Re: [Scrollkeeper-devel] move to libxml2

[<las...@Su...>]

Norman Walsh wrote:
> =

> / L=E1szl=F3 Kov=E1cs <las...@Su...> was heard to say:
> | the new XSL stylesheet based TOC extraction. libxml2 version 2.3.8
> | contains a DocBook SGML parser that creates a DOM tree from an SGML
> | document.
> =

> From any SGML document, or just a DocBook document? How general is this=
?
> Does it support tag omission, etc.?
> =

>                                         Be seeing you,

DocBook only. And I wouldnt think it is bug free. After using it for 2-3
days I have submitted about 3-4 patches about tags not being handled
correctly. Also the encoding detection is a major problem. It doesnt
seem to be possible for DocBook SGML files. I think the DcoBook SGML
parser is just an intermediary solution, people would be expected to use
DocBook XML for docs, not DocBook SGML.

Laszlo