On 2007-02-19 19:59 +0100, Gary Hennigan wrote:
> Package: emacs21-el
> Version: 21.4a+1-3
> Severity: normal
> Tags: patch
>
>
> If you look at the ldap-search-internal function in the file ldap.el
> there is a regex that parses the lines output from the invocation of
> "ldap-ldapsearch-prog". It parses the string into three substrings and
> then sets the variable "name" to the first substring and the variable
> "value" to the third substring. In the latest ldap.el it then tries to
> invoke "string-match" on "value". Unfortunately that match always
> fails because it looks like it's assuming that part of the regex match
> for the second substring, which isn't stored, is contained in
> "value". There's either a problem with the regex used to parse the
> string into substrings, or you need to get rid of the "string-match",
> which is what was done in the previous version of ldap.el in
> emacs21-el 21.4a+1-1. Here's the diff for the latter that fixed the
> problem for me:
>
> --- /usr/share/emacs/21.4/lisp/net/ldap.el 2007-01-04 00:53:33.000000000
> -0700
> +++ /usr/share/emacs/21.4/lisp/net/ldap.el 2007-02-19 11:31:26.000000000
> -0700
> @@ -587,13 +587,12 @@
> (while (looking-at "^\\(\\w*\\)[=:\t ]+\\(<[\t
> ]*file://\\)?\\(.*\\)$")
> (setq name (match-string 1)
> value (match-string 3))
> - (if (string-match "^< file:///\\(.*\\)$" value)
> (save-excursion
> (set-buffer bufval)
> (erase-buffer)
> (insert-file-contents-literally value)
> (delete-file value)
> - (setq value (buffer-substring (point-min) (point-max)))))
> + (setq value (buffer-substring (point-min) (point-max))))
> (setq record (cons (list name value)
> record))
> (forward-line 1))
Could you please test whether this bug is still present in the emacs22
(not emacs21) package in Lenny? While I think it is fixed, the
ldap-search-internal function looks significantly different there, and I
have no opportunity to test it.
TIA,
Sven
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
