X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FImport.pm;h=c7a96e1eabcc1214e6cdc61572ba6ed70db64750;hb=69329215485cf2ab9d8cd1fa7faf65d8ec42dc0b;hp=463b44e2e7998e8f4261b696a4f62cb1528cc3b1;hpb=aacd0f52d9f4ba6821ad34b30a123966958c7e0c;p=public-inbox.git diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm index 463b44e2..c7a96e1e 100644 --- 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) = @_; @@ -49,7 +50,14 @@ sub gfi_start { $self->lock_acquire; local $/ = "\n"; - chomp($self->{tip} = $git->qx(qw(rev-parse --revs-only), $self->{ref})); + my $ref = $self->{ref}; + chomp($self->{tip} = $git->qx(qw(rev-parse --revs-only), $ref)); + if ($self->{path_type} ne '2/38' && $self->{tip}) { + local $/ = "\0"; + my @tree = $git->qx(qw(ls-tree -r -z --name-only), $ref); + chomp @tree; + $self->{-tree} = { map { $_ => 1 } @tree }; + } my $git_dir = $git->{git_dir}; my @cmd = ('git', "--git-dir=$git_dir", qw(fast-import @@ -88,19 +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 = ''; @@ -115,7 +117,23 @@ 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 $cur = PublicInbox::MIME->new(\$buf); + \$buf; +} + +sub cat_blob { + my ($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 = $mime->header('Subject'); @@ -185,6 +203,7 @@ sub get_mark { my ($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; } @@ -238,7 +257,8 @@ sub remove { if (defined $path) { print $w "D $path\n\n" or wfail; } else { - print $w "M 100644 :$blob _/D\n\n" or wfail; + clean_tree_v2($self, $w, 'd'); + print $w "M 100644 :$blob d\n\n" or wfail; } $self->{nchg}++; (($self->{tip} = ":$commit"), $cur); @@ -311,12 +331,21 @@ sub v1_mid0 ($) { 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; } $mids->[0]; } +sub clean_tree_v2 ($$$) { + my ($self, $w, $keep) = @_; + my $tree = $self->{-tree} or return; #v2 only + delete $tree->{$keep}; + foreach (keys %$tree) { + print $w "D $_\n" or wfail; + } + %$tree = ($keep => 1); +} # returns undef on duplicate # returns the :MARK of the most recent commit @@ -362,7 +391,7 @@ sub add { # v2: we need this for Xapian if ($self->{want_object_info}) { - chomp(my $oid = $self->get_mark(":$blob")); + my $oid = $self->get_mark(":$blob"); $self->{last_object} = [ $oid, $n, \$str ]; } my $ref = $self->{ref}; @@ -382,6 +411,7 @@ sub add { if ($tip ne '') { print $w 'from ', ($parent ? $parent : $tip), "\n" or wfail; } + clean_tree_v2($self, $w, $path); print $w "M 100644 :$blob $path\n\n" or wfail; $self->{nchg}++; $self->{tip} = ":$commit"; @@ -416,23 +446,25 @@ sub atfork_child { } } -sub digest2mid ($) { - my ($dig) = @_; +sub digest2mid ($$) { + 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 { - my ($oid, $buf) = @_; - my $cmt_msg = "purged $oid\n"; + my ($oids, $buf) = @_; + my $cmt_msg = 'purged '.join(' ',@$oids)."\n"; + @$oids = (); foreach my $i (0..$#$buf) { my $l = $buf->[$i]; @@ -456,6 +488,9 @@ sub purge_oids { my ($r, $w) = $self->gfi_start; my @buf; my $npurge = 0; + my @oids; + my ($done, $mark); + my $tree = $self->{-tree}; while (<$rd>) { if (/^reset (?:.+)/) { push @buf, "reset $tmp\n"; @@ -472,25 +507,33 @@ sub purge_oids { my $n = read($rd, my $buf, $len) or die "read: $!"; $len == $n or die "short read ($n < $len)"; push @buf, $buf; - } elsif (/^M 100644 ([a-f0-9]+) /) { - my $oid = $1; + } elsif (/^M 100644 ([a-f0-9]+) (\w+)/) { + my ($oid, $path) = ($1, $2); if ($purge->{$oid}) { - my $lf = <$rd>; - if ($lf eq "\n") { - my $out = join('', @buf); - $out =~ s/^/# /sgm; - warn "purge rewriting\n", $out, "\n"; - clean_purge_buffer($oid, \@buf); - $out = join('', @buf); - $w->print(@buf, "\n") or wfail; - @buf = (); - $npurge++; - } else { - die "expected LF: $lf\n"; - } + push @oids, $oid; + delete $tree->{$path}; } else { + $tree->{$path} = 1; push @buf, $_; } + } elsif (/^D (\w+)/) { + my $path = $1; + push @buf, $_ if $tree->{$path}; + } elsif ($_ eq "\n") { + if (@oids) { + my $out = join('', @buf); + $out =~ s/^/# /sgm; + warn "purge rewriting\n", $out, "\n"; + clean_purge_buffer(\@oids, \@buf); + $npurge++; + } + $w->print(@buf, "\n") or wfail; + @buf = (); + } elsif ($_ eq "done\n") { + $done = 1; + } elsif (/^mark :(\d+)$/) { + push @buf, $_; + $mark = $1; } else { push @buf, $_; } @@ -498,7 +541,9 @@ sub purge_oids { if (@buf) { $w->print(@buf) or wfail; } - $w = $r = undef; + die 'done\n not seen from fast-export' unless $done; + chomp(my $cmt = $self->get_mark(":$mark")) if $npurge; + $self->{nchg} = 0; # prevent _update_git_info until update-ref: $self->done; my @git = ('git', "--git-dir=$git->{git_dir}"); @@ -517,7 +562,9 @@ sub purge_oids { $err++; } } + _update_git_info($self, 0); die "Failed to purge $err object(s)\n" if $err; + $cmt; } 1;