X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FImport.pm;h=ee5ca2ea11b4f383e514ecdf117ac1dabe2c14c1;hb=94ae705673cb03045a109041eec9a6704b8a735b;hp=95d654f667221711de6897ffd632b197bd6d10cf;hpb=6ae5319be740b48370f32fe043d112bf09dd18da;p=public-inbox.git diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm index 95d654f6..ee5ca2ea 100644 --- a/lib/PublicInbox/Import.pm +++ b/lib/PublicInbox/Import.pm @@ -7,14 +7,16 @@ # requires read-only access. package PublicInbox::Import; use strict; -use warnings; -use base qw(PublicInbox::Lock); +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 { @@ -23,17 +25,17 @@ sub new { 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}; - $git ||= $ibx->git; + $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, @@ -46,34 +48,35 @@ sub gfi_start { return ($self->{in}, $self->{out}) if $self->{pid}; - my ($out_r, $out_w); + 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 ($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: $!"; - ($in_r, $out_w); + @ret; } sub wfail () { die "write to fast-import failed: $!" } @@ -137,7 +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 = $mime->header('Subject'); @@ -159,7 +162,7 @@ sub progress { my ($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; } @@ -175,13 +178,16 @@ 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']); - 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, '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; } @@ -212,13 +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); @@ -251,7 +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", @@ -268,29 +274,27 @@ 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_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, @@ -316,27 +320,18 @@ sub extract_cmt_info ($;$) { warn "no name in From: $from or Sender: $sender\n"; } - my $hdr = $mime->header_obj; - - my $subject = $hdr->header('Subject'); - $subject = '(no subject)' unless defined $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 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; } @@ -350,13 +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]; @@ -374,9 +368,9 @@ sub clean_tree_v2 ($$$) { # returns undef on duplicate # returns the :MARK of the most recent commit sub add { - my ($self, $mime, $check_cb, $smsg) = @_; # mime = Email::MIME + 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') { @@ -395,7 +389,7 @@ 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}++; @@ -408,7 +402,7 @@ sub add { # v2: we need this for Xapian if ($smsg) { $smsg->{blob} = $self->get_mark(":$blob"); - $smsg->{bytes} = $n; + $smsg->{raw_bytes} = $n; $smsg->{-raw_email} = \$raw_email; } my $ref = $self->{ref}; @@ -420,7 +414,7 @@ sub add { } 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; @@ -467,20 +461,29 @@ sub init_bare { } } +# 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 { @@ -508,11 +511,11 @@ sub digest2mid ($$) { sub rewrite_commit ($$$$) { my ($self, $oids, $buf, $mime) = @_; - my ($name, $email, $at, $ct, $subject); + my ($author, $at, $ct, $subject); if ($mime) { - ($name, $email, $at, $ct, $subject) = extract_cmt_info($mime); + ($author, $at, $ct, $subject) = extract_cmt_info($mime); } else { - $name = $email = ''; + $author = '<>'; $subject = 'purged '.join(' ', @$oids); } @$oids = (); @@ -521,7 +524,7 @@ sub rewrite_commit ($$$$) { 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"; @@ -553,7 +556,7 @@ sub replace_oids { push @buf, "reset $tmp\n"; } elsif (/^commit (?:.+)/) { if (@buf) { - $w->print(@buf) or wfail; + print $w @buf or wfail; @buf = (); } push @buf, "commit $tmp\n"; @@ -590,7 +593,7 @@ sub replace_oids { rewrite_commit($self, \@oids, \@buf, $mime); $nreplace++; } - $w->print(@buf, "\n") or wfail; + print $w @buf, "\n" or wfail; @buf = (); } elsif ($_ eq "done\n") { $done = 1; @@ -603,7 +606,7 @@ sub replace_oids { } 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; @@ -647,7 +650,10 @@ version 1.0 =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; @@ -663,18 +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"; @@ -689,8 +694,8 @@ version 1.0 =head1 DESCRIPTION -An importer and remover for public-inboxes which takes L -messages as input and stores them in a git repository as +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. @@ -708,7 +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 @@ -719,12 +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, @@ -748,7 +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