implement MailingList::Messages::num_paragram() and make test tools for it.

it return the number of paragraph in the message ($self).
modify MailingList::Messages::get_content_type() which removes ";" for
convenience.

4 files changed, 181 insertions, 2 deletions

diff --git a/fml/lib/MailingList/Messages.pm b/fml/lib/MailingList/Messages.pm
index 9b5ddde7..0fe60af9 100644
--- a/fml/lib/MailingList/Messages.pm
+++ b/fml/lib/MailingList/Messages.pm
@@ -766,7 +766,69 @@ sub is_empty
 sub get_content_type
 {
     my ($self) = @_;
-    $self->{ content_type };
+    my $type = $self->{ content_type };
+    $type =~ s/;//;
+    $type;
+}
+
+
+=head2 C<num_paragraph()>
+
+return the number of paragraphs in the message ($self).
+
+=cut
+
+sub num_paragraph
+{
+    my ($self) = @_;
+
+    # exit ASAP if the message is empty. 
+    return 0 if $self->is_empty();
+
+    my $pb      = $self->{ offset_begin };
+    my $pe      = $self->{ offset_end };
+    my $bodylen = $self->size;
+    my $content = $self->{ content };
+
+    my $i  = 0; # the number of paragraphs
+    my $p  = $pb;
+    my $pp = $p;
+
+    # skip "\n" in the first and end of the buffer
+    while (substr($$content, $p, 1) eq "\n") { $p++;}
+    while (substr($$content, $pe -1, 1) eq "\n") { $pe--;} 
+
+    my (@pmap) = ($pb);
+  LINE:
+    while ($p < $pe) {
+	$pp = index($$content, "\n\n", $p);
+	if ($pp < $p ||    # not found
+	    $pp >= $pe ) { # over the end of buffer boundary
+
+	    push(@pmap, $pe); # the end of the last paragraph
+	    last LINE; 
+	}
+	else {
+	    # skip trailing "\n" after "\n\n"
+	    while (substr($$content, $pp, 1) eq "\n") { $pp++;}
+
+	    push(@pmap, $pp) if $pp > 0;
+
+	    $p = $pp;
+	}
+    }
+
+    # XXX debug
+    if (0) {
+	for (my $i = 0; $i < $#pmap; $i++ ) {
+	    my $p  = $pmap[ $i ];
+	    my $pp = $pmap[ $i + 1 ];
+	    print STDERR "($p,$pp)<", substr($$content, $p, $pp - $p) , ">\n";
+	}
+	print STDERR "( @pmap )\n"; 
+    }
+
+    $#pmap;
 }
 
 
diff --git a/regress/base/get_paragraph_info.pl b/regress/base/get_paragraph_info.pl
new file mode 100755
index 00000000..a78ac5d4
--- /dev/null
+++ b/regress/base/get_paragraph_info.pl
@@ -0,0 +1,89 @@
+use strict;
+use File::Basename;
+use vars qw(%opts);
+
+use Getopt::Std;
+getopts('d', \%opts);
+
+my $dir = dirname($0);
+$| = 1;
+
+my $msg = shift @ARGV || "$dir/msg_mp";
+
+use FileHandle;
+my $fh = new FileHandle $msg;
+
+my $header;
+my $boundary;
+my $content_type;
+my $body;
+
+while (<$fh>) {
+    s/\r\n$/\n/;
+
+    if (1 .. /^$/) {
+	$header .= $_;
+	if (/boundary=\"(.*)\"/) {
+	    $boundary = $1;
+	}
+	if (/Content-Type: (\S+\/\S+\w)/) {
+	    $content_type = $1;
+	}
+    }
+    else {
+	$body .= $_; 
+    }
+}
+
+my $original_length = length($body);
+
+use MailingList::Messages;
+my $m = new MailingList::Messages { 
+    content_type   => ($content_type || 'multipart/mixed'),
+    boundary       => $boundary,
+    content        => \$body,
+    debug          => 1,
+};
+
+$m->print;
+my $mp = $m->get_first_plaintext_message ;
+print "// first plain/text message\n";
+$mp->print;
+print "//\n";
+debug( $mp );
+
+exit 0;
+
+
+sub debug
+{
+    my ($m) = @_;
+
+    unless (defined $m) {
+	use Carp;
+	croak("debug gets non-object\n");
+    }
+
+    my $p;
+    my $total=0;
+    my $i = 0;
+    for ($p = $m; defined $p ; $p = $p->{ next }) {
+	my $size = $p->size;
+	my $h    = $p->get_content_header;
+	$size   += length($h) if defined $h;
+
+	$total += $size;
+
+	$i++;
+	print STDERR "--------- $i ---------\n"; 
+	print STDERR "num_p:", $p->num_paragraph, "\n\n";
+	next;
+
+	print STDERR "type: ", $p->get_content_type, "\n";
+	print STDERR " hdr:{", $h,                   "}\n";
+	print STDERR "body:{", $p->get_content_body, "}\n";
+	print STDERR "size: ", $size, " / $total (<= $original_length)\n";
+	print STDERR "\n";
+	sleep 3;
+    }
+}
diff --git a/regress/base/mime.pl b/regress/base/mime.pl
index 995ec64c..5add052c 100755
--- a/regress/base/mime.pl
+++ b/regress/base/mime.pl
@@ -14,16 +14,18 @@ use MIME::QuotedPrint;
 &try_mime(" ");
 &try_mime("　");
 &try_mime("\n");
+&try_mime(" ", 3);
 
 exit 0;
 
 
 sub try_mime
 {
-    my ($separator) = @_;
+    my ($separator, $xbuf) = @_;
     my $buf = "\n";
 
     my $orig_str = "うじゃ". $separator ."あじゃじゃ";
+    $orig_str = "うじゃ". $separator . $xbuf . "あじゃじゃ" if $xbuf;
     my $in_str_b64 = encode_mime_string($orig_str);
     my $in_str_qp  = encode_mime_string($orig_str, { encode => 'qp' });
 
diff --git a/regress/testmails/text=plain2 b/regress/testmails/text=plain2
new file mode 100644
index 00000000..aa0d7773
--- /dev/null
+++ b/regress/testmails/text=plain2
@@ -0,0 +1,26 @@
+From owner-elena@elena.sapporo.iij.ad.jp  Fri Jan 26 00:11:07 2001
+Return-Path: <fukachan@fml.org>
+Received: from hikari.fml.org (hikari.fml.org [192.168.1.6])
+	by elena.fml.org (Postfix) with ESMTP id C995C1D8C9
+	for <mailbox@elena.fml.org>; Tue,  2 Jan 2001 17:08:05 +0900 (JST)
+Received: by hikari.fml.org (8.9.3/3.4W6+Hikari-2.1) with ESMTP id RAA17788 for <caesar@fml.org>; Tue, 2 Jan 2001 17:08:04 +0900 (JST)
+Received: by iris.fml.org (Postfix, from userid 0)
+	id 1BCBE15024; Tue,  2 Jan 2001 17:08:03 +0900 (JST)
+From: fukachan@sapporo.iij.ad.jp (Ken'ichi Fukamachi)
+To: elena@nuinui.net
+Reply-To: rudo@nuinui.net
+Subject: [elena:00300] test 01
+	appended_mesg
+Message-Id: <20010102080803.1BCBE15024@iris.fml.org>
+Date: Tue,  2 Jan 2001 17:08:03 +0900 (JST)
+Mime-Version: 1.0
+Content-Type: text/plain; charset=ISO-2022-JP
+Content-Transfer-Encoding: 7bit
+
+test message
+
+
+
+
+-- rudo's mabuachi
+