Mats Lidell writes:
>>>>> Stephen J Turnbull <stephen(a)xemacs.org>
> You just chose a bad place to start.
Maybe there are no good places to start ;-)
Not at all. Jeff Sparkes' work on GTK+ is quite isolated from the
rest of XEmacs, for example. Although I disagree with his proposal to
add delete-trailing-whitespace, if whitespace.el didn't exist that
would be very plausible, and not dangerous at all. But the syntax
stuff and anything related to parse-partial-sexp is hairy. That code
is not based on a real parser; it requires ugly heuristics like
"defuns start with an open-paren-syntax character in column 0".
I guess the real problem is that I don't know what I should test
what the right response is! So if you have some good tests for
parse-partial-sexp laying around...
Not at the moment. I guess what I would do is a sledgehammer test
taking something with horridly complex code (efs.el in Lisp, and
lisp.h for C would probably do), and start from character zero and
test every pair of positions. :-)
Without that I think I will make
up some simple tests and go for a patch for getting views.
That's the right way to go.
XEmacs-Beta mailing list