This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [libstdc++] automate documentation.html changes, basic_string doxygen hooks
- From: Phil Edwards <phil at jaj dot com>
- To: Jonathan Wakely <cow at compsoc dot man dot ac dot uk>
- Cc: libstdc++ at gcc dot gnu dot org
- Date: Tue, 3 Sep 2002 14:47:33 -0400
- Subject: Re: [libstdc++] automate documentation.html changes, basic_string doxygen hooks
- References: <20020831203328.A15650@disaster.jaj.com> <20020902133751.GA73210@compsoc.man.ac.uk>
On Mon, Sep 02, 2002 at 02:37:51PM +0100, Jonathan Wakely wrote:
> On Sat, Aug 31, 2002 at 08:33:28PM -0400, Phil Edwards wrote:
>
> > string, wstring, and basic_string hav never been doxygenated. This adds
> > the initial hooks to make them at least show up the output.
>
> Am I right in thinking the algorithm Doxygen comments don't come out in
> the manpage versions yet?
You are correct. I don't know why not.
There are problems with comments from <ext/algorithm> overwriting those in
<algorithm>, and doxygen even complains about it, but the only solution
it understands is "add a prefix to <algorithm> to distinguish it from
<ext/algorithm>". Grr.
> Once I've got the wwwdocs as XHTML I'll be glad to continue adding Doxygen
> comments to the sources, starting with basic_string if you like.
basic_string or anything else, as you prefer. I've started
pseudo-standardizing the comments used for container claases for
consistency's sake. You might save yourself a chunk of time by copying
text from stl_{vector,list,deque,whatever}.h.
> I've merged this with my XHTML changes (adding </li> was required there
> anyway :-) and found a couple of problems:
>
> [[:digit:]] as a shell wildcard is a Bash-ism (well, POSIXsim, whatever),
> which chokes /bin/sh on BSD.
> I don't know if [0-9] is totally portable either, but it WorksForMe.
Sigh. Let's use [0-9].
> $(wildcard ...) is a GNU make feature, not supported on BSD make, so you
> can't regenerate documentation.html without either marking that target
> as .PHONY or using gmake (or so my 10 minutes testing indicated.)
Yes, hence my comment. :-) If there's a portable why of doing that,
I'm all ears.
> I think I need to (slightly) change makedoc.awk to convert the docs to XHTML,
> but I'll check that against a clean tree and submit it with the XHTML
> patch.
Cool.
Phil
--
I would therefore like to posit that computing's central challenge, viz. "How
not to make a mess of it," has /not/ been met.
- Edsger Dijkstra, 1930-2002