use PublicInbox::MID qw(mids mid2path);
use PublicInbox::Address;
use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
-use PublicInbox::ContentId qw(content_digest);
+use PublicInbox::ContentHash qw(content_digest);
use PublicInbox::MDA;
+use PublicInbox::Eml;
use POSIX qw(strftime);
sub new {
$info =~ m!\A100644 blob ([a-f0-9]{40})\t!s or die "not blob: $info";
my $oid = $1;
my $msg = _cat_blob($r, $w, $oid) or die "BUG: cat-blob $1 failed";
- my $cur = PublicInbox::MIME->new($msg);
+ my $cur = PublicInbox::Eml->new($msg);
my $cur_s = $cur->header('Subject');
$cur_s = '' unless defined $cur_s;
my $cur_m = $mime->header('Subject');
my ($self, $msg) = @_;
return unless $self->{pid};
print { $self->{out} } "progress $msg\n" or wfail;
- $self->{in}->getline eq "progress $msg\n" or die
+ readline($self->{in}) eq "progress $msg\n" or die
"progress $msg not received\n";
undef;
}
}
# returns undef on non-existent
-# ('MISMATCH', Email::MIME) on mismatch
-# (:MARK, Email::MIME) on success
+# ('MISMATCH', PublicInbox::Eml) on mismatch
+# (:MARK, PublicInbox::Eml) on success
#
# v2 callers should check with Xapian before calling this as
# it is not idempotent.
sub remove {
- my ($self, $mime, $msg) = @_; # mime = Email::MIME
+ my ($self, $mime, $msg) = @_; # mime = PublicInbox::Eml or Email::MIME
my $path_type = $self->{path_type};
my ($path, $err, $cur, $blob);
"$ts $zone";
}
-sub extract_cmt_info ($) {
- my ($mime) = @_;
+sub extract_cmt_info ($;$) {
+ my ($mime, $smsg) = @_;
my $sender = '';
my $from = $mime->header('From');
}
}
if (defined $email) {
+ # Email::Address::XS may leave quoted '<' in addresses,
+ # which git-fast-import doesn't like
+ $email =~ tr/<>//d;
+
# quiet down wide character warnings with utf8::encode
utf8::encode($email);
} else {
utf8::encode($subject);
my $at = git_timestamp(my @at = msg_datestamp($hdr));
my $ct = git_timestamp(my @ct = msg_timestamp($hdr));
+ if ($smsg) {
+ $smsg->{ds} = $at[0];
+ $smsg->{ts} = $ct[0];
+ }
($name, $email, $at, $ct, $subject);
}
my $hdr = $mime->header_obj;
my $mids = mids($hdr);
- if (!scalar(@$mids)) { # spam often has no Message-Id
+ if (!scalar(@$mids)) { # spam often has no Message-ID
my $mid0 = digest2mid(content_digest($mime), $hdr);
append_mid($hdr, $mid0);
return $mid0;
# returns undef on duplicate
# returns the :MARK of the most recent commit
sub add {
- my ($self, $mime, $check_cb) = @_; # mime = Email::MIME
+ my ($self, $mime, $check_cb, $smsg) = @_;
- my ($name, $email, $at, $ct, $subject) = extract_cmt_info($mime);
+ my ($name, $email, $at, $ct, $subject) = extract_cmt_info($mime, $smsg);
my $path_type = $self->{path_type};
my $path;
if ($path_type eq '2/38') {
print $w $raw_email, "\n" or wfail;
# v2: we need this for Xapian
- if ($self->{want_object_info}) {
- my $oid = $self->get_mark(":$blob");
- $self->{last_object} = [ $oid, $n, \$raw_email ];
+ if ($smsg) {
+ $smsg->{blob} = $self->get_mark(":$blob");
+ $smsg->{bytes} = $n;
+ $smsg->{-raw_email} = \$raw_email;
}
my $ref = $self->{ref};
my $commit = $self->{mark}++;
$? == 0 or die join(' ', @$cmd) . " failed: $?\n";
}
+my @INIT_FILES = ('HEAD' => "ref: refs/heads/master\n",
+ 'description' => <<EOD,
+Unnamed repository; edit this file 'description' to name the repository.
