user/dev discussion of public-inbox itself
 help / color / mirror / code / Atom feed
Search results ordered by [date|relevance]  view[summary|nested|Atom feed]
thread overview below | download mbox.gz: |
* [PATCH 6/6] admin: improve warnings and errors for missing modules
  2019-05-15  6:33  7% [PATCH 0/6] make Search::Xapian optional for v2, NNTP, etc Eric Wong
@ 2019-05-15  6:33  5% ` Eric Wong
  0 siblings, 0 replies; 2+ results
From: Eric Wong @ 2019-05-15  6:33 UTC (permalink / raw)
  To: meta

Since we lazy-load Xapian now, some errors may become
more cryptic or buried.  Try to improve that by making
Admin show better errors.
---
 lib/PublicInbox/Admin.pm  | 60 +++++++++++++++++++++++++++++++++++++++
 script/public-inbox-index | 32 ++++++++++++---------
 script/public-inbox-init  |  7 +++--
 script/public-inbox-purge | 53 ++++++++++++++++++++++++++--------
 4 files changed, 124 insertions(+), 28 deletions(-)

diff --git a/lib/PublicInbox/Admin.pm b/lib/PublicInbox/Admin.pm
index d0a8dd0..3eff5cd 100644
--- a/lib/PublicInbox/Admin.pm
+++ b/lib/PublicInbox/Admin.pm
@@ -41,4 +41,64 @@ sub resolve_repo_dir {
 	}
 }
 
+# TODO: make Devel::Peek optional, only used for daemon
+my @base_mod = qw(Email::MIME Date::Parse Devel::Peek);
+my @over_mod = qw(DBD::SQLite DBI);
+my %mod_groups = (
+	-index => [ @base_mod, @over_mod ],
+	-base => \@base_mod,
+	-search => [ @base_mod, @over_mod, 'Search::Xapian' ],
+);
+
+sub scan_ibx_modules ($$) {
+	my ($mods, $ibx) = @_;
+	if (!$ibx->{indexlevel} || $ibx->{indexlevel} ne 'basic') {
+		$mods->{'Search::Xapian'} = 1;
+	} else {
+		$mods->{$_} = 1 foreach @over_mod;
+	}
+}
+
+sub check_require {
+	my (@mods) = @_;
+	my $err = {};
+	while (my $mod = shift @mods) {
+		if (my $groups = $mod_groups{$mod}) {
+			push @mods, @$groups;
+		} else {
+			eval "require $mod";
+			$err->{$mod} = $@ if $@;
+		}
+	}
+	scalar keys %$err ? $err : undef;
+}
+
+sub missing_mod_msg {
+	my ($err) = @_;
+	my @mods = map { "`$_'" } sort keys %$err;
+	my $last = pop @mods;
+	@mods ? (join(', ', @mods)."' and $last") : $last
+}
+
+sub require_or_die {
+	my $err = check_require(@_) or return;
+	die missing_mod_msg($err)." required for $0\n";
+}
+
+sub indexlevel_ok_or_die ($) {
+	my ($indexlevel) = @_;
+	my $req;
+	if ($indexlevel eq 'basic') {
+		$req = '-index';
+	} elsif ($indexlevel =~ /\A(?:medium|full)\z/) {
+		$req = '-search';
+	} else {
+		die <<"";
+invalid indexlevel=$indexlevel (must be `basic', `medium', or `full')
+
+	}
+	my $err = check_require($req) or return;
+	die missing_mod_msg($err) ." required for indexlevel=$indexlevel\n";
+}
+
 1;
diff --git a/script/public-inbox-index b/script/public-inbox-index
index 53def9a..cf001cc 100755
--- a/script/public-inbox-index
+++ b/script/public-inbox-index
@@ -10,18 +10,14 @@ use strict;
 use warnings;
 use Getopt::Long qw(:config gnu_getopt no_ignore_case auto_abbrev);
 my $usage = "public-inbox-index REPO_DIR";
-use PublicInbox::Config;
 use PublicInbox::Admin qw(resolve_repo_dir);
+PublicInbox::Admin::require_or_die('-index');
+require PublicInbox::Config;
 
 my $config = eval { PublicInbox::Config->new } || eval {
 	warn "public-inbox unconfigured for serving, indexing anyways...\n";
 	undef;
 };
-eval { require PublicInbox::SearchIdx };
-if ($@) {
-	print STDERR "Search::Xapian required for $0\n";
-	exit 1;
-}
 
 my $reindex;
 my $prune;
@@ -56,26 +52,34 @@ defined($config) and $config->each_inbox(sub {
 	}
 });
 
