-# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
#
# git fast-import-based ssoma-mda MDA replacement
use strict;
use parent qw(PublicInbox::Lock);
use v5.10.1;
-use PublicInbox::Spawn qw(spawn popen_rd);
+use PublicInbox::Spawn qw(run_die popen_rd);
use PublicInbox::MID qw(mids mid2path);
use PublicInbox::Address;
-use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
+use PublicInbox::Smsg;
+use PublicInbox::MsgTime qw(msg_datestamp);
use PublicInbox::ContentHash qw(content_digest);
use PublicInbox::MDA;
use PublicInbox::Eml;
use POSIX qw(strftime);
+sub default_branch () {
+ state $default_branch = do {
+ my $r = popen_rd([qw(git config --global init.defaultBranch)],
+ { GIT_CONFIG => undef });
+ chomp(my $h = <$r> // '');
+ close $r;
+ $h eq '' ? 'refs/heads/master' : "refs/heads/$h";
+ }
+}
+
sub new {
# we can't change arg order, this is documented in POD
# and external projects may rely on it:
my ($class, $git, $name, $email, $ibx) = @_;
- my $ref = 'refs/heads/master';
+ my $ref;
if ($ibx) {
- $ref = $ibx->{ref_head} // 'refs/heads/master';
+ $ref = $ibx->{ref_head};
$name //= $ibx->{name};
$email //= $ibx->{-primary_address};
$git //= $ibx->git;
git => $git,
ident => "$name <$email>",
mark => 1,
- ref => $ref,
- -inbox => $ibx,
+ ref => $ref // default_branch,
+ ibx => $ibx,
path_type => '2/38', # or 'v2'
lock_path => "$git->{git_dir}/ssoma.lock", # v2 changes this
bytes_added => 0,
sub gfi_start {
my ($self) = @_;
- return ($self->{in}, $self->{out}) if $self->{pid};
+ return ($self->{in}, $self->{out}) if $self->{in};
- my ($out_r, $out_w);
+ my ($in_r, $out_r, $out_w);
pipe($out_r, $out_w) or die "pipe failed: $!";
- my $git = $self->{git};
$self->lock_acquire;
-
- local $/ = "\n";
- 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
- --quiet --done --date-format=raw));
- my ($in_r, $pid) = popen_rd(\@cmd, undef, { 0 => $out_r });
- $out_w->autoflush(1);
- $self->{in} = $in_r;
- $self->{out} = $out_w;
- $self->{pid} = $pid;
- $self->{nchg} = 0;
- binmode $out_w, ':raw' or die "binmode :raw failed: $!";
- binmode $in_r, ':raw' or die "binmode :raw failed: $!";
+ eval {
+ my ($git, $ref) = @$self{qw(git ref)};
+ local $/ = "\n";
+ chomp($self->{tip} = $git->qx(qw(rev-parse --revs-only), $ref));
+ die "fatal: rev-parse --revs-only $ref: \$?=$?" if $?;
+ if ($self->{path_type} ne '2/38' && $self->{tip}) {
+ my $t = $git->qx(qw(ls-tree -r -z --name-only), $ref);
+ die "fatal: ls-tree -r -z --name-only $ref: \$?=$?" if $?;
+ $self->{-tree} = { map { $_ => 1 } split(/\0/, $t) };
+ }
+ $in_r = $self->{in} = $git->popen(qw(fast-import
+ --quiet --done --date-format=raw),
+ undef, { 0 => $out_r });
+ $out_w->autoflush(1);
+ $self->{out} = $out_w;
+ $self->{nchg} = 0;
+ };
+ if ($@) {
+ $self->lock_release;
+ die $@;
+ }
($in_r, $out_w);
}
return if $tip eq '';
print $w "ls $tip $path\n" or wfail;
local $/ = "\n";
- defined(my $info = <$r>) or die "EOF from fast-import: $!";
+ my $info = <$r> // die "EOF from fast-import: $!";
$info =~ /\Amissing / ? undef : $info;
}
my ($r, $w, $oid) = @_;
print $w "cat-blob $oid\n" or wfail;
local $/ = "\n";
- my $info = <$r>;
- defined $info or die "EOF from fast-import / cat-blob: $!";
