From: Reece Dunn (msclrhd_at_[hidden])
Date: 2003-04-17 08:53:28
William E. Kempf wrote:
> > NOTE: I am thinking about writing an XSLT file that will generate >
>Already done. You're reinventing the wheel, and should
>instead be helping out the development efforts of BoostBook.
I said that I was *thinking* about it. Anyway, I'd be happy to help out the
BoostBook development efforts.
I am currently using a different XML structure/XSLT fileset that I have been
using for a while now, that has features such as MathML support and unicode
mappings for IE6 (as XSLT -> HTML/CSS) based on David Carlisle's work at
I have updated this XSLT code to use a CSS file passed as a parameter to it.
The syntax highlighter now uses CSS class types so the CSS file can
configure the highlighting accordingly. I am now using boost.css for the
boost version of the document, and default.css for my internal version (no
difference except the CSS file used).
boost.css provides black on white text with softer colours for the syntax
highlighting based on the feedback I have recieved.
>Regarding the use of names starting with double underscore:
This has now been fixed in both identifiers and header guards.
>" __cdecl" - This is Microsoft-specific, so it may be a good idea to remove
>(it can always be set in the project settings, instead).
This has now been removed from the test files.
>Some headers, such as list_output.hpp, are not stand-alone. Including it
>gives several errors about undefined names.
I am working on this.
>Some names don't conform to Boost's coding standard, such as
>"spaceAroundList()" (instead of "space_around_list()").
This has been fixed to conform to Boost coding standards.
 Work on the documentation
This is a work-in-progress.
 Clean up the code
I have made progress in this area, by removing redundant constructs.
 Examples and test files
I have split the examples based on the technology that they use.
I have begun to document where I want to extend the existing code.
NOTE: The code is not 100% backward compatible, but is compatible for the
The rationale for this is because it is still a beta library, and as such is
likely to be in a constant state of flux. If/when the library becomes part
of boost, *then* I will apply stricter version controlling and backwards
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk