On Thu, Dec 16, 2021 at 05:43:43PM -0600, Jacob Bachmeyer wrote:
Gavin Smith wrote:
On Thu, Dec 16, 2021 at 11:46:08AM -0800, Per Bothner wrote:
I know nothing about it - appears to have been added by Gavin:
2020-11-25 Gavin Smith <gavinsmith0123@gmail.com>
data-manual attribute
* tp/Texinfo/Convert/HTML.pm (_convert_xref_commands):
Set data-manual attribute instead of class="texi-manual"
on links to other Texinfo manuals.
The purpose of this was to support locally installed HTML manuals better.
The idea was that the contents of the data-manual attribute could be used
for the name of a manual to be found in a search path.
Use of data-* attributes was suggested in another thread and is valid HTML5:
https://developer.mozilla.org/en-US/docs/Learn/HTML/Howto/Use_data_attributes
https://lists.gnu.org/archive/html/bug-texinfo/2020-10/msg00085.html
It is valid HTML5, but epub requires XHTML 1.1 and it is *not* valid there.
Most epub readers will *probably* silently ignore such invalid attributes,
but I would not be surprised if there is an embedded reader out there that
crashes in this case.
Would distinct HTML4, HTML5, and XHTML 1.1 output modes, probably with a
common base, be feasible? HTML4 is preferable as a strict document format
and would eschew custom attributes in favor of CSS classes. HTML5 is
preferable for "web applications", such as a Web-based JavaScript Info
reader. XHTML 1.1 is required for epub standards conformance.
We barely have the manpower to maintain one HTML output mode. I seriously
doubt we'd do a good job of maintining three. I say if there is a good
practical reason to output valid XML/XHTML then we should output that instead,
to the extent that epub readers actually need it.