[RFC] dumping the CED in favor of local variables

[Paul W. Frields]

On Wed, 2005-04-27 at 17:09 -0700, Karsten Wade wrote:
> On Wed, 2005-04-27 at 16:50 -0700, Karsten Wade wrote:
> > I'd like to recommend we dump the usage of CED files in favor of the
> > local variables designation.  The CED file is a pre-compiled DTD subset.
> 
> For the record, this is a *complete reversal* of my previous position.
> If Dave Pawson is lurking out there, he can laugh at me.  The rest of
> you who care can swift-kick when you see/meet me next.
> 
> After working with the local variable, I am willing to undergo the pain
> to change over to it.  If there is even any pain.
> 
> The only problem is _mixing_ the two formats.  If I write using the
> Local Variable and you load the CED file, our indenting will be off.  At
> the least, diffs will be noisy.
> 
> > <!-- Keep this comment at the end of the file
> > Local variables:
> > mode: xml
> > xml-parent-document:("parent-file.xml" "book" "chapter")
> > End:
> 
> D'oh!
> 
> that should be:
> 
> sgml-parent-document:("parent-file.xml" "book" "chapter")
> 
> I did s/sg/x/g without verifying. :)

Karsten, I'm having a big problem with this on my system.  I am looking
at the style chapter you committed to CVS last night.  Although the
indenting works *somewhat*, using the Emacs "C-c <" command to get a tag
list is completely screwed up now.  I am not using a .ced file at all.

Here's a blow-by-blow example:

1. From shell, "emacs docs-style-en.xml"
2. Move to line 13, right below the <title> element for the <chapter>
3. Use Emacs command "C-c <" to get a tag list
4. The sole completion is </chapter>!

Obviously, I should be able to open a new <para> here without a problem.
When using DTD loading/parsing with .ced files, it worked flawlessly.
Now, at first I thought this might be some weird artifact of my using
the emacs RPMs from Jens Petersen's emacs-cvs-testing repository, but
not so.  I reverted to the standard emacs-21.3-21.FC3 that's in Core +
updates, and the same problem occurs.

Does no one else have this problem?  If you're going to test this,
please remember the system documenting guidelines... open up a new
account, and follow the instructions in the current Doc Guide release to
test.  If you're an Emacs expert, you probably have all sorts of cool
stuff going on which to me, to quote Arthur C. Clarke only slightly out
of context, is indistinguishable from magic.

If someone can show me what I'm doing wrong, then great, I'm all for
improvements.  If not, then let's leave well enough alone.  The number
of people adding tags to parent files in guides is relatively small,
compared to the number of people who will -- or should, by the guidance
we've set up -- be editing single-file tutorials where this isn't a
concern.

-- 
Paul W. Frields, RHCE                          http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
 Fedora Documentation Project: http://fedora.redhat.com/projects/docs/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.fedoraproject.org/pipermail/docs/attachments/20050428/1149e9e3/attachment.bin