Yes, the rumors are true: I'm planning to release the documentation for Python 1.5.2 one more time. The current target date is 17 March. Here's my planned schedule for short-term activity: Feb 29: Release documentation packages to Doc-SIG for review Mar 17: Release documentation version 1.5.2p2 (last 1.5.2 release) Apr 17: Finish merging 1.5.2p2 into documentation for 1.6 There are a couple of things I'd like help with. ;) First, I try to go throught the HTML version every so often using major browsers on major platforms. I *don't* have access to all platforms, though, and just don't have time to test for each major browser for each platform, and that's only gotten worse the last couple of releases. Second, I haven't been able to "close-read" the documentation for a long time; this is valuable to catch confusing passages, the random typo, and effects of whitespace-gobbling nanoviruses. Catching statements that are just plain wrong is also an important goal. Clearly, the second goal can be met while meeting the first goal; if you shoot for the second, the first falls out, at least for one browser/platform combination, but it requires a more substantial time commitment than browser-testing the HTML. I've often used a printed version for the close-reading instead of using the HTML. What I'm doing is asking for volunteers to help; with enough eyes, the work is highly partionable. If each of us picks one or more documents (the library reference can be partitioned by chapter) and a particular browser/platform combo, this shouldn't be too hard. I'd like to hit at least the following browser/platform combinations: MSIE 5.0 Windows, MacOS(?) Navigator 4.x Windows, MacOS, Unix Opera Windows, MacOS(?), Linux Mozilla M13 Windows, MacOS, Unix (I haven't been able to run the M13 binary for Solaris 2.6 on my 2.7 box; has anyone succeeded at this? The M12 runs OK... for a little while. I've not yet checked M13 for other platforms.) Could someone with a MacOS box please check that all the filenames, at least for the generated HTML, aren't too long or otherwise cause problems with HFS? Thanks! I can coordinate as well as do a portion of the work; who'll volunteer? -Fred -- Fred L. Drake, Jr. <fdrake at acm.org> Corporation for National Research Initiatives
On Mon, 28 Feb 2000, Fred L. Drake, Jr. wrote:
What I'm doing is asking for volunteers to help; with enough eyes, the work is highly partionable. If each of us picks one or more documents (the library reference can be partitioned by chapter) and a particular browser/platform combo, this shouldn't be too hard.
Okay, post some URLs... i have Navigator 3 and 4 on Irix and Linux.
I'd like to hit at least the following browser/platform combinations:
MSIE 5.0 Windows, MacOS(?) Navigator 4.x Windows, MacOS, Unix Opera Windows, MacOS(?), Linux Mozilla M13 Windows, MacOS, Unix
Is Mozilla even worth it? I've downloaded two or three releases and never had it stay up for more than a couple of minutes even on ordinary pages, or on its own included test pages. -- ?!ng
Ka-Ping Yee writes:
Okay, post some URLs... i have Navigator 3 and 4 on Irix and Linux.
I've dropped tarballs at: ftp://ftp.python.org/pub/python/doc/.test/ Online versions are available at: http://www.python.org/doc/1.5.2p2/ (not yet linked to from anywhere).
Is Mozilla even worth it? I've downloaded two or three releases and never had it stay up for more than a couple of minutes even on ordinary pages, or on its own included test pages.
I tried M12 on Solaris, Linux, and Windows, and it mostly worked OK (HTML widget portion of the display), aside from some weirdness on Solaris, and generally leaking like a sieve. I still haven't played with M13 on Linux or Windows. It still seems slow, but that's probably an issue of the application code rather than the renderer (feels snappy); I suspect they're not caching network resources, or have a bug in that. I'm not so concerned about Mozilla now as six months from now; they *are* becoming much more stable & capable. I don't want to feel I need to do another release simply because they've fixed the usability issues related to their interface. But I've not intention of holding anything up for Mozilla either. -Fred -- Fred L. Drake, Jr. <fdrake at acm.org> Corporation for National Research Initiatives
how do you want us to relate "errors" back to you? directly, or on this list, or on another list, or bug-tracker? i've been programming python for a while, but i would just gloss over subtle python inconsistencies. so i'll stare through the html code, run it through the w3c validator (they do matching of tags and such which is useful), and tell my friends to sift through it with their browsers. (i know various folks at companies that do web-sites so they have a lot of the browser/os/platform test-beds, i think). i can get a more extensive listing of browser/os/platform combinations that i'm checking soonish. it'll likely be versions of Netscape and IE on Mac and Windows and Linux platforms (where applicable). beyond that, i'm running IE 5.0 on WinNT 4.0 and Win2000 and M13 on WinNT 4.0 (which i haven't had that many problems with other than the ui). i've found a few problems with the code posted at the url. lines 267 and 276 of .../lib/typesseq-strings.html have stray unmatched <var> and </var> tags. looking at the web-page (not the html), it's in section 2.1.5.2 Mutabe Sequence Types in the table. you can actually see two of the parens are in a var enclosure and shouldn't be. 9th and 12th row, first column in the table under 2.1.5.2. also links on the modules and index pages turn up 404's. i may untar the tarball and see if i have the same problems. /will
-----Original Message-----
Ka-Ping Yee writes:
Okay, post some URLs... i have Navigator 3 and 4 on Irix and Linux.