- $info =~ /\A[a-f0-9]{40} blob ([0-9]+)\n\z/ or return;
+ my $info = <$r> // die "EOF from fast-import / cat-blob: $!";
+ $info =~ /\A[a-f0-9]{40,} blob ([0-9]+)\n\z/ or return;
my $left = $1;
my $offset = 0;
my $buf = '';
my $n;
while ($left > 0) {
- $n = read($r, $buf, $left, $offset);
- defined($n) or die "read cat-blob failed: $!";
+ $n = read($r, $buf, $left, $offset) //
+ die "read cat-blob failed: $!";
$n == 0 and die 'fast-export (cat-blob) died';
$left -= $n;
$offset += $n;
}
- $n = read($r, my $lf, 1);
- defined($n) or die "read final byte of cat-blob failed: $!";
+ $n = read($r, my $lf, 1) //
+ die "read final byte of cat-blob failed: $!";
die "bad read on final byte: <$lf>" if $lf ne "\n";
# fixup some bugginess in old versions:
my ($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";
+ $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 $cur = PublicInbox::Eml->new($msg);
- my $cur_s = $cur->header('Subject');
- $cur_s = '' unless defined $cur_s;
- my $cur_m = $mime->header('Subject');
- $cur_m = '' unless defined $cur_m;
+ my $cur_s = $cur->header('Subject') // '';
+ my $cur_m = $mime->header('Subject') // '';
if ($cur_s ne $cur_m || norm_body($cur) ne norm_body($mime)) {
return ('MISMATCH', $cur);
}
sub checkpoint {
my ($self) = @_;
- return unless $self->{pid};
+ return unless $self->{in};
print { $self->{out} } "checkpoint\n" or wfail;
undef;
}
sub progress {
my ($self, $msg) = @_;
- return unless $self->{pid};
+ return unless $self->{in};
print { $self->{out} } "progress $msg\n" or wfail;
readline($self->{in}) eq "progress $msg\n" or die
"progress $msg not received\n";
my $env = { GIT_INDEX_FILE => $index };
run_die([@cmd, qw(read-tree -m -v -i), $self->{ref}], $env);
}
- run_die([@cmd, 'update-server-info']);
- my $ibx = $self->{-inbox};
- ($ibx && $self->{path_type} eq '2/38') and eval {
- require PublicInbox::SearchIdx;
- my $s = PublicInbox::SearchIdx->new($ibx);
- $s->index_sync({ ref => $self->{ref} });
- };
- eval { run_die([@cmd, qw(gc --auto)]) } if $do_gc;
+ my $ibx = $self->{ibx};
+ eval { run_die([@cmd, 'update-server-info']) } if $ibx;
+ if ($ibx && $ibx->version == 1 && -d "$ibx->{inboxdir}/public-inbox" &&
+ eval { require PublicInbox::SearchIdx }) {
+ eval {
+ my $s = PublicInbox::SearchIdx->new($ibx);
+ $s->index_sync({ ref => $self->{ref} });
+ };
+ warn "$ibx->{inboxdir} index failed: $@\n" if $@;
+ }
+ if ($do_gc) {
+ my @quiet = (-t STDERR ? () : '-q');
+ eval { run_die([@cmd, qw(gc --auto), @quiet]) }
+ }
}
sub barrier {
# used for v2
sub get_mark {
my ($self, $mark) = @_;
- die "not active\n" unless $self->{pid};
+ die "not active\n" unless $self->{in};
my ($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";
+ my $oid = <$r> // die "get-mark failed, need git 2.6.0+\n";
chomp($oid);
$oid;
}
(($self->{tip} = ":$commit"), $cur);
}
-sub git_timestamp {
- my ($ts, $zone) = @_;
+sub git_timestamp ($) {
+ my ($ts, $zone) = @{$_[0]};
$ts = 0 if $ts < 0; # git uses unsigned times
"$ts $zone";
}
sub extract_cmt_info ($;$) {
my ($mime, $smsg) = @_;
# $mime is PublicInbox::Eml, but remains Email::MIME-compatible
+ $smsg //= bless {}, 'PublicInbox::Smsg';
+
+ $smsg->populate($mime);
my $sender = '';
- my $hdr = $mime->header_obj;
