Zsh Mailing List Archive
Messages sorted by:
Reverse Date,
Date,
Thread,
Author
zsh-3.1.7
- X-seq: zsh-workers 11658
- From: Peter Stephenson <pws@xxxxxxxxxxxxxxxxxxxxxxxxx>
- To: zsh-workers@xxxxxxxxxxxxxx (Zsh hackers list)
- Subject: zsh-3.1.7
- Date: Tue, 30 May 2000 12:31:32 +0100
- Mailing-list: contact zsh-workers-help@xxxxxxxxxxxxxx; run by ezmlm
I have finally uploaded this. Announcements and tidying up will follow
eventually. There's no particular hurry, since it's best if it propagates
to mirrors before all and sundry try to retrieve it. I get a clean
compilation and corect tests on Linux and Solaris with this.
If anyone happens to know how to release a module on sourceforge, the files
zsh-3.1.7.tar.gz zsh-3.1.7-doc.tar.gz are in /home/groups/zsh. Otherwise
I'll look at it sometime.
Now it's out, anything held over from checking in might as well go in ---
even if something turns up and we need to produce a revision soon, it might
as well have the latest stuff in it. I don't envisage having to branch the
archive at this point (but I would say that).
I have spent ridiculous amounts of time fighting with trivial problems
including but not limited to:
- modlist.yo wasn't built properly here. Probably another feature of
Solaris sed which needs working round. (Not yet done.)
- modlist.yo wasn't in .distfiles, so the compiling machine always thought
it had to remake it. This was bad news if you didn't have yodl. (Fixed.)
- making the doc distribution doesn't work on sourceforge because tex
etc. aren't there, unless anyone knows better. Argh. I've just made it
here. I think as long as the YODL files are up-to-date before making
that should be OK.
- The link from sourceforge to zsh.org was bad, presumably temporary.
- The autoconf versions there and here differ in producing tests for awk
versions, one giving `mawk gawk', the other `gawk mawk'. It's not
important, but I suppose it's a good idea to try to synchronise autoconf
versions as much as possible.
- I'm also unhappy about the way autoconf puts in the cygwin test --- it
seems to run into the mingw32 test incorrectly (line 1558). I don't know
if an upgrade might help that. I haven't dared change it by hand.
--
Peter Stephenson <pws@xxxxxxxxxxxxxxxxxxxxxxxxx>
Cambridge Silicon Radio, Unit 300, Science Park, Milton Road,
Cambridge, CB4 0XL, UK Tel: +44 (0)1223 392070
Messages sorted by:
Reverse Date,
Date,
Thread,
Author