svn commit: r1932428 - in spamassassin/trunk: . lib/Mail lib/Mail/SpamAssassin

Author: fkento
Date: Sat Mar 21 04:35:43 2026
New Revision: 1932428

Log:
Add multipart_alternative_preferred_part config option

New setting allows choosing which part of a multipart/alternative
section to use for rendered body text (text/html or text/plain),
skipping the other alternative. Defaults to empty (all parts included,
preserving existing behavior).

Modified:
   spamassassin/trunk/UPGRADE
   spamassassin/trunk/lib/Mail/SpamAssassin.pm
   spamassassin/trunk/lib/Mail/SpamAssassin/Conf.pm
   spamassassin/trunk/lib/Mail/SpamAssassin/Message.pm

Modified: spamassassin/trunk/UPGRADE
==============================================================================
--- spamassassin/trunk/UPGRADE	Sat Mar 21 04:31:44 2026	(r1932427)
+++ spamassassin/trunk/UPGRADE	Sat Mar 21 04:35:43 2026	(r1932428)
@@ -44,6 +44,12 @@ Note for Users Upgrading to SpamAssassin
   If no AuthRes results are available or the policy is missing, the plugin
   falls back to its own Mail::DMARC::PurePerl-based validation as before.
 
+- New configuration option: multipart_alternative_preferred_part
+  When set to "text/html" or "text/plain", only the preferred part type
+  from multipart/alternative sections will be used for rendered body text,
+  skipping the other alternative. By default this is unset and all parts
+  are included as before.
+
 - New Mail::SpamAssassin::Plugin::NeuralNetwork
   This plugin checks messages using Fast Artificial Neural Network library.
 

Modified: spamassassin/trunk/lib/Mail/SpamAssassin.pm
==============================================================================
--- spamassassin/trunk/lib/Mail/SpamAssassin.pm	Sat Mar 21 04:31:44 2026	(r1932427)
+++ spamassassin/trunk/lib/Mail/SpamAssassin.pm	Sat Mar 21 04:35:43 2026	(r1932428)
@@ -621,6 +621,7 @@ sub parse {
     normalize=>$self->{conf}->{normalize_charset},
     body_part_scan_size=>$self->{conf}->{body_part_scan_size},
     rawbody_part_scan_size=>$self->{conf}->{rawbody_part_scan_size},
+    multipart_alternative_preferred_part=>$self->{conf}->{multipart_alternative_preferred_part},
     master_deadline=>$master_deadline, suppl_attrib=>$suppl_attrib });
 
   # bug 5069: The goal here is to get rendering plugins to do things

Modified: spamassassin/trunk/lib/Mail/SpamAssassin/Conf.pm
==============================================================================
--- spamassassin/trunk/lib/Mail/SpamAssassin/Conf.pm	Sat Mar 21 04:31:44 2026	(r1932427)
+++ spamassassin/trunk/lib/Mail/SpamAssassin/Conf.pm	Sat Mar 21 04:35:43 2026	(r1932428)
@@ -3809,7 +3809,34 @@ Like body_part_scan_size, for "rawbody"
     default => 500000,
     type => $CONF_TYPE_NUMERIC,
   });
-  
+
+=item multipart_alternative_preferred_part (default: empty)
+
+Specifies which part of a multipart/alternative section to prefer for
+rendered body text.  Valid values are C<text/html> and C<text/plain>.
+When set, only the preferred part type will be used from
+multipart/alternative sections, skipping the other alternative.  When
+empty (the default), all parts are included as before.
+
+=cut
+
+  push (@cmds, {
+    setting => 'multipart_alternative_preferred_part',
+    default => '',
+    type => $CONF_TYPE_STRING,
+    code => sub {
+      my ($self, $key, $value, $line) = @_;
+      unless (defined $value && $value !~ /^$/) {
+        return $MISSING_REQUIRED_VALUE;
+      }
+      if ($value eq 'text/html' || $value eq 'text/plain') {
+        $self->{multipart_alternative_preferred_part} = $value;
+      } else {
+        return $INVALID_VALUE;
+      }
+    }
+  });
+
 =item rbl_timeout t [t_min] [zone]		(default: 15 3)
 
