From: Alexey K. <ak...@ib...> - 2020-03-08 14:36:58
|
Hello, I think we don't need chuncked html. Regards, Alexey On 08.03.2020 16:13, Mark Rotteveel wrote: > As I'm looking at the asciidoc conversion, our current choice of > having a (single) PDF and a chunked HTML output on the website is a > bit of a pain point. > > Asciidoctor does not support chunked HTML output (there is an > experimental, third-party plugin, but that plugin is tied to a > specific, out-of-date asciidoctor version). > > The current options to produce chunked HTML output are: > > 1. Use separate asciidoc files per section > 2. Generate docbook from asciidoc and then use xsl to generate chunked > HTML > > Option 1 has the downsides that to generate a PDF, an additional > 'include' file needs to be maintained to be able to generate a single > document in addition to the 'chunked' files. This has maintenance > overhead and could be something that can easily be overlooked when > adding sections. I also haven't yet figured out how to have navigation > between sections in the (faked) chunked output this way. > > This last concern might be addressable by using > Antora[https://antora.org/] to generate the HTML documentation (eg see > https://docs.antora.org/antora/2.2/ for an example), but that feels > like a huge increase in scope. > > Option 2 might run into some problems as asciidoctor generates docbook > 5, not docbook 4.5, so I'm not sure if this would cause problems with > our current stylesheets. > > Option 2 seems to be the least intrusive way to retain chunked HTML > output. However, before I spend a lot of time going that route, I'm > wondering: > > - Do we need to retain the ability to generate chunked HTML output? > - And, what is the reason that we currently use chunked HTML output? > > Personally, I'm not really a fan of chunked HTML (searching through a > single page HTML document is much simpler), but maybe I'm overlooking > something. > > Mark |