X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FImport.pm;h=ee5ca2ea11b4f383e514ecdf117ac1dabe2c14c1;hb=94ae705673cb03045a109041eec9a6704b8a735b;hp=bfa7a8053297a50c44c5e1f4fe173fbd7f882bbf;hpb=9294716ecc685c7b21ec21e2dbbbeb2c8f62c477;p=public-inbox.git diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm index bfa7a805..ee5ca2ea 100644 --- a/lib/PublicInbox/Import.pm +++ b/lib/PublicInbox/Import.pm @@ -1,35 +1,41 @@ -# Copyright (C) 2016-2018 all contributors +# Copyright (C) 2016-2020 all contributors # License: AGPL-3.0+ # # git fast-import-based ssoma-mda MDA replacement -# This is only ever run by public-inbox-mda and public-inbox-learn, -# not the WWW or NNTP code which only requires read-only access. +# This is only ever run by public-inbox-mda, public-inbox-learn +# and public-inbox-watch. Not the WWW or NNTP code which only +# requires read-only access. package PublicInbox::Import; use strict; -use warnings; -use base qw(PublicInbox::Lock); -use PublicInbox::Spawn qw(spawn); -use PublicInbox::MID qw(mids mid_mime mid2path); +use parent qw(PublicInbox::Lock); +use v5.10.1; +use PublicInbox::Spawn qw(spawn popen_rd); +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::Smsg; +use PublicInbox::MsgTime qw(msg_datestamp); +use PublicInbox::ContentHash qw(content_digest); use PublicInbox::MDA; +use PublicInbox::Eml; use POSIX qw(strftime); 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 $ref = 'refs/heads/master'; if ($ibx) { - $ref = $ibx->{ref_head} || 'refs/heads/master'; - $name ||= $ibx->{name}; - $email ||= $ibx->{-primary_address}; + $ref = $ibx->{ref_head} // 'refs/heads/master'; + $name //= $ibx->{name}; + $email //= $ibx->{-primary_address}; + $git //= $ibx->git; } bless { git => $git, ident => "$name <$email>", mark => 1, ref => $ref, - inbox => $ibx, + ibx => $ibx, path_type => '2/38', # or 'v2' lock_path => "$git->{git_dir}/ssoma.lock", # v2 changes this bytes_added => 0, @@ -42,37 +48,35 @@ sub gfi_start { return ($self->{in}, $self->{out}) if $self->{pid}; - my ($in_r, $in_w, $out_r, $out_w); - pipe($in_r, $in_w) or die "pipe failed: $!"; + my (@ret, $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 { $_ => 1 } @tree }; + eval { + my ($git, $ref) = @$self{qw(git ref)}; + local $/ = "\n"; + chomp($self->{tip} = $git->qx(qw(rev-parse --revs-only), $ref)); + if ($self->{path_type} ne '2/38' && $self->{tip}) { + local $/ = "\0"; + my @t = $git->qx(qw(ls-tree -r -z --name-only), $ref); + chomp @t; + $self->{-tree} = { map { $_ => 1 } @t }; + } + my @cmd = ('git', "--git-dir=$git->{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; + $self->{nchg} = 0; + @ret = ($in_r, $out_w); + }; + if ($@) { + $self->lock_release; + die $@; } - - my $git_dir = $git->{git_dir}; - my @cmd = ('git', "--git-dir=$git_dir", qw(fast-import - --quiet --done --date-format=raw)); - my $rdr = { 0 => fileno($out_r), 1 => fileno($in_w) }; - my $pid = spawn(\@cmd, undef, $rdr); - die "spawn fast-import failed: $!" unless defined $pid; - $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: $!"; - ($in_r, $out_w); + @ret; } sub wfail () { die "write to fast-import failed: $!" } @@ -102,7 +106,7 @@ sub _cat_blob ($$$) { local $/ = "\n"; my $info = <$r>; defined $info or die "EOF from fast-import / cat-blob: $!"; - $info =~ /\A[a-f0-9]{40} blob (\d+)\n\z/ or return; + $info =~ /\A[a-f0-9]{40} blob ([0-9]+)\n\z/ or return; my $left = $1; my $offset = 0; my $buf = ''; @@ -136,7 +140,7 @@ sub check_remove_v1 { $info =~ m!