 All DNS queries are made at the beginning of a check and we try to read

Modified: spamassassin/trunk/lib/Mail/SpamAssassin/Message.pm
==============================================================================
--- spamassassin/trunk/lib/Mail/SpamAssassin/Message.pm	Sat Mar 21 04:31:44 2026	(r1932427)
+++ spamassassin/trunk/lib/Mail/SpamAssassin/Message.pm	Sat Mar 21 04:35:43 2026	(r1932428)
@@ -120,6 +120,7 @@ sub new {
   $self->{suppl_attrib} = $opts->{'suppl_attrib'};
   $self->{body_part_scan_size} = $opts->{'body_part_scan_size'} || 0;
   $self->{rawbody_part_scan_size} = $opts->{'rawbody_part_scan_size'} || 0;
+  $self->{multipart_alternative_preferred_part} = $opts->{'multipart_alternative_preferred_part'} || '';
 
   if ($self->{suppl_attrib}) {  # caller-provided additional information
     # pristine_body_length is currently used by an eval test check_body_length
@@ -735,6 +736,7 @@ sub finish {
   delete $self->{'normalize'};
   delete $self->{'body_part_scan_size'};
   delete $self->{'rawbody_part_scan_size'};
+  delete $self->{'multipart_alternative_preferred_part'};
   delete $self->{'pristine_msg'};
   delete $self->{'pristine_body'};
   delete $self->{'pristine_headers'};
@@ -1254,6 +1256,18 @@ sub get_mimepart_digests {
 
 # ---------------------------------------------------------------------------
 
+# Search children/descendants of a node for a leaf part matching $type.
+# Returns the part if found, undef otherwise.
+sub _find_part_by_type {
+  my ($node, $type) = @_;
+  my @q = @{$node->{'body_parts'} || []};
+  while (my $n = shift @q) {
+    return $n if $n->is_leaf() && $n->{'type'} eq $type;
+    push @q, @{$n->{'body_parts'}} if !$n->is_leaf();
+  }
+  return undef;
+}
+
 # common code for get_rendered_body_text_array,
 # get_visible_rendered_body_text_array, get_invisible_rendered_body_text_array
 #
@@ -1266,10 +1280,7 @@ sub get_body_text_array_common {
   $self->{$key} = [];
 
   my $scansize = $self->{body_part_scan_size};
-
-  # Find all parts which are leaves
-  my @parts = $self->find_parts(qr/./,1);
-  return $self->{$key} unless @parts;
+  my $preferred_alt = $self->{multipart_alternative_preferred_part};
 
   # the html metadata may have already been set, so let's not bother if it's
   # already been done.
@@ -1278,13 +1289,25 @@ sub get_body_text_array_common {
   my $subject = $method_name eq 'invisible_rendered' ? ''
                : ($self->get_header('subject') || "\n");
 
-  # Go through each part
+  # Walk the MIME tree using a queue.  For multipart/alternative nodes,
+  # if a preferred part type is set and found, enqueue only that part
+  # instead of all children (skipping the other alternative).
+  my @queue = ($self);
   my $text = '';
-  for (my $pt = 0 ; $pt <= $#parts ; $pt++ ) {
-    my $p = $parts[$pt];
-
-    # put a blank line between parts ...
-    $text .= "\n"  if $text ne '';
+  while (my $p = shift @queue) {
+    if (!$p->is_leaf()) {
+      if ($preferred_alt && $p->{'type'} eq 'multipart/alternative') {
+        my $preferred_part = _find_part_by_type($p, $preferred_alt);
+        if ($preferred_part) {
+          unshift @queue, $preferred_part;
+        } else {
+          unshift @queue, @{$p->{'body_parts'}};
+        }
+      } else {
+        unshift @queue, @{$p->{'body_parts'}};
+      }
+      next;
+    }
 
     my($type, $rnd) = $p->$method_name();  # decode this part
     # Only text/* types are rendered ...
@@ -1315,6 +1338,9 @@ sub get_body_text_array_common {
         }
       }
 
+      # put a blank line between parts ...
+      $text .= "\n"  if $text ne '';
+
       # Add to rendered text
       $text .= $rnd;