X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=blobdiff_plain;f=script%2Fpublic-inbox-index;h=a04be9fcb4f18456777f3eb0e0295c9cd283ce96;hp=61f21d7060c345378ab1e96893e4b589968a89b5;hb=refs%2Fheads%2Fmaster;hpb=3f4c9ce3d2c0796c6c3f352b5b586ddd7a85ffb0 diff --git a/script/public-inbox-index b/script/public-inbox-index index 61f21d70..a04be9fc 100755 --- a/script/public-inbox-index +++ b/script/public-inbox-index @@ -1,68 +1,156 @@ -#!/usr/bin/perl -w -# Copyright (C) 2015 all contributors -# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) -# Basic tool to create a Xapian search index for a git repository -# configured for public-inbox. +#!perl -w +# Copyright (C) 2015-2021 all contributors +# License: AGPL-3.0+ +# Basic tool to create a Xapian search index for a public-inbox. # Usage with libeatmydata -# highly recommended: eatmydata public-inbox-index GIT_DIR +# highly recommended: eatmydata public-inbox-index INBOX_DIR use strict; -use warnings; +use v5.10.1; use Getopt::Long qw(:config gnu_getopt no_ignore_case auto_abbrev); -my $usage = "public-inbox-index GIT_DIR"; -use PublicInbox::Config; -eval { require PublicInbox::SearchIdx }; -if ($@) { - print STDERR "Search::Xapian required for $0\n"; - exit 1; +my $help = < -1, compact => 0, max_size => undef, fsync => 1, + 'update-extindex' => [], # ":s@" optional arg sets '' if no arg given +}; +GetOptions($opt, qw(verbose|v+ reindex rethread compact|c+ jobs|j=i prune + fsync|sync! xapian_only|xapian-only dangerous + indexlevel|index-level|L=s max_size|max-size=s + batch_size|batch-size=s + since|after=s until|before=s + sequential-shard|seq-shard + no-update-extindex update-extindex|E=s@ + fast-noop|F skip-docdata all C=s@ help|h)) + or die $help; +if ($opt->{help}) { print $help; exit 0 }; +die "--jobs must be >= 0\n" if defined $opt->{jobs} && $opt->{jobs} < 0; +if ($opt->{xapian_only} && !$opt->{reindex}) { + die "--xapian-only requires --reindex\n"; +} +if ($opt->{reindex} && delete($opt->{'fast-noop'})) { + warn "--fast-noop ignored with --reindex\n"; } -my $reindex; -my %opts = ( '--reindex' => \$reindex ); -GetOptions(%opts) or die "bad command-line args\n$usage"; +# require lazily to speed up --help +require PublicInbox::Admin; +PublicInbox::Admin::require_or_die('-index'); +PublicInbox::Admin::do_chdir(delete $opt->{C}); -my @dirs; +my $cfg = PublicInbox::Config->new; # Config is loaded by Admin +$opt->{-use_cwd} = 1; +my @ibxs = PublicInbox::Admin::resolve_inboxes(\@ARGV, $opt, $cfg); +PublicInbox::Admin::require_or_die('-index'); +unless (@ibxs) { print STDERR $help; exit 1 } -sub resolve_git_dir { - my ($cd) = @_; - my @cmd = qw(git rev-parse --git-dir); - my $cmd = join(' ', @cmd); - my $pid = open my $fh, '-|'; - defined $pid or die "forking $cmd failed: $!\n"; - if ($pid == 0) { - if (defined $cd) { - chdir $cd or die "chdir $cd failed: $!\n"; - } - exec @cmd; - die "Failed to exec $cmd: $!