Re: ready for review: qcad (essentially done, I think)

[J Chapman Flack <flack%cs.purdue.edu@localhost> See http://spamassassin.org/tag/ for more details. Report problems to http://sf.net/tracker/?func=add&group_id=1&atid=200001 0.0 SF_CHICKENPOX_PERIOD BODY: Text interparsed with . 0.0 SF_CHICKENPOX_SLASH BODY: Text interparsed with / 0.0 SF_CHICKENPOX_MINUS BODY: Text interparsed with - 0.0 SF_CHICKENPOX_UNDERSCORE BODY: Text interparsed with _ 0.0 SF_CHICKENPOX_APOSTROPHE BODY: Text interparsed with ' -0.0 AWL AWL: From: address is in the auto white-list]

Oops, I think I replied to Thomas yesterday instead of to the list.

> pkglint is slightly unhappy:
> WARN: Makefile:41: Trailing white space.
> WARN: Makefile:15: USE_X11 should be set to YES or yes
> WARN: Makefile:16: GNU_CONFIGURE should be set to YES or yes

fixed, committed.

> DISTNAME=               qcad-2.0.4.0-1.src
> PKGNAME=                qcad-2.0.4.0
> You can use five digits too, just make it 2.0.4.0.1.

That's good to know, but at the moment I haven't changed it.  I'm not sure
what ribbonsoft intends with the -1, I've never seen -<anything but 1> on
any of their downloads.  I trust the package could be updated at any time
to use a fifth digit if it becomes necessary - it shouldn't create any
ambiguity with version checking or break anything - and if it doesn't
become necessary I think keeping the numbering simple is a Good Thing.
What do others think?

> # XXX how to determine the below dynamically?
> MAKE_ENV+=              QMAKESPEC=${QTDIR}/mkspecs/default
> PREPEND_PATH+=          ${QTDIR}/bin
> What do you mean?

oops, forgot to remove that comment, fixed now.  I originally had
mkspecs/netbsd-g++ because I hadn't noticed default, wondered how to
choose the mkspecs dynamically when building on other systems, and
Iain Hibbert set me straight about default.

> How does is this supposed to work when you build from source manually?
> Do you have to go into each dir and do 'make prepare' manually?

To build directly from the ribbonsoft distribution, you run a shell script
instead of make, and the script does that.  I thought about writing a very
trivial Makefile that just runs the shell script, but I decided it was worth
better integrating into the pkgsrc way of doing things (for benefit of
buildlink3 etc.) by reverse-engineering the script and getting the Makefile
to do the right things.  Any dissent? :)

> The Makefile.common is so short, I'd suggested just including
> it in the appropriate Makefiles.

Even if it's short though, I like having it factored out, for that much less
chance that somebody changes something in one place and misses duplicates.
Is a short Makefile.common a problem for anything?

> The TODO contains no real items.

It was originally my list of 'am I doing this right?' questions; it got a
lot shorter after Iain's suggestions.  It can disappear when we're sure
there are no real items left.  Maybe that's now.  :)

Peter I. Hansen writes:
> I built it on 1.6 with no problems.

Cool. Did you try qcad-manual-<your language here> and qcad-partlibrary also?

Btw, the DESCR for -partlibrary explains what to do in QCad to actually *see*
the partlibrary after it's installed (because the distributed version of the
manual seems to omit that detail).  I'm thinking maybe that should be
a MESSAGE as well, no?

Thanks for testing!

-Chap


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
pkgsrc-wip-review mailing list
pkgsrc-wip-review%lists.sourceforge.net@localhost
https://lists.sourceforge.net/lists/listinfo/pkgsrc-wip-review