Re: [sympa-fr] Accumulation de « Re : » en sujet des messages

[David Prévot <adresse@cachée>]

Cyril Quémeras a écrit :
>     Bonsoir David, bonsoir à tous,

Bonsoir Cyril et les autres,

>>>  Nous avons installé Sympa (5.3.4) sur Debian Lenny avec Exim
> [...]

>>  C'est un pb corrigé dans les versions ultérieures de Sympa. (les
>>  versions 5.3 ont été arrétées en 2007)
> 
> Merci, j'ai intégré le traitement du sujet de la version 5.4.7 
[...]

>    Peux-tu m'indiquer quel patch tu as appliqué STP ?

En pièce jointe, le résultat d'un « diff -u ». J'ai juste remplacé le
traitement du sujet par celui de la version 5.4.7 (actuellement dans Sid
et Squeeze). Réponse sur la liste, vu que le fichier fait moins de 5 ko.

Amicalement

David

--- /usr/lib/sympa/bin/List.pm  2008-10-19 17:37:24.000000000 -0400
+++ /mnt/donnees/Important/globenet/home/taffit/List.pm 2009-10-21 
15:54:23.000000000 -0400
@@ -2381,38 +2381,57 @@
        
        ## Search previous subject tagging in Subject
        my $custom_subject = $self->{'admin'}{'custom_subject'};
-       my $tag_regexp = $custom_subject;
-       $tag_regexp =~ s/([\[\]\*\-\(\)\+\{\}\?])/\\$1/g;  ## cleanup, just 
in case dangerous chars were left
-       $tag_regexp =~ s/\[\S+\]/\.\+/g;
-       $tag_regexp =~ s/\s+/\\s+/g; ## consider folding
-       
-       ## Add subject tag
-       $message->{'msg'}->head->delete('Subject');
-       my $parsed_tag;
-       &tt2::parse_tt2({'list' => {'name' => $self->{'name'},
-                                   'sequence' => $self->{'stats'}->[0]
-                                   }},
-                       [$custom_subject], \$parsed_tag);
-       
-       ## If subject is tagged, replace it with new tag
-       $subject_field =~ s/\s*\[$tag_regexp\]//;
-       ## Encode subject using initial charset
-
-       ## Don't try to encode the subject if it was not originaly encoded 
non-ASCII.
-       if ($message->{'subject_charset'} or $subject_field !~ 
/[^\x00-\x7E]/) {
-           $subject_field = MIME::EncWords::encode_mimewords([
-                                                              
[Encode::decode('utf8', '['.$parsed_tag.'] '), &Language::GetCharset()],
-                                                              
[Encode::decode('utf8', $subject_field), $message->{'subject_charset'}]
-                                                              ], 
Encoding=>'A', Field=>'Subject');
-       }else {
-           $subject_field = MIME::EncWords::encode_mimewords([
-                                                              
[Encode::decode('utf8', '['.$parsed_tag.']'), &Language::GetCharset()]
-                                                              ], 
Encoding=>'A', Field=>'Subject') . ' ' . $subject_field;
-       }
+    
+       ## Remove unwanted headers if present.
+        ## tag_regexp will be used to remove the custom subject if it is 
already present in the message subject.
+        ## Remember that the value of custom_subject can be "dude number 
[%list.sequence"%]" whereas the actual
+        ## subject will contain "dude number 42".
+        my $tag_regexp = $custom_subject;
+        $tag_regexp =~ s/([\[\]\*\-\(\)\+\{\}\?])/\\$1/g;  ## cleanup, just 
in case dangerous chars were left
+        $tag_regexp =~ s/\\\[%\S+%\\\]/[^\]]\+/g; ## Replaces variables 
declarations by "[^\]]+"
+        $tag_regexp =~ s/\s+/\\s+/g; ## Takes spaces into account
+
+        ## Add subject tag
+        $message->{'msg'}->head->delete('Subject');
+        my $parsed_tag;
+        &tt2::parse_tt2({'list' => {'name' => $self->{'name'},
+                                    'sequence' => $self->{'stats'}->[0]
+                                    }},
+                        [$custom_subject], \$parsed_tag);
+
+        ## If subject is tagged, replace it with new tag
+        ## Splitting the subject in two parts :
+        ##   - what is before the custom subject (probably some "Re:")
+        ##   - what is after it : the orginal subject sent to the list.
+        ## The custom subject is not kept.
+        my $before_tag = '';
+        my $after_tag = $subject_field;
+        $after_tag =~ s/.*\[$tag_regexp\]//;
+
+        if($subject_field =~ /(.*)\s*\[$tag_regexp\](.*)/) {
+            $before_tag = $1;
+            $after_tag = $2;
+            $after_tag =~ s/^\s*(.*)\s*$/$1/; ## Remove leading and trailing 
blanks
+        }
+
+        ## Encode subject using initial charset
 
-       $message->{'msg'}->head->add('Subject', $subject_field);
+        ## Don't try to encode the subject if it was not originaly encoded 
non-ASCII.
+        if ($message->{'subject_charset'} or $subject_field !~ 
/[^\x00-\x7E]/) {
+            $subject_field = MIME::EncWords::encode_mimewords([
+                                                               
[Encode::decode('utf8', $before_tag), $message->{'subject_charset'}],
+                                                               
[Encode::decode('utf8', '['.$parsed_tag.'] '), &Language::GetCharset()],
+                                                               
[Encode::decode('utf8', $after_tag), $message->{'subject_charset'}]
+                                                               ], 
Encoding=>'A', Field=>'Subject');
+        }else {
+            $subject_field = $before_tag . ' ' .  
MIME::EncWords::encode_mimewords([
+                                                                             
       [Encode::decode('utf8', '['.$parsed_tag.']'), &Language::GetCharset()]
+                                                                             
       ], Encoding=>'A', Field=>'Subject') . ' ' . $after_tag;
+        }
+
+        $message->{'msg'}->head->add('Subject', $subject_field);
     }
-    
+
     ## Remove unwanted headers if present.
     if ($Conf{'remove_headers'}) {
         foreach my $field (@{$Conf{'remove_headers'}}) {

Attachment: signature.asc
Description: OpenPGP digital signature