- my $from = $hdr->header('From') // '';
+ my $from = delete($smsg->{From}) // '';
my ($email) = PublicInbox::Address::emails($from);
my ($name) = PublicInbox::Address::names($from);
if (!defined($name) || !defined($email)) {
- $sender = $hdr->header('Sender') // '';
+ $sender = $mime->header('Sender') // '';
$name //= (PublicInbox::Address::names($sender))[0];
$email //= (PublicInbox::Address::emails($sender))[0];
}
warn "no name in From: $from or Sender: $sender\n";
}
- my $subject = $hdr->header('Subject') // '(no subject)';
- # MIME decoding can create nulls replace them with spaces to protect git
- $subject =~ tr/\0/ /;
+ my $subject = delete($smsg->{Subject}) // '(no subject)';
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 $at = git_timestamp(delete $smsg->{-ds});
+ my $ct = git_timestamp(delete $smsg->{-ts});
+ ("$name <$email>", $at, $ct, $subject);
}
# kill potentially confusing/misleading headers
+our @UNWANTED_HEADERS = (qw(Bytes Lines Content-Length),
+ qw(Status X-Status));
sub drop_unwanted_headers ($) {
- my ($mime) = @_;
-
- $mime->header_set($_) for qw(bytes lines content-length status);
- $mime->header_set($_) for @PublicInbox::MDA::BAD_HEADERS;
+ my ($eml) = @_;
+ for (@UNWANTED_HEADERS, @PublicInbox::MDA::BAD_HEADERS) {
+ $eml->header_set($_);
+ }
}
# used by V2Writable, too
}
sub v1_mid0 ($) {
- my ($mime) = @_;
- my $hdr = $mime->header_obj;
- my $mids = mids($hdr);
+ my ($eml) = @_;
+ my $mids = mids($eml);
if (!scalar(@$mids)) { # spam often has no Message-ID
- my $mid0 = digest2mid(content_digest($mime), $hdr);
- append_mid($hdr, $mid0);
+ my $mid0 = digest2mid(content_digest($eml), $eml);
+ append_mid($eml, $mid0);
return $mid0;
}
$mids->[0];
sub add {
my ($self, $mime, $check_cb, $smsg) = @_;
- my ($name, $email, $at, $ct, $subject) = extract_cmt_info($mime, $smsg);
+ my ($author, $at, $ct, $subject) = extract_cmt_info($mime, $smsg);
my $path_type = $self->{path_type};
my $path;
if ($path_type eq '2/38') {
# spam check:
if ($check_cb) {
- $mime = $check_cb->($mime) or return;
+ $mime = $check_cb->($mime, $self->{ibx}) or return;
}
my $blob = $self->{mark}++;
# v2: we need this for Xapian
if ($smsg) {
$smsg->{blob} = $self->get_mark(":$blob");
- $smsg->{bytes} = $n;
- $smsg->{-raw_email} = \$raw_email;
+ $smsg->set_bytes($raw_email, $n);
+ if (my $oidx = delete $smsg->{-oidx}) { # used by LeiStore
+ my $eidx_git = delete $smsg->{-eidx_git};
+
+ # we need this sharedkv to dedupe blobs added in the
+ # same fast-import transaction
+ my $u = $self->{uniq_skv} //= do {
+ require PublicInbox::SharedKV;
+ my $x = PublicInbox::SharedKV->new;
+ $x->dbh;
+ $x;
+ };
+ return if !$u->set_maybe($smsg->oidbin, 1);
+ return if (!$oidx->vivify_xvmd($smsg) &&
+ $eidx_git->check($smsg->{blob}));
+ }
}
my $ref = $self->{ref};
my $commit = $self->{mark}++;
}
print $w "commit $ref\nmark :$commit\n",
- "author $name <$email> $at\n",
+ "author $author $at\n",
"committer $self->{ident} $ct\n" or wfail;
print $w "data ", (length($subject) + 1), "\n",
$subject, "\n\n" or wfail;
$self->{tip} = ":$commit";
}
-sub run_die ($;$$) {
- my ($cmd, $env, $rdr) = @_;
- my $pid = spawn($cmd, $env, $rdr);
- waitpid($pid, 0) == $pid or die join(' ', @$cmd) .' did not finish';