\n"; +my (@eidx, %eidx_seen); +my $update_extindex = $opt->{'update-extindex'}; +if (!scalar(@$update_extindex) && (my $ALL = $cfg->ALL)) { + # extindex and normal inboxes may have different owners + push(@$update_extindex, 'all') if -w $ALL->{topdir}; +} +@$update_extindex = () if $opt->{'no-update-extindex'}; +if (scalar @$update_extindex) { + PublicInbox::Admin::require_or_die('-search'); + require PublicInbox::ExtSearchIdx; +} +for my $ei_name (@$update_extindex) { + my $es = $cfg->lookup_ei($ei_name); + my $topdir; + if (!$es && -d $ei_name) { # allow dirname or config section name + $topdir = $ei_name; + } elsif ($es) { + $topdir = $es->{topdir}; } else { - my $dir = eval { - local $/; - <$fh>; - }; - close $fh or die "error in $cmd: $!\n"; - chomp $dir; - return $cd if ($dir eq '.' && defined $cd); - $dir; + die "extindex `$ei_name' not configured or found\n"; } + my $o = { %$opt }; + delete $o->{indexlevel} if ($o->{indexlevel}//'') eq 'basic'; + $eidx_seen{$topdir} //= + push(@eidx, PublicInbox::ExtSearchIdx->new($topdir, $o)); } - -if (@ARGV) { - @dirs = map { resolve_git_dir($_) } @ARGV; -} else { - @dirs = (resolve_git_dir()); +my $mods = {}; +my @eidx_unconfigured; +foreach my $ibx (@ibxs) { + # detect_indexlevel may also set $ibx->{-skip_docdata} + my $detected = PublicInbox::Admin::detect_indexlevel($ibx); + # XXX: users can shoot themselves in the foot, with opt->{indexlevel} + $ibx->{indexlevel} //= $opt->{indexlevel} // ($opt->{xapian_only} ? + 'full' : $detected); + PublicInbox::Admin::scan_ibx_modules($mods, $ibx); + if (@eidx && $ibx->{-unconfigured}) { + push @eidx_unconfigured, " $ibx->{inboxdir}\n"; + } } +warn <{compact} = 0 if !$mods->{'Search::Xapian'}; -foreach my $dir (@dirs) { - index_dir($dir); -} +PublicInbox::Admin::require_or_die(keys %$mods); +my $env = PublicInbox::Admin::index_prepare($opt, $cfg); +local %ENV = (%ENV, %$env) if $env; +require PublicInbox::InboxWritable; +PublicInbox::Xapcmd::check_compact() if $opt->{compact}; +PublicInbox::Admin::progress_prepare($opt); +for my $ibx (@ibxs) { + $ibx = PublicInbox::InboxWritable->new($ibx); + if ($opt->{compact} >= 2) { + PublicInbox::Xapcmd::run($ibx, 'compact', $opt->{compact_opt}); + } + $ibx->{-no_fsync} = 1 if !$opt->{fsync}; + $ibx->{-dangerous} = 1 if $opt->{dangerous}; + $ibx->{-skip_docdata} //= $opt->{'skip-docdata'}; -sub index_dir { - my ($git_dir) = @_; - -d $git_dir or die "$git_dir does not appear to be a git repository\n"; - my $s = PublicInbox::SearchIdx->new($git_dir, 1); - $s->index_sync({ reindex => $reindex }); + my $ibx_opt = $opt; + if (defined(my $s = $ibx->{lc('indexSequentialShard')})) { + defined(my $v = $cfg->git_bool($s)) or die <{name}.indexSequentialShard not boolean +EOL + $ibx_opt = { %$opt, 'sequential-shard' => $v }; + } + my $nidx = PublicInbox::Admin::index_inbox($ibx, undef, $ibx_opt); + last if $ibx_opt->{quit}; + if (my $copt = $opt->{compact_opt}) { + local $copt->{jobs} = 0 if $ibx_opt->{'sequential-shard'}; + PublicInbox::Xapcmd::run($ibx, 'compact', $copt); + } + last if $ibx_opt->{quit}; + next if $ibx->{-unconfigured} || !$nidx; + for my $eidx (@eidx) { + $eidx->attach_inbox($ibx); + } +} +my $pr = $opt->{-progress}; +for my $eidx (@eidx) { + $pr->("indexing $eidx->{topdir} ...\n") if $pr; + $eidx->eidx_sync($opt); + last if $opt->{quit}; }