On Jan 28, 2007, at 4:48 PM, James Howison wrote:
So things like comma vs semi-colon would be handled in the style file,
ie if you had a style that was right in all other respects but wanted
to swap comma vs semi-colon you'd be sub-classing the style for this
specific document?
Yes, though CSL has no notion of inter-style subclassing. That would
have been too complicated.
I think that's ok, a GUI can hide that complexity anyway, but does it
imply embedding the tweaked style file in the document (Matthias's
'project specific settings' are implemented as 'document specific
settings')?
Forgive my ignorance, but does CSL currently have the ability to
specify in-text styles? (I'm guessing not, since we're talking about
how to describe the lowest level flags)
It has hard-coded style options: the generic, and an optional
"subsequent" to handle typically note-based styles.
It *would* be possible to allow user-defined styles, but I'd have to
think more if that's a good idea or not (KISS and such).
Do you imagine the style file specifying 'bundles' of flags or a set
of named formatters?
I suppose in the subsequent form of a citation is sort of like a named
formatter (?). I just don't allow any old template.
The bibtex approach is named formatters. If named formatters are used
then it's hard to see how one could change just one of them, but if
'bundles' of flags are used, then maybe it's pretty easy to allow just
one to vary?
When I say "user-defined styles" above, I'm basically assuming someone
does:
<item style="foo">
...
</item>
... in the style file, and then the user can select a "foo" style
option locally.
But that then couples the formatting of a document to CSL.
Bruce
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
