Guido van Rossum [Fri, 23 May 1997 21:21:43 +0000 (21:21 +0000)]
Added last mtime on whole faq.
Added more elaborate explanation of version conflict.
Set fake header on commit so show() shows the new mtime.
Reorder name/email fields to be after the log message.
Guido van Rossum [Fri, 23 May 1997 19:18:35 +0000 (19:18 +0000)]
Nits.
Use TARGET=rlog instead of _blank for rlog output.
Add missing mailto: to author's email address in Last changed link.
Forget about the domainname in the cookie; the default is just fine.
Guido van Rossum [Fri, 23 May 1997 18:53:06 +0000 (18:53 +0000)]
Reorg TO DO list.
Run rlog commands as pipe so we can cgi.escape(output).
Added self.escape(), like cgi.escape() but also escapes " -- for use
as double-quoted attribute values inside HTML tags.
Fix complains from Mark Lutz: HTTP_HOST is not always set.
Guido van Rossum [Fri, 23 May 1997 17:45:04 +0000 (17:45 +0000)]
- Suppress the visible text in the epilogue when "edit=no" is specified.
- Add translation of *foobar* into <I>foobar</I> (but not inside <PRE>).
- Optimization for the translation: only translate when @ or / or * seen.
- Add some layout to front page lay-out.
Jack Jansen [Fri, 23 May 1997 15:35:14 +0000 (15:35 +0000)]
- Added Py_GetProgramFullPath
- Added PyMac_StopGUSISpin and have it called at exit time (fixes crash
when exiting with sockets open)
- Added PLstr... funcs needed by gusi: those provided by StdCLib are
broken under cfm68k
Guido van Rossum [Fri, 23 May 1997 00:01:41 +0000 (00:01 +0000)]
PyObject_Compare can now return an error. Unfortunately, there are a
few places where we don't know how to test for them without losing
speed; don't know yet how to handle that.
Guido van Rossum [Thu, 22 May 1997 15:21:57 +0000 (15:21 +0000)]
Mostly small changes.
- Fixed broken HTML in edit form that swallowed <BR> tags
- Removed "another one" link in roulette, replaced with "use Reload" message
- show() now has edit=1 by default; changed all calls
- Display last-changed date, author, email when present and edit=1
- Added back pointers at bottom
- Added a copy of the bootstrap script at the end of the file
Guido van Rossum [Wed, 21 May 1997 21:31:39 +0000 (21:31 +0000)]
Lots of new features:
- got rid of the separate search page
- added an index (lists the questions with links to the answers)
- add a mechanism to add new entries
- add a way to list most recently edited entries first
Guido van Rossum [Tue, 20 May 1997 18:21:34 +0000 (18:21 +0000)]
Got rid of c_error in favor of errno (and EDOM/ERANGE).
Assume that errno usage is thread-safe -- most vendors do this by
making in a macro that refers to a per-thread storage area.
Guido van Rossum [Tue, 20 May 1997 17:06:11 +0000 (17:06 +0000)]
Plug leak of stack frame object in exception handling code.
Also delay DECREF calls until after the structures have been updated
(for reentrancy awareness).
Guido van Rossum [Fri, 16 May 1997 16:21:38 +0000 (16:21 +0000)]
Address the following problem on DOS and Win 3.1, reported by Jim
Ahlstrom:
Arraymodule.c has static functions H_getitem and h_getitem, and a
few others which differ only in case. These are a problem on
Windows 3.1, since a case-sensitive link causes Winsock to fail
(hey, it's not my fault). Please convert H_etc to HH_etc etc.
Guido van Rossum [Fri, 16 May 1997 14:23:33 +0000 (14:23 +0000)]
Got rid of all the last_name_* bogosities. I don't think the
complexity saved much any more. A simple benchmark (grail) showed
that there were 3 times as many misses as hits, and the same number of
times again the code was bypassed altogether due to the existence of
setattro/getattro.
Guido van Rossum [Thu, 15 May 1997 21:43:21 +0000 (21:43 +0000)]
Made a start with api.tex, the Python-C API Reference Manual.
Removed extref.tex (which provided the starting point).
Also removed qua.tex, which is out of date and no longer needed.
Barry Warsaw [Thu, 15 May 1997 18:27:49 +0000 (18:27 +0000)]
Catch nis.error and raise TestFailed instead. This catches the
problem where Python is configured with the nismodule but NIS isn't
installed, or the map names don't correspond to those hardcoded in
nismodule.c (which is bogus in and of itself).