CEDET

>>> David Ponce <david.ponce@...> seems to think that:
>Hi Eric,
>
>I committed the following changes mostly to remove obsolete code from
>Semantic.

Great, thanks.

>On an another subject, did you get a chance to check if wisent-dot
>works with my latest changes to wisent-dot files (posted in
>"Auto-generation of lexical analyzers for cogre .dot files")?  If so,
>can I commit them?

They do not.  (The blocks don't seem to be grabbed and their contents
parsed.  If you check in the changes, I would like to investigate to
experiment more with the new goodies.


>P.S.: It seems that ML are no more archived at SF since 2004-01-15 :-(

I have found the mailing list archives to be unreliable on any given
day.  It will probably re-appear again on some random future date.

Thanks
Eric

-- 
          Eric Ludlam:                 zappo@..., eric@...
   Home: http://www.ludlam.net            Siege: http://www.siege-engine.com
Emacs: http://cedet.sourceforge.net               GNU: http://www.gnu.org

>>> David Ponce <david.ponce@...> seems to think that:
>Eric,
>
>[...]
> >>On an another subject, did you get a chance to check if wisent-dot
> >>works with my latest changes to wisent-dot files (posted in
> >>"Auto-generation of lexical analyzers for cogre .dot files")?  If so,
> >>can I commit them?
> >
> >
> > They do not.  (The blocks don't seem to be grabbed and their contents
> > parsed.  If you check in the changes, I would like to investigate to
> > experiment more with the new goodies.
>[...]
>
>I finally got the support file for .dot files ;-)
>
>FYI, it seems that the link in wisent-dot.el is no more valid.  After
>googling a little, I got it from this address:
>
><http://www.research.att.com/sw/tools/graphviz/extras/graphviz-dot-mode.el>
>
>After applying your suggested patch to the dot syntax table, I did a
>quick test on cogre/test.dot with my modified wisent-dot.el,
>wisent-dot.wy, and re-generated wisent-dot-wy.el, and obtained the
>following parse tree:
>
>(("structs" digraph
  [ ... ]

That's good.  I'll have to update the help.  Now that it is at att,
perhaps someone will accept my patch. ;)

It could be I do not have enough of your new stuff.  CVS update this
morning was so slow I had to cancel.  I hope I it isn't related to my
new computer, ip address, and versions of whatever.  Perhaps when I
successfully update I too will enjoy happy parsing.

Last time I updated, I found I had to do a "make clean" before a
rebuild, and I had to run make twice for your changes to bootstrap in
properly.  I'm still a couple features away from a Beta 2 release so
hopefully a dist rebuild is ok.

>It seems to me that the dot parser works well.  However it looks like
>an overrode `semantic-tag-components' is missing to correctly define
>overlays in child tags.

I will investigate when I next synchronize.

>Anyway, I checked-in the new wisent-dot.el and wisent-dot.wy files so
>you can play with them ;-)
  [ ... ]

Thanks!
Eric

-- 
          Eric Ludlam:                 zappo@..., eric@...
   Home: http://www.ludlam.net            Siege: http://www.siege-engine.com
Emacs: http://cedet.sourceforge.net               GNU: http://www.gnu.org

Eric,

[...]
 >>On an another subject, did you get a chance to check if wisent-dot
 >>works with my latest changes to wisent-dot files (posted in
 >>"Auto-generation of lexical analyzers for cogre .dot files")?  If so,
 >>can I commit them?
 >
 >
 > They do not.  (The blocks don't seem to be grabbed and their contents
 > parsed.  If you check in the changes, I would like to investigate to
 > experiment more with the new goodies.
[...]

I finally got the support file for .dot files ;-)

FYI, it seems that the link in wisent-dot.el is no more valid.  After
googling a little, I got it from this address:

<http://www.research.att.com/sw/tools/graphviz/extras/graphviz-dot-mode.el>

After applying your suggested patch to the dot syntax table, I did a
quick test on cogre/test.dot with my modified wisent-dot.el,
wisent-dot.wy, and re-generated wisent-dot-wy.el, and obtained the
following parse tree:

(("structs" digraph
   (:members
    (("NODE" generic-node
      (:attributes
       (("shape" attribute
         (:value "record")
         (reparse-symbol node-description)
         [26 38])))
      (reparse-symbol graph-contents)
      [20 40])
     ("class1" node
      (:attributes
       (("label" attribute
         (:value "\"{class1|||}\"")
         (reparse-symbol node-description)
         [50 69])))
      (reparse-symbol graph-contents)
      [42 71])
     ("class2" node
      (:attributes
       (("shape" attribute
         (:value "record")
         (reparse-symbol node-description)
         [81 93])
        ("label" attribute
         (:value "\"{class2|||}\"")
         (reparse-symbol node-description)
         [94 113])))
      (reparse-symbol graph-contents)
      [73 115])
     ("class3" node
      (:attributes
       (("shape" attribute
         (:value "record")
         (reparse-symbol node-description)
         [125 137])
        ("label" attribute
         (:value "\"{class3|||}\"")
         (reparse-symbol node-description)
         [138 157])))
      (reparse-symbol graph-contents)
      [117 159])
     ("class1" link
      (:to "class2")
      (reparse-symbol graph-contents)
      [163 180])
     ("class1" link
      (:to "class3")
      (reparse-symbol graph-contents)
      [182 199])))
   nil #<overlay from 1 to 201 in test.dot>))

It seems to me that the dot parser works well.  However it looks like
an overrode `semantic-tag-components' is missing to correctly define
overlays in child tags.

Anyway, I checked-in the new wisent-dot.el and wisent-dot.wy files so
you can play with them ;-)

Good luck!

David