+my @inboxes;
+my $mods = {};
+
 foreach my $dir (@dirs) {
-	if (!ref($dir)) {
+	my $ibx = $dir;
+	if (!ref($ibx)) {
 		unless (-d $dir) {
 			die "$dir does not appear to be an inbox repository\n";
 		}
-		my $ibx = {
+		$ibx = PublicInbox::Inbox->new({
 			mainrepo => $dir,
 			name => 'unnamed',
 			indexlevel => $indexlevel,
 			version => -f "$dir/inbox.lock" ? 2 : 1,
-		};
-		$dir = PublicInbox::Inbox->new($ibx);
-	} elsif (defined $indexlevel && !defined($dir->{indexlevel})) {
+		});
+	} elsif (defined $indexlevel && !defined($ibx->{indexlevel})) {
 		# XXX: users can shoot themselves in the foot, with this...
-		$dir->{indexlevel} = $indexlevel;
+		$ibx->{indexlevel} = $indexlevel;
 	}
-
-	index_inbox($dir);
+	push @inboxes, $ibx;
+	PublicInbox::Admin::scan_ibx_modules($mods, $ibx);
 }
 
+PublicInbox::Admin::require_or_die(keys %$mods);
+
+require PublicInbox::SearchIdx;
+index_inbox($_) for @inboxes;
+
 sub index_inbox {
 	my ($repo) = @_;
 	if (ref($repo) && ($repo->{version} || 1) == 2) {
diff --git a/script/public-inbox-init b/script/public-inbox-init
index 9f0bd1b..5516e79 100755
--- a/script/public-inbox-init
+++ b/script/public-inbox-init
@@ -7,8 +7,10 @@ use strict;
 use warnings;
 my $usage = "public-inbox-init NAME REPO_DIR HTTP_URL ADDRESS [ADDRESS..]";
 use Getopt::Long qw/:config gnu_getopt no_ignore_case auto_abbrev/;
-use PublicInbox::Config;
-use PublicInbox::Inbox;
+use PublicInbox::Admin;
+PublicInbox::Admin::require_or_die('-base');
+require PublicInbox::Config;
+require PublicInbox::Inbox;
 use File::Temp qw/tempfile/;
 use File::Basename qw/dirname/;
 use File::Path qw/mkpath/;
@@ -24,6 +26,7 @@ my %opts = ( 'V|version=i' => \$version,
 	     'S|skip=i' => \$skip,
 );
 GetOptions(%opts) or usage();
+PublicInbox::Admin::indexlevel_ok_or_die($indexlevel) if defined $indexlevel;
 my $name = shift @ARGV or usage();
 my $mainrepo = shift @ARGV or usage();
 my $http_url = shift @ARGV or usage();
diff --git a/script/public-inbox-purge b/script/public-inbox-purge
index 264bcde..381826d 100755
--- a/script/public-inbox-purge
+++ b/script/public-inbox-purge
@@ -7,16 +7,16 @@
 use strict;
 use warnings;
 use Getopt::Long qw(:config gnu_getopt no_ignore_case auto_abbrev);
-use PublicInbox::Config;
-use PublicInbox::MIME;
 use PublicInbox::Admin qw(resolve_repo_dir);
-use PublicInbox::Filter::Base;
-*REJECT = *PublicInbox::Filter::Base::REJECT;
+PublicInbox::Admin::check_require('-index');
+require PublicInbox::Filter::Base;
+require PublicInbox::Config;
+require PublicInbox::MIME;
+require PublicInbox::V2Writable;
 
