import: cat_blob drops leading 'From ' lines like Inbox

[public-inbox.git] / lib / PublicInbox / Import.pm

diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 73290eed5a04d37a8328eb39bd217cf2258fd6a7..b25427ee552628237bc094225f2daf24e98dcdac 100644 (file)
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -14,6 +14,7 @@ use PublicInbox::Address;
 use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
 use PublicInbox::ContentId qw(content_digest);
 use PublicInbox::MDA;
+use POSIX qw(strftime);
 
 sub new {
        my ($class, $git, $name, $email, $ibx) = @_;
@@ -95,19 +96,13 @@ sub _check_path ($$$$) {
        $info =~ /\Amissing / ? undef : $info;
 }
 
-sub check_remove_v1 {
-       my ($r, $w, $tip, $path, $mime) = @_;
-
-       my $info = _check_path($r, $w, $tip, $path) or return ('MISSING',undef);
-       $info =~ m!\A100644 blob ([a-f0-9]{40})\t!s or die "not blob: $info";
-       my $blob = $1;
-
-       print $w "cat-blob $blob\n" or wfail;
+sub _cat_blob ($$$) {
+       my ($r, $w, $oid) = @_;
+       print $w "cat-blob $oid\n" or wfail;
        local $/ = "\n";
-       $info = <$r>;
+       my $info = <$r>;
        defined $info or die "EOF from fast-import / cat-blob: $!";
-       $info =~ /\A[a-f0-9]{40} blob (\d+)\n\z/ or
-                               die "unexpected cat-blob response: $info";
+       $info =~ /\A[a-f0-9]{40} blob (\d+)\n\z/ or return;
        my $left = $1;
        my $offset = 0;
        my $buf = '';
@@ -122,7 +117,26 @@ sub check_remove_v1 {
        $n = read($r, my $lf, 1);
        defined($n) or die "read final byte of cat-blob failed: $!";
        die "bad read on final byte: <$lf>" if $lf ne "\n";
-       my $cur = PublicInbox::MIME->new(\$buf);
+
+       # fixup some bugginess in old versions:
+       $buf =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+       \$buf;
+}
+
+sub cat_blob {
+       my ($self, $oid) = @_;
+       my ($r, $w) = $self->gfi_start;
+       _cat_blob($r, $w, $oid);
+}
+
+sub check_remove_v1 {
+       my ($r, $w, $tip, $path, $mime) = @_;
+
+       my $info = _check_path($r, $w, $tip, $path) or return ('MISSING',undef);
+       $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_s = $cur->header('Subject');
        $cur_s = '' unless defined $cur_s;
        my $cur_m = $mime->header('Subject');
@@ -192,6 +206,7 @@ sub get_mark {
        my ($r, $w) = $self->gfi_start;
        print $w "get-mark $mark\n" or wfail;
        defined(my $oid = <$r>) or die "get-mark failed, need git 2.6.0+\n";
+       chomp($oid);
        $oid;
 }
 
@@ -319,7 +334,7 @@ sub v1_mid0 ($) {
        my $mids = mids($hdr);
 
        if (!scalar(@$mids)) { # spam often has no Message-Id
-               my $mid0 = digest2mid(content_digest($mime));
+               my $mid0 = digest2mid(content_digest($mime), $hdr);
                append_mid($hdr, $mid0);
                return $mid0;
        }
@@ -379,7 +394,7 @@ sub add {
 
        # v2: we need this for Xapian
        if ($self->{want_object_info}) {
-               chomp(my $oid = $self->get_mark(":$blob"));
+               my $oid = $self->get_mark(":$blob");
                $self->{last_object} = [ $oid, $n, \$str ];
        }
        my $ref = $self->{ref};
@@ -434,18 +449,19 @@ sub atfork_child {
        }
 }
 
-sub digest2mid ($) {
-       my ($dig) = @_;
+sub digest2mid ($$) {
+       my ($dig, $hdr) = @_;
        my $b64 = $dig->clone->b64digest;
        # Make our own URLs nicer:
        # See "Base 64 Encoding with URL and Filename Safe Alphabet" in RFC4648
        $b64 =~ tr!+/=!-_!d;
 
-       # We can make this more meaningful with a date prefix or other things,
-       # but this is only needed for crap that fails to generate a Message-ID
-       # or reuses one.  In other words, it's usually spammers who hit this
-       # so they don't deserve nice Message-IDs :P
-       $b64 . '@localhost';
+       # Add a date prefix to prevent a leading '-' in case that trips
+       # up some tools (e.g. if a Message-ID were a expected as a
+       # command-line arg)
+       my $dt = msg_datestamp($hdr);
+       $dt = POSIX::strftime('%Y%m%d%H%M%S', gmtime($dt));
+       "$dt.$b64" . '@z';
 }
 
 sub clean_purge_buffer {