\A100644 blob ([a-f0-9]{40})\t!s or die "not blob: $info"; my $oid = $1; my $msg = _cat_blob($r, $w, $oid) or die "BUG: cat-blob $1 failed"; - my $cur = PublicInbox::MIME->new($msg); + my $cur = PublicInbox::Eml->new($msg); my $cur_s = $cur->header('Subject'); $cur_s = '' unless defined $cur_s; my $cur_m = $mime->header('Subject'); @@ -158,7 +162,7 @@ sub progress { my ($self, $msg) = @_; return unless $self->{pid}; print { $self->{out} } "progress $msg\n" or wfail; - $self->{in}->getline eq "progress $msg\n" or die + readline($self->{in}) eq "progress $msg\n" or die "progress $msg not received\n"; undef; } @@ -174,14 +178,17 @@ sub _update_git_info ($$) { my $env = { GIT_INDEX_FILE => $index }; run_die([@cmd, qw(read-tree -m -v -i), $self->{ref}], $env); } - run_die([@cmd, 'update-server-info'], undef); - ($self->{path_type} eq '2/38') and eval { - require PublicInbox::SearchIdx; - my $inbox = $self->{inbox} || $git_dir; - my $s = PublicInbox::SearchIdx->new($inbox); - $s->index_sync({ ref => $self->{ref} }); - }; - eval { run_die([@cmd, qw(gc --auto)], undef) } if $do_gc; + eval { run_die([@cmd, 'update-server-info']) }; + my $ibx = $self->{ibx}; + if ($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 $@; + } + eval { run_die([@cmd, qw(gc --auto)]) } if $do_gc; } sub barrier { @@ -211,13 +218,13 @@ sub get_mark { } # returns undef on non-existent -# ('MISMATCH', Email::MIME) on mismatch -# (:MARK, Email::MIME) on success +# ('MISMATCH', PublicInbox::Eml) on mismatch +# (:MARK, PublicInbox::Eml) on success # # v2 callers should check with Xapian before calling this as # it is not idempotent. sub remove { - my ($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); @@ -250,7 +257,7 @@ sub remove { } my $ident = $self->{ident}; my $now = now_raw(); - $msg ||= 'rm'; + $msg //= 'rm'; my $len = length($msg) + 1; print $w "commit $ref\nmark :$commit\n", "author $ident $now\n", @@ -267,31 +274,33 @@ sub remove { (($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 $zone"; } -sub extract_author_info ($) { - my ($mime) = @_; +sub extract_cmt_info ($;$) { + my ($mime, $smsg) = @_; + # $mime is PublicInbox::Eml, but remains Email::MIME-compatible + $smsg //= bless {}, 'PublicInbox::Smsg'; + + $smsg->populate($mime); my $sender = ''; - my $from = $mime->header('From'); - $from ||= ''; + my $from = delete($smsg->{From}) // ''; my ($email) = PublicInbox::Address::emails($from); my ($name) = PublicInbox::Address::names($from); if (!defined($name) || !defined($email)) { - $sender = $mime->header('Sender'); - $sender ||= ''; - if (!defined($name)) { - ($name) = PublicInbox::Address::names($sender); - } - if (!defined($email)) { - ($email) = PublicInbox::Address::emails($sender); - } + $sender = $mime->header('Sender') // ''; + $name //= (PublicInbox::Address::names($sender))[0]; + $email //= (PublicInbox::Address::emails($sender))[0]; } 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 { @@ -310,14 +319,19 @@ sub extract_author_info ($) { $name = ''; warn "no name in From: $from or Sender: $sender\n"; } - ($name, $email); + + my $subject = delete($smsg->{Subject}) // '(no subject)'; + utf8::encode($subject); + my $at = git_timestamp(delete $smsg->{-ds}); + my $ct = git_timestamp(delete $smsg->{-ts}); + ("$name <$email>", $at, $ct, $subject); } # kill potentially confusing/misleading headers sub drop_unwanted_headers ($) { my ($mime) = @_; - $mime->header_set($_) for qw(bytes lines content-length status); + $mime->header_set($_) for qw(Bytes Lines Content-Length Status); $mime->header_set($_) for @PublicInbox::MDA::BAD_HEADERS; } @@ -331,13 +345,12 @@ sub append_mid ($$) { } sub v1_mid0 ($) { - my ($mime) = @_; - my $hdr = $mime->header_obj; - my $mids = mids($hdr); + my ($eml) = @_; + my $mids = mids($eml); - if (!scalar(@$mids)) { # spam often has no Message-Id - my $mid0 = digest2mid(content_digest($mime), $hdr); - append_mid($hdr, $mid0); + if (!scalar(@$mids)) { # spam often has no Message-ID + my $mid0 = digest2mid(content_digest($eml), $eml); + append_mid($eml, $mid0); return $mid0; } $mids->[0]; @@ -355,18 +368,10 @@ sub clean_tree_v2 ($$$) { # returns undef on duplicate # returns the :MARK of the most recent commit sub add { - my ($self, $mime, $check_cb) = @_; # mime = Email::MIME - - my ($name, $email) = extract_author_info($mime); - my $hdr = $mime->header_obj; - my @at = msg_datestamp($hdr); - my @ct = msg_timestamp($hdr); - my $author_time_raw = git_timestamp(@at); - my $commit_time_raw = git_timestamp(@ct); - my $subject = $mime->header('Subject'); - $subject = '(no subject)' unless defined $subject; - my $path_type = $self->{path_type}; + my ($self, $mime, $check_cb, $smsg) = @_; + my ($author, $at, $ct, $subject) = extract_cmt_info($mime, $smsg); + my $path_type = $self->{path_type}; my $path; if ($path_type eq '2/38') { $path = mid2path(v1_mid0($mime)); @@ -384,20 +389,21 @@ sub add { # spam check: if ($check_cb) { - $mime = $check_cb->($mime) or return; + $mime = $check_cb->($mime, $self->{ibx}) or return; } my $blob = $self->{mark}++; - my $str = $mime->as_string; - my $n = length($str); + my $raw_email = $mime->{-public_inbox_raw} // $mime->as_string; + my $n = length($raw_email); $self->{bytes_added} += $n; print $w "blob\nmark :$blob\ndata ", $n, "\n" or wfail; - print $w $str, "\n" or wfail; + 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, $n, \$str ]; + if ($smsg) { + $smsg->{blob} = $self->get_mark(":$blob"); + $smsg->{raw_bytes} = $n; + $smsg->{-raw_email} = \$raw_email; } my $ref = $self->{ref}; my $commit = $self->{mark}++; @@ -407,12 +413,9 @@ sub add { print $w "reset $ref\n" or wfail; } - # Mime decoding can create nulls replace them with spaces to protect git - $subject =~ tr/\0/ /; - utf8::encode($subject); print $w "commit $ref\nmark :$commit\n", - "author $name <$email> $author_time_raw\n", - "committer $self->{ident} $commit_time_raw\n" or wfail; + "author $author $at\n", + "committer $self->{ident} $ct\n" or wfail; print $w "data ", (length($subject) + 1), "\n", $subject, "\n\n" or wfail; if ($tip ne '') { @@ -427,25 +430,60 @@ sub add { sub run_die ($;$$) { my ($cmd, $env, $rdr) = @_; my $pid = spawn($cmd, $env, $rdr); - defined $pid or die "spawning ".join(' ', @$cmd)." failed: $!"; waitpid($pid, 0) == $pid or die join(' ', @$cmd) .' did not finish'; $? == 0 or die join(' ', @$cmd) . " failed: $?\n"; } +my @INIT_FILES = ('HEAD' => "ref: refs/heads/master\n", + 'description' => < <{git}->{git_dir} if ref($dir); + require File::Path; + File::Path::mkpath([ map { "$dir/$_" } qw(objects/info refs/heads) ]); + for (my $i = 0; $i < @INIT_FILES; $i++) { + my $f = $dir.'/'.