+EOD
+ 'config' => <<EOC);
+[core]
+ repositoryFormatVersion = 0
+ filemode = true
+ bare = true
+[repack]
+ writeBitmaps = true
+EOC
+
sub init_bare {
- my ($dir) = @_;
- my @cmd = (qw(git init --bare -q), $dir);
- run_die(\@cmd);
- # set a reasonable default:
- @cmd = (qw/git config/, "--file=$dir/config",
- 'repack.writeBitmaps', 'true');
- run_die(\@cmd);
+ my ($dir) = @_; # or self
+ $dir = $dir->{git}->{git_dir} if ref($dir);
+ require File::Path;
+ File::Path::mkpath([ map { "$dir/$_" } qw(objects/info refs/heads) ]);
+ for (my $i = 0; $i < @INIT_FILES; $i++) {
+ my $f = $dir.'/'.$INIT_FILES[$i++];
+ next if -f $f;
+ open my $fh, '>', $f or die "open $f: $!";
+ print $fh $INIT_FILES[$i] or die "print $f: $!";
+ close $fh or die "close $f: $!";
+ }
}
sub done {
push @buf, "reset $tmp\n";
} elsif (/^commit (?:.+)/) {
if (@buf) {
- $w->print(@buf) or wfail;
+ print $w @buf or wfail;
@buf = ();
}
push @buf, "commit $tmp\n";
rewrite_commit($self, \@oids, \@buf, $mime);
$nreplace++;
}
- $w->print(@buf, "\n") or wfail;
+ print $w @buf, "\n" or wfail;
@buf = ();
} elsif ($_ eq "done\n") {
$done = 1;
}
close $rd or die "close fast-export failed: $?";
if (@buf) {
- $w->print(@buf) or wfail;
+ print $w @buf or wfail;
}
die 'done\n not seen from fast-export' unless $done;
chomp(my $cmt = $self->get_mark(":$mark")) if $nreplace;
=head1 SYNOPSIS
- use Email::MIME;
+ use PublicInbox::Eml;
+ # PublicInbox::Eml exists as of public-inbox 1.5.0,
+ # Email::MIME was used in older versions
+
use PublicInbox::Git;
use PublicInbox::Import;
"Date: Thu, 01 Jan 1970 00:00:00 +0000\n" .
"Message-ID: <m\@example.org>\n".
"\ntest message";
- my $parsed = Email::MIME->new($message);
+ my $parsed = PublicInbox::Eml->new($message);
my $ret = $im->add($parsed);
if (!defined $ret) {
warn "duplicate: ",
$im->done;
# to remove a message
- my $junk = Email::MIME->new($message);
+ my $junk = PublicInbox::Eml->new($message);
my ($mark, $orig) = $im->remove($junk);
if ($mark eq 'MISSING') {
print "not found\n";
=head1 DESCRIPTION
-An importer and remover for public-inboxes which takes L<Email::MIME>
-messages as input and stores them in a git repository as
+An importer and remover for public-inboxes which takes C<PublicInbox::Eml>
+or L<Email::MIME> messages as input and stores them in a git repository as
documented in L<https://public-inbox.org/public-inbox-v1-format.txt>,
except it does not allow duplicate Message-IDs.
=head2 add
- my $parsed = Email::MIME->new($message);
+ my $parsed = PublicInbox::Eml->new($message);
$im->add($parsed);
Adds a message to to the git repository. This will acquire
=head2 remove
- my $junk = Email::MIME->new($message);
+ my $junk = PublicInbox::Eml->new($message);
my ($code, $orig) = $im->remove($junk);
Removes a message from the repository. On success, it returns
a ':'-prefixed numeric code representing the git-fast-import
-mark and the original messages as an Email::MIME object.
+mark and the original messages as a PublicInbox::Eml
+(or Email::MIME) object.
If the message could not be found, the code is "MISSING"
and the original message is undef. If there is a mismatch where
the "Message-ID" is matched but the subject and body do not match,
=head1 COPYRIGHT
-Copyright (C) 2016 all contributors L<mailto:meta@public-inbox.org>
+Copyright (C) 2016-2020 all contributors L<mailto:meta@public-inbox.org>
License: AGPL-3.0+ L<http://www.gnu.org/licenses/agpl-3.0.txt>