Gunnar Evermann <ge204(a)eng.cam.ac.uk> wrote:
Jerry, could you look into this?
[snip]
From: cmsmcq(a)acm.org
Subject: psgml package (Wing change breaks sgml-split-element)
[snip]
Somewhere along the way, apparently in the course of
integrating psgml 1.15 with Xemacs, a change seems to
have been made to the function sgml-insert-end-tag,
so that instead of closing the current element, it
closes the largest element it can.
That change was removed from the XEmacs PSGML package when I submitted
my monster patch to sync us up with the latest PSGML release. The
current package does not contain the code mentioned in Michael's
message, and has not for several months. Like Michael, I found that it
broke things in strange ways. At the time, I sent a message to
xemacs-beta asking if anyone objected to me removing Ben's code. There
was one response, from somebody who agreed that that piece of code
needed to go.
I was going to look up URLs for my message and the reply, but the
xemacs-beta indexer appears to be broken. Not only does searching for
"wing;change" turn up no results when restricted to the year 2000 files,
but searching for "Jerry" turns up no results when restricted to the
same files. This is clearly wrong, since I have sent several messages
to the list this year, each with my name at the end.
A more elaborate change might allow the choice between closing the
currently open element and the largest in the list of endable
elements to depend on a user option when sgml-insert-end-tag is
invoked directly, and on an option passed by the caller when
it is invoked by another function. But I'll leave that to
people with better elisp skills than I have.
This should be discussed with Lennart Staflin, the PSGML maintainer.
--
Jerry James