pointed out a case where we are already scraping data about members from doc files: http://hg.python.org/tracker/python-dev/file/69984a770ab5/extensions/jnosy.p... So there is another case where we would benefit from storing names with associated attributes in a centralized, structured way. Sorry if this is a departure from the topic of this issue. I may create a separate issue for this or raise the idea in another forum. ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Changes by Serhiy Storchaka : Removed file: http://bugs.python.org/file26504/doc-nonascii-names.patch _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Serhiy Storchaka added the comment: Non-ascii characters already used in a lot (43-50) of doc files. LC_ALL=C find Doc/ -type f -name '*.rst' -exec egrep --color "$(printf '[\x80-\xFF]+')" '{}' + All touched files already contains non-ascii characters (and Misc/HISTORY contains invalid UTF-8 sequence). ---------- _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________

Antoine Pitrou added the comment: Ok, then I've committed the patch. Closing the issue now, thank you. ---------- resolution: -> fixed stage: patch review -> committed/rejected status: open -> closed versions: -Python 2.7 _______________________________________ Python tracker http://bugs.python.org/issue15444 _______________________________________