$INIT_FILES[$i++]; + next if -f $f; + open my $fh, '>', $f or die "open $f: $!"; + print $fh $INIT_FILES[$i] or die "print $f: $!"; + close $fh or die "close $f: $!"; + } +} + +# true if locked and active +sub active { !!$_[0]->{out} } + sub done { my ($self) = @_; my $w = delete $self->{out} or return; - my $r = delete $self->{in} or die 'BUG: missing {in} when done'; - print $w "done\n" or wfail; - my $pid = delete $self->{pid} or die 'BUG: missing {pid} when done'; - waitpid($pid, 0) == $pid or die 'fast-import did not finish'; - $? == 0 or die "fast-import failed: $?"; - - _update_git_info($self, 1) if delete $self->{nchg}; - - $self->lock_release; - + eval { + my $r = delete $self->{in} or die 'BUG: missing {in} when done'; + print $w "done\n" or wfail; + my $pid = delete $self->{pid} or + die 'BUG: missing {pid} when done'; + waitpid($pid, 0) == $pid or die 'fast-import did not finish'; + $? == 0 or die "fast-import failed: $?"; + }; + my $wait_err = $@; + my $nchg = delete $self->{nchg}; + if ($nchg && !$wait_err) { + eval { _update_git_info($self, 1) }; + warn "E: $self->{git}->{git_dir} update info: $@\n" if $@; + } + $self->lock_release(!!$nchg); $self->{git}->cleanup; + die $wait_err if $wait_err; } sub atfork_child { @@ -471,33 +509,45 @@ sub digest2mid ($$) { "$dt.$b64" . '@z'; } -sub clean_purge_buffer { - my ($oids, $buf) = @_; - my $cmt_msg = 'purged '.join(' ',@$oids)."\n"; +sub rewrite_commit ($$$$) { + my ($self, $oids, $buf, $mime) = @_; + my ($author, $at, $ct, $subject); + if ($mime) { + ($author, $at, $ct, $subject) = extract_cmt_info($mime); + } else { + $author = '<>'; + $subject = 'purged '.join(' ', @$oids); + } @$oids = (); - + $subject .= "\n"; foreach my $i (0..$#$buf) { my $l = $buf->[$i]; - if ($l =~ /^author .* (\d+ [\+-]?\d+)$/) { - $buf->[$i] = "author <> $1\n"; - } elsif ($l =~ /^data (\d+)/) { - $buf->[$i++] = "data " . length($cmt_msg) . "\n"; - $buf->[$i] = $cmt_msg; + if ($l =~ /^author .* ([0-9]+ [\+-]?[0-9]+)$/) { + $at //= $1; + $buf->[$i] = "author $author $at\n"; + } elsif ($l =~ /^committer .* ([0-9]+ [\+-]?[0-9]+)$/) { + $ct //= $1; + $buf->[$i] = "committer $self->{ident} $ct\n"; + } elsif ($l =~ /^data ([0-9]+)/) { + $buf->[$i++] = "data " . length($subject) . "\n"; + $buf->[$i] = $subject; last; } } } -sub purge_oids { - my ($self, $purge) = @_; - my $tmp = "refs/heads/purge-".((keys %$purge)[0]); +# returns the new commit OID if a replacement was done +# returns undef if nothing was done +sub replace_oids { + my ($self, $mime, $replace_map) = @_; # oid => raw string + my $tmp = "refs/heads/replace-".((keys %$replace_map)[0]); my $old = $self->{'ref'}; my $git = $self->{git}; my @export = (qw(fast-export --no-data --use-done-feature), $old); - my ($rd, $pid) = $git->popen(@export); + my $rd = $git->popen(@export); my ($r, $w) = $self->gfi_start; my @buf; - my $npurge = 0; + my $nreplace = 0; my @oids; my ($done, $mark); my $tree = $self->{-tree}; @@ -506,11 +556,11 @@ sub purge_oids { 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"; - } elsif (/^data (\d+)/) { + } elsif (/^data ([0-9]+)/) { # only commit message, so $len is small: my $len = $1; # + 1 for trailing "\n" push @buf, $_; @@ -519,11 +569,15 @@ sub purge_oids { push @buf, $buf; } elsif (/^M 100644 ([a-f0-9]+) (\w+)/) { my ($oid, $path) = ($1, $2); - if ($purge->{$oid}) { + $tree->{$path} = 1; + my $sref = $replace_map->{$oid}; + if (defined $sref) { push @oids, $oid; - delete $tree->{$path}; + my $n = length($$sref); + push @buf, "M 100644 inline $path\ndata $n\n"; + push @buf, $$sref; # hope CoW works... + push @buf, "\n"; } else { - $tree->{$path} = 1; push @buf, $_; } } elsif (/^D (\w+)/) { @@ -531,49 +585,54 @@ sub purge_oids { 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); - $npurge++; + if (!