- $? == 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
+my @INIT_FILES = ('HEAD' => undef, # filled in at runtime
'config' => <<EOC);
[core]
repositoryFormatVersion = 0
EOC
sub init_bare {
- my ($dir) = @_; # or self
+ my ($dir, $head) = @_; # 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++];
+ $INIT_FILES[1] //= 'ref: '.default_branch."\n";
+ my @fn_contents = @INIT_FILES;
+ $fn_contents[1] = "ref: refs/heads/$head\n" if defined $head;
+ while (my ($fn, $contents) = splice(@fn_contents, 0, 2)) {
+ my $f = $dir.'/'.$fn;
next if -f $f;
open my $fh, '>', $f or die "open $f: $!";
- print $fh $INIT_FILES[$i] or die "print $f: $!";
+ print $fh $contents or die "print $f: $!";
close $fh or die "close $f: $!";
}
}
+# true if locked and active
+sub active { !!$_[0]->{out} }
+
sub done {
my ($self) = @_;
my $w = delete $self->{out} or return;
- my $r = delete $self->{in} or die 'BUG: missing {in} when done';
- print $w "done\n" or wfail;
- my $pid = delete $self->{pid} or die 'BUG: missing {pid} when done';
- waitpid($pid, 0) == $pid or die 'fast-import did not finish';
- $? == 0 or die "fast-import failed: $?";
-
- _update_git_info($self, 1) if delete $self->{nchg};
-
- $self->lock_release;
-
+ eval {
+ my $r = delete $self->{in} or die 'BUG: missing {in} when done';
+ print $w "done\n" or wfail;
+ close $r or die "fast-import failed: $?"; # ProcessPipe::CLOSE
+ };
+ my $wait_err = $@;
+ my $nchg = delete $self->{nchg};
+ if ($nchg && !$wait_err) {
+ eval { _update_git_info($self, 1) };
+ warn "E: $self->{git}->{git_dir} update info: $@\n" if $@;
+ }
+ $self->lock_release(!!$nchg);
$self->{git}->cleanup;
+ die $wait_err if $wait_err;
}
sub atfork_child {
}
}
-sub digest2mid ($$) {
- my ($dig, $hdr) = @_;
+sub digest2mid ($$;$) {
+ my ($dig, $hdr, $fallback_time) = @_;
my $b64 = $dig->clone->b64digest;
# Make our own URLs nicer:
# See "Base 64 Encoding with URL and Filename Safe Alphabet" in RFC4648
# Add a date prefix to prevent a leading '-' in case that trips
# up some tools (e.g. if a Message-ID were a expected as a
# command-line arg)
- my $dt = msg_datestamp($hdr);
+ my $dt = msg_datestamp($hdr, $fallback_time);
$dt = POSIX::strftime('%Y%m%d%H%M%S', gmtime($dt));
"$dt.$b64" . '@z';
}
sub rewrite_commit ($$$$) {
my ($self, $oids, $buf, $mime) = @_;
- my ($name, $email, $at, $ct, $subject);
+ my ($author, $at, $ct, $subject);
if ($mime) {
- ($name, $email, $at, $ct, $subject) = extract_cmt_info($mime);
+ ($author, $at, $ct, $subject) = extract_cmt_info($mime);
} else {
- $name = $email = '';
+ $author = '<>';
$subject = 'purged '.join(' ', @$oids);
}
@$oids = ();
my $l = $buf->[$i];
if ($l =~ /^author .* ([0-9]+ [\+-]?[0-9]+)$/) {
$at //= $1;
- $buf->[$i] = "author $name <$email> $at\n";
+ $buf->[$i] = "author $author $at\n";
} elsif ($l =~ /^committer .* ([0-9]+ [\+-]?[0-9]+)$/) {
$ct //= $1;
$buf->[$i] = "committer $self->{ident} $ct\n";
my $parsed = PublicInbox::Eml->new($message);
my $ret = $im->add($parsed);
if (!defined $ret) {
- warn "duplicate: ",
- $parsed->header_obj->header_raw('Message-ID'), "\n";
+ warn "duplicate: ", $parsed->header_raw('Message-ID'), "\n";
} else {
print "imported at mark $ret\n";
}