On Fri, May 10, 2019 at 05:34:48PM +0100, Gavin Smith wrote: > On 5/10/19, Patrice Dumas <[email protected]> wrote: > > On Thu, May 09, 2019 at 07:13:05PM +0100, Gavin Smith wrote: > >> I think there should be a more reliable way to identify links to > >> index nodes from the table of contents in index.html in HTML output. > >> Currently the only way is to check if "Index" appears in the name of > >> the page. This would be used in an "HTML-Info" system, providing the > >> functionality of Info using HTML. > > > > I may be missing something, but it seems to me that the first node with > > printindex is used, see _prepare_global_targets() in > > Texinfo/Convert/HTML.pm. > > > > But maybe you are referring to something else? > > I hadn't realised that a <link rel="index"> element was already > output in the file header. I was thinking of the <a> elements in the > table of contents or main menu. But the <link> elements are good too. > The only change I'd make is to output a <link> element for every node > containing a @printindex, instead of just the first one. Do you think > that would be okay?
That would be possible to do that, and it seems to be ok in term of validity of the document. But I do not know how it will be interpreted by the webbrowser. I tried to search on the internet some information on <link> with the same rel= used multiple time but I didn't find anything interesting. In any case, I guess it would be ok to do it, as it seems to me that no webbrowser uses the <link rel= > informations we provide in the documents. -- Pat