I've dropped tarballs at:
ftp://ftp.python.org/pub/python/doc/.test/
Online versions are available at:
http://www.python.org/doc/1.5.2p2/
(not yet linked to from anywhere).
Is Mozilla even worth it? I've downloaded two or three releases and never had it stay up for more than a couple of minutes even on ordinary pages, or on its own included test pages.
I tried M12 on Solaris, Linux, and Windows, and it mostly worked OK (HTML widget portion of the display), aside from some weirdness on Solaris, and generally leaking like a sieve. I still haven't played with M13 on Linux or Windows. It still seems slow, but that's probably an issue of the application code rather than the renderer (feels snappy); I suspect they're not caching network resources, or have a bug in that. I'm not so concerned about Mozilla now as six months from now; they *are* becoming much more stable & capable. I don't want to feel I need to do another release simply because they've fixed the usability issues related to their interface. But I've not intention of holding anything up for Mozilla either.
-Fred
-- Fred L. Drake, Jr. <fdrake at acm.org> Corporation for National Research Initiatives
_______________________________________________ Doc-SIG maillist - Doc-SIG@python.org http://www.python.org/mailman/listinfo/doc-sig
For everyone: I've posted new packages at: ftp://ftp.python.org/pub/python/doc/.test/ This version is also online at: http://www.python.org/doc/1.5.2p2/ A major, systematic navigational failure reported by Peter Funk has been fixed, as well as a number of small fixes being integrated, and a note on the removal of multi-arg [].append has been added. will writes:
how do you want us to relate "errors" back to you? directly, or on this list, or on another list, or bug-tracker?
The bug tracker is probably the best, but python-docs@python.org is good too.
i've been programming python for a while, but i would just gloss over subtle python inconsistencies. so i'll stare through the html code, run it through the w3c validator (they do matching of tags and such
I don't know that the HTML is good to look at; it definately shows that it's generated! The validator is a good idea.
which is useful), and tell my friends to sift through it with their browsers. (i know various folks at companies that do web-sites so they have a lot of the browser/os/platform test-beds, i think). i can
I hope you're only asking people that don't mind checking this stuff for free!
beyond that, i'm running IE 5.0 on WinNT 4.0 and Win2000 and M13 on WinNT 4.0 (which i haven't had that many problems with other than the ui).
Great! Hopefully I'll manage to install M13 on my home machines this weekend.
i've found a few problems with the code posted at the url. lines 267 and 276 of .../lib/typesseq-strings.html have stray unmatched <var> and </var> tags. looking at the web-page (not the html), it's in section 2.1.5.2 Mutabe Sequence Types in the table. you can actually see two of the parens are in a var enclosure and shouldn't be. 9th and 12th row, first column in the table under 2.1.5.2.
I see -- this will be hard to fix, but I know where the problem is. I'll see what I can do. For anyone here who really understands LaTeX2HTML's implementation (there are so many people in that group! ;), the problem is this: in perl/python.perl:do_cmd_optional(), I need to be able to determine if the current markup context include <var>. If so, it needs to be closed and re-opened (potentially along with other elements) to insert "unmolested" square brackets. If anyone has any ideas on this, or a patch to perl/python.perl, I'd love to hear from you! ;)
also links on the modules and index pages turn up 404's. i may untar the tarball and see if i have the same problems.
Try the new version, & let me know which specific links are broken if that's still a problem. -Fred -- Fred L. Drake, Jr. <fdrake at acm.org> Corporation for National Research Initiatives
participants (3)
-
Fred L. Drake, Jr. -
Ka-Ping Yee -
will