$mime) { + my $out = join('', @buf); + $out =~ s/^/# /sgm; + warn "purge rewriting\n", $out, "\n"; + } + rewrite_commit($self, \@oids, \@buf, $mime); + $nreplace++; } - $w->print(@buf, "\n") or wfail; + print $w @buf, "\n" or wfail; @buf = (); } elsif ($_ eq "done\n") { $done = 1; - } elsif (/^mark :(\d+)$/) { + } elsif (/^mark :([0-9]+)$/) { push @buf, $_; $mark = $1; } else { push @buf, $_; } } + 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 $npurge; + chomp(my $cmt = $self->get_mark(":$mark")) if $nreplace; $self->{nchg} = 0; # prevent _update_git_info until update-ref: $self->done; my @git = ('git', "--git-dir=$git->{git_dir}"); - run_die([@git, qw(update-ref), $old, $tmp]) if $npurge; + run_die([@git, qw(update-ref), $old, $tmp]) if $nreplace; run_die([@git, qw(update-ref -d), $tmp]); - return if $npurge == 0; + return if $nreplace == 0; - run_die([@git, qw(-c gc.reflogExpire=now gc --prune=all)]); + run_die([@git, qw(-c gc.reflogExpire=now gc --prune=all --quiet)]); + + # check that old OIDs are gone my $err = 0; - foreach my $oid (keys %$purge) { + foreach my $oid (keys %$replace_map) { my @info = $git->check($oid); if (@info) { - warn "$oid not purged\n"; + warn "$oid not replaced\n"; $err++; } } _update_git_info($self, 0); - die "Failed to purge $err object(s)\n" if $err; + die "Failed to replace $err object(s)\n" if $err; $cmt; } @@ -583,15 +642,18 @@ __END__ =head1 NAME -PublicInbox::Import - message importer for public-inbox +PublicInbox::Import - message importer for public-inbox v1 inboxes =head1 VERSION version 1.0 -=head1 SYNOPSYS +=head1 SYNOPSIS + + use PublicInbox::Eml; + # PublicInbox::Eml exists as of public-inbox 1.5.0, + # Email::MIME was used in older versions - use Email::MIME; use PublicInbox::Git; use PublicInbox::Import; @@ -607,18 +669,17 @@ version 1.0 "Date: Thu, 01 Jan 1970 00:00:00 +0000\n" . "Message-ID: \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: ", - $parsed->header_obj->header_raw('Message-ID'), "\n"; + warn "duplicate: ", $parsed->header_raw('Message-ID'), "\n"; } else { print "imported at mark $ret\n"; } $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"; @@ -633,9 +694,9 @@ version 1.0 =head1 DESCRIPTION -An importer and remover for public-inboxes which takes L -messages as input and stores them in a ssoma repository as -documented in L, +An importer and remover for public-inboxes which takes C +or L messages as input and stores them in a git repository as +documented in L, except it does not allow duplicate Message-IDs. It requires L and L to be installed. @@ -652,7 +713,7 @@ Initialize a new PublicInbox::Import object. =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 @@ -663,12 +724,13 @@ is called, but L may be called on them. =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, @@ -692,7 +754,7 @@ The mail archives are hosted at L =head1 COPYRIGHT -Copyright (C) 2016 all contributors L +Copyright (C) 2016-2020 all contributors L License: AGPL-3.0+ L