Re: A Roadmap for TexInfo without Info

[Per Bothner]

On 11/26/2016 09:24 AM, Gavin Smith wrote:
> On 26 November 2016 at 16:59, Per Bothner <address@hidden> wrote:
> > > The important issue is (1).  Making the new format readable with Web
> > > browsers is the correct direction, IMO, but it needs to solve the few
> > > basic deficiencies in the current browsers due to which they still
> > > cannot beat any Info reader.  One of the most important ones is
> > > index-based search, for example.
> >
> >
> > I'm not sure what you mean by "index-based search".  That sounds like
> > searching by making use of a pre-computed index, which is different from
> > plain searching the actual text, which I though our goal is.
>
> HTML can't be a full replacement for Info unless features like the
> index-based search are available.

Your phrasing confuses the format with the application.  Both HTML and Info are
file formats, and neither has "index-based search available".  The applications
we care about can either calculate on-the-fly whatever tables and
indexes they need, or we can have makeinfo put whatever we want into the
HTML file(s).  If it's stuff we don't want to display when JavaScript is 
disabled,
then we can put it either in a comment, or an attribute, or (for big tables)
in a separate HTML file that JavaScript can load into an <iframe>.

I'm still unclear what you mean by "indexed-based search".  Do you mean
the 'i' / 'index-search' command?  That's easy to implement, especially
if we use a single non-split HTML file: Find the element with the index;
search through the index; navigate to the corresponding entry.  If the manual
is split there is an extra step to load the index in a hidden <iframe>.

Plain-text search is similar but we have more places to search (the entire
manual), plus that the search has to navigate the element hierarchy.

> This kind of technical problem will be resolved when someone produces
> something useful, like working on one of the projects I mentioned in
> this message
>
> http://lists.gnu.org/archive/html/bug-texinfo/2016-11/msg00008.html
> * modify texi2any to have an option to output HTML5
> * create a HTML documentation reader in JavaScript with advanced features
> * work on distribution or installation of HTML files. for example,
> this could be done by a GNU/Linux distribution that wants to make HTML
> documentation available.
>
> Another project would be to process HTML manuals in Emacs Lisp.

All these are part of the roadmap.  I think it is good to have a roadmap
so we know what we're aiming for, and what the useful sub-projects are.
--
        --Per Bothner
address@hidden   http://per.bothner.com/