Re: I have diagnosed and fixed a problem with Wanderlust under Emacs 23

[Tetsurou Okazaki <okazaki@be.to>]

At Tue, 18 Nov 2008 16:41:47 -0500,
Lloyd Zusman wrote:
> 
> There has been a problem with Wanderlust and Emacs 23 ever since that
> version of Emacs came out. This is a description of the problem:
> 
> http://article.gmane.org/gmane.mail.wanderlust.general.japanese/6088
> 
> After much digging, I have found the cause of the problem and I have
> a solution. It turns out that in the `wl-summary-detect-mark-position'
> function in wl-summary.el, a dummy message is written into a temp buffer
> with special characters (ascii 200 and 201) being used as the temp-mark
> and the wl-summary-new-uncached-mark. Then, after this dummy message
> is created, these characters are searched for within the temp buffer
> using `search-forward' to determine the positions in the summary
> line where these marks are supposed to appear.
> 
> Well, in Emacs 23, the underlying coding system is UTF-8, and therefore,
> `search-forward' doesn't find these characters in the temp buffer given
> the way that they were created. This causes the character position of
> the mark to be returned as `nil' by `wl-summary-detect-mark-position',
> which in turn causes the code that formats the Summary buffer to fail.
> 
> The way I fixed this is to force the temp buffer to enable the use of
> multibyte characters within searches. This causes `search-forward' to
> succeed, and then the rest of the code for properly formatting the
> Summary buffer also succeeds.
> 
> I'm enclosing a patch. Could someone who understands Wanderlust and
> Emacs better than I do review this patch, and correct it, if necessary?
> And then, could someone apply the resulting patch to the Wanderlust
> code base, so we can then properly use it under Emacs 23?

Applied, thanks.

-- 
Tetsurou Okazaki