This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

PATCH: convert online docs to XHTML


The attached bzip2'd patch is a 6000 line unified diff to convert the
libstdc++-v3/docs/html tree to XHTML 1.0 Transitional, following the
main GCC policy for the online docs.
(I can submit it again as a context diff if preferred, but the unified
diff is much smaller and IMHO is more readable for these HTML changes)

The XHTML pages can be seen at http://www.compsoc.man.ac.uk/~cow/tmp/gcc/

http://gcc.gnu.org/codingconventions.html says not to change anything
in 17_intro without permission.  I've changed contribute.html, howto.html
and license.html, I'm pretty confident those files aren't taken from
upstream pkgs or auto-generated, so I'm assuming it's OK?
I didn't change the LWG lists or auto-generated pages.

I changed "op()" to "operator()", and marked up a couple of other names
as <code>. I diffed the formatted output from the pages (using lynx -dump)
to check I didn't change any of the text, only markup. I've not changed
any text apart from the above operator, moving a sentence out of a title
(in "What the standard says" in 22_locale/locale.html, "See chapter 22 in
the standard" is now below the heading, not part of it), and fixing a
couple of spelling mistakes due to typos (e.g. "replcement")

All the pages validate as XHTML if they have the boilerplate header added
to specify the DOCTYPE, XML namespace etc. so if this patch goes in the
server can preprocess the libstdc++ pages (except for the four listed above
that I didn't alter, not sure if they can be excluded easily?)

I had to change makedoc.awk so that the <ul> and <li> elements are nested
correctly. The generated output looks the same, but is valid XHTML now, so
I think I've done it right. This change is at the very top of the patch,
if you want to check it without trawling through all the HTML.

I could knock together a bit of text listing the main things to look out
for when updating the docs, to ensure they're still valid XHTML. Would
that be useful?


2002-09-04  Jonathan Wakely <jw@kayari.org>

	* docs/html/makedoc.awk: Nest elements correctly for XHTML conversion.
	* docs/html/configopts.html, docs/html/explanations.html,
	docs/html/install.html, docs/html/17_intro/contribute.html,
	docs/html/17_intro/howto.html, docs/html/17_intro/license.html,
	docs/html/18_support/howto.html, docs/html/19_diagnostics/howto.html,
	docs/html/20_util/howto.html, docs/html/22_locale/codecvt.html,
	docs/html/22_locale/ctype.html, docs/html/22_locale/howto.html,
	docs/html/22_locale/locale.html, docs/html/22_locale/messages.html,
	docs/html/23_containers/howto.html, docs/html/24_iterators/howto.html,
	docs/html/25_algorithms/howto.html, docs/html/26_numerics/howto.html,
	docs/html/27_io/howto.html, docs/html/ext/howto.html,
	docs/html/ext/sgiexts.html, docs/html/faq/index.html: Convert to XHTML.


Jon


--
"It is practically impossible to teach good programming to students
 that have had a prior exposure to BASIC: as potential programmers
 they are mentally mutilated beyond hope of regeneration."
          - Edsger Dijkstra

Attachment: xhtml.patch.bz2
Description: Binary data


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]