Mail Archives: djgpp-workers/2003/04/05/15:17:58
Hello.
Juan Manuel Guerrero wrote:
[snip]
> Unfortunately I have been very busy the last few months so I had to stop
> following this newsgroup completely. This seems to have lead to some double
> efforts.
Oh well, these things happen. Actually, supporting 'hh', 'z', etc. in printf
wasn't actually that hard. I wrote some test cases for them (see
tests/libc/ansi/stdio/printf2.c), which we can still use. It is probably worth
checking that the test behaves the same way before and after applying your
patch.
I also added support to scanf (well, _doscan) for them.
Is there a similar positional patch for scanf? From looking at the Single Unix
Specification, version 2, it seems that scanf supports %n$ too.
[snip]
> Because I had to submit that version to the GNU gettext maintainer, I
> would like to see my code getting syncronized with actual CVS _doprnt.c.
> I will either remove my code, or adjust it, or in a last resort, I will
> have to remove your code from _doprnt.c
I think removing my code from _doprnt may be the easiest way to do it. I can't
remember if I made any bugfixes at the same time. I don't think so. But it's
worth checking the comments in CVS log and doing some diffs. You could use the
web interface to CVS for that:
http://www.delorie.com/bin/cvsweb.cgi/djgpp/src/libc/ansi/stdio/doprnt.c
I fixed some bugs in _doprnt or _doscan or perhaps both. I can't really
remember. IIRC I fixed a buffer-overflow bug in _doscan, when I added support
for 'hh', 'z', etc.
> Please let _doprnt.c one or two days unaltered, so I can adjust and submit
> my proposed changes.
OK.
Thanks, bye, Rich =]
--
Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]
- Raw text -