huge refactor of encoding handling

[public-inbox.git] / public-inbox-learn

diff --git a/public-inbox-learn b/public-inbox-learn
index d770f0f7b348fe82b23f04e62d14a405b5575d59..bb130b215fc686ac82b37ec4351460251e7ef167 100755 (executable)
--- a/public-inbox-learn
+++ b/public-inbox-learn
@@ -5,7 +5,7 @@ my $usage = "$0 (spam|ham) < /path/to/message";
 use strict;
 use warnings;
 use PublicInbox::Config;
-use Email::Simple;
+use Email::MIME;
 use Email::Address;
 use IPC::Run qw/run/;
 my $train = shift or die "usage: $usage\n";
@@ -14,22 +14,18 @@ if ($train !~ /\A(?:ham|spam)\z/) {
 }
 
 my $pi_config = PublicInbox::Config->new;
-my $simple;
-{
-       local $/;
-       $simple = Email::Simple->new(<>);
-}
+my $mime = Email::MIME->new(eval { local $/; <> });
 
 # get all recipients
 my %dests;
 foreach my $h (qw(Cc To)) {
-       foreach my $recipient (Email::Address->parse($simple->header($h))) {
+       foreach my $recipient (Email::Address->parse($mime->header($h))) {
                $dests{lc($recipient->address)} = 1;
        }
 }
 
-my $in = $simple->as_string;
-$simple->body_set("");
+my $in = $mime->as_string;
+$mime->body_set('');
 
 my $err = 0;
 my @output = qw(> /dev/null > /dev/null);
@@ -54,19 +50,22 @@ foreach my $recipient (keys %dests) {
                        }
                }
        } else { # $train eq "ham"
-               my $from = $simple->header("From");
-               my @from = Email::Address->parse($from);
-               my $name = $from[0]->name;
-               defined $name or $name = "";
-               my $email = $from[0]->address;
-               defined $email or $email = "";
+               require PublicInbox::MDA;
+               require PublicInbox::Filter;
+
+               # no checking for errors here, we assume the message has
+               # been reviewed by a human at this point:
+               PublicInbox::Filter->run($mime);
+
+               my ($name, $email, $date) =
+                               PublicInbox::MDA->author_info($mime);
                local $ENV{GIT_AUTHOR_NAME} = $name;
                local $ENV{GIT_AUTHOR_EMAIL} = $email;
-               local $ENV{GIT_AUTHOR_DATE} = $simple->header("Date");
+               local $ENV{GIT_AUTHOR_DATE} = $date;
 
                # Ham messages are trained when they're marked into
-               # a SEEN state, so this is idempotent
-               run([qw(ssoma-mda -1), $git_dir], \$in, \$out, \$err);
+               # a SEEN state, so this is idempotent:
+               run([PublicInbox::MDA->cmd, $git_dir], \$in, \$out, \$err);
                if ($err !~ /CONFLICT/) {
                        $err = 1;
                }