2 # Copyright (C) 2020 all contributors <meta@public-inbox.org>
3 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
4 # Basic tool to create a Xapian search index for a public-inbox.
7 use Getopt::Long qw(:config gnu_getopt no_ignore_case auto_abbrev);
8 my $help = <<EOF; # the following should fit w/o scrolling in 80x24 term:
9 usage: public-inbox-eindex [options] EINDEX_DIR [INBOX_DIR]
11 Create and update external (detached) search indices
13 --no-fsync speed up indexing, risk corruption on power outage
14 -L LEVEL `medium', or `full' (default: full)
15 --all index all configured inboxes
16 --jobs=NUM set or disable parallelization (NUM=0)
17 --batch-size=BYTES flush changes to OS after a given number of bytes
18 --max-size=BYTES do not index messages larger than the given size
19 --verbose | -v increase verbosity (may be repeated)
21 BYTES may use `k', `m', and `g' suffixes (e.g. `10m' for 10 megabytes)
22 See public-inbox-eindex(1) man page for full documentation.
24 my $opt = { quiet => -1, compact => 0, max_size => undef, fsync => 1 };
25 GetOptions($opt, qw(verbose|v+ reindex rethread compact|c+ jobs|j=i
27 indexlevel|index-level|L=s max_size|max-size=s
28 batch_size|batch-size=s
29 skip-docdata all help|h))
31 if ($opt->{help}) { print $help; exit 0 };
32 die "--jobs must be >= 0\n" if defined $opt->{jobs} && $opt->{jobs} < 0;
34 # require lazily to speed up --help
35 my $eidx_dir = shift(@ARGV) // die "E: $help";
36 require PublicInbox::Admin;
37 my $cfg = PublicInbox::Config->new;
38 my @ibxs = PublicInbox::Admin::resolve_inboxes(\@ARGV, $opt, $cfg);
39 PublicInbox::Admin::require_or_die(qw(-search));
40 require PublicInbox::ExtSearchIdx;
41 my $eidx = PublicInbox::ExtSearchIdx->new($eidx_dir, $opt);
42 $eidx->attach_inbox($_) for @ibxs;
43 $eidx->eidx_sync($opt);