On Sun, 14 May 2000, Dave Love wrote:
> >>>>> "EZ" == Eli Zaretskii <[EMAIL PROTECTED]> writes:
>
> EZ> (1) how to transform the
> EZ> cross-references, including to other Texinfo documents (assuming those
> EZ> were also converted to HTML),
>
> Yes, but I think that just needs a convention defining.
In principle, yes; but the devil is in the details ;-).
It's not trivial. In order to work, there's a need for a universal
agreement with whatever standard we adopt, and that's not easy,
considering texi2html and whatever tools are out there. More about
this below.
> EZ> and (2) how to make splitting work right even when makeinfo does
> EZ> an extra pass (e.g., to support @contents in the beginning of the
> EZ> document).
>
> Do you remember what was wrong with it?
Not off the top of my head, no. I just remember several annoying
complications, on top of makeinfo's already not-so-clean architecture,
to say the least.
> As far as I know, it worked right with @contents
Previously, @contents was working correctly only if it was near the
end (except in TeX). Now you can put @contents anywhere in the
document. This required to add a second pass.
> EZ> The second group of problems was the reason that I suggested a
> EZ> separate program to implement the split stage. I even sent to
> EZ> Karl a program I wrote for the DJGPP FAQ list that splits an HTML
> EZ> document by nodes. (Dave, if you are interested, I can send it
> EZ> to you as well.)
>
> If it implements the link meta-information, OK, but I don't see that
> in the version @delorie.com.
Sorry, I don't understand what do you mean by link meta-information.
> EZ> However, since Texinfo 4.0 was already very late, we never
> EZ> finalized these issues.
>
> :-( I'd probably have worked on it if asked, as I'd already put in a
> good deal of work after rms leant on me and it seemed to have been
> dumped.
Given the fact that the pretest was very long, Karl decided to release
without split-HTML support. So it was not a question of somebody who
was willing to work on this. And your code is all still there, modulo
some minor changes, so it's not lost.
> EZ> There were other problems as well (e.g., what file-naming scheme
> EZ> to use for the split files), but I think the above are the main
> EZ> ones.
>
> Do you remember what was wrong with what I chose?
The names themselves were not that bad (except for the 8+3 DOS names,
but that could be solved). The problem was how do you compute a
reference to another HTML file that was also split: you don't know the
name of the specific sub-file where the referenced node lives in the
other document, and HTML files lack tag tables that remove this
problem in Info.