-my $usage = "$0 [--all] [INBOX_DIRS] </path/to/message";
+{ no warnings 'once'; *REJECT = *PublicInbox::Filter::Base::REJECT }
 
-eval { require PublicInbox::V2Writable } or die
-	"DBI, DBD::SQLite and Search::Xapian required for purge\n";
+my $usage = "$0 [--all] [INBOX_DIRS] </path/to/message";
 my $config = eval { PublicInbox::Config->new };
 my $cfgfile = PublicInbox::Config::default_file();
 my ($all, $force);
@@ -63,11 +63,6 @@ if ($all) {
 				mainrepo => $dir,
 			});
 		};
-
-		# somebody could "rm -r" all the Xapian directories;
-		# let them purge the overview, at least
-		$ibx->{indexlevel} ||= 'basic' unless $ibx->search;
-
 		push @inboxes, $ibx;
 	}
 
@@ -77,6 +72,40 @@ if ($all) {
 	}
 }
 
+foreach my $ibx (@inboxes) {
+	my $lvl = $ibx->{indexlevel};
+	if (defined $lvl) {
+		PublicInbox::Admin::indexlevel_ok_or_die($lvl);
+		next;
+	}
+
+	# Undefined indexlevel, so `full'...
+	# Search::Xapian exists and the DB can be read, at least, fine
+	$ibx->search and next;
+
+	# it's possible for a Xapian directory to exist, but Search::Xapian
+	# to go missing/broken.  Make sure it's purged in that case:
+	$ibx->over or die "no over.sqlite3 in $ibx->{mainrepo}\n";
+
+	# $ibx->{search} is populated by $ibx->over call
+	my $xdir_ro = $ibx->{search}->xdir(1);
+	my $npart = 0;
+	foreach my $part (<$xdir_ro/*>) {
+		if (-d $part && $part =~ m!/\d+\z!) {
+			my $bytes = 0;
+			$bytes += -s $_ foreach glob("$part/*");
+			$npart++ if $bytes;
+		}
+	}
+	if ($npart) {
+		PublicInbox::Admin::require_or_die('-search');
+	} else {
+		# somebody could "rm -r" all the Xapian directories;
+		# let them purge the overview, at least
+		$ibx->{indexlevel} ||= 'basic';
+	}
+}
+
 my $data = do { local $/; scalar <STDIN> };
 $data =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
 my $n_purged = 0;
-- 
EW


^ permalink raw reply	[relevance 5%]

* [PATCH 0/6] make Search::Xapian optional for v2, NNTP, etc
@ 2019-05-15  6:33  7% Eric Wong
  2019-05-15  6:33  5% ` [PATCH 6/6] admin: improve warnings and errors for missing modules Eric Wong
  0 siblings, 1 reply; 2+ results
From: Eric Wong @ 2019-05-15  6:33 UTC (permalink / raw)
  To: meta

Search::Xapian is missing from CentOS 7 and probably other
distros; so make it optional as all of our NNTP and a large
chunk of our WWW functionality doesn't rely on it, anymore.

Xapian databases is also the most expensive in terms of space
for indexing overhead, so indexlevel=basic users may appreciate
not having to install Xapian at all.

This has passed all test profiles (ci/profiles.sh) on Debian 9
and FreeBSD 11.2 (x86-64); but I haven't dogfooded it enough
to be comfortable having it in 'master'...  and I might not be
online much the next few days in case something goes sideways
on public-inbox.org.

The following changes since commit 2d51a2accd28465c915f91ab48c7e52a81d87a68:

  tests: get rid of unnecessary Cwd module use (2019-05-14 04:00:10 +0000)

are available in the Git repository at:

  https://80x24.org/public-inbox.git xap-optional

for you to fetch changes up to d0e8bfd866ed1e924e8d9f551939eecbea4920ef:

  admin: improve warnings and errors for missing modules (2019-05-15 06:15:34 +0000)

----------------------------------------------------------------
Eric Wong (6):
      inbox: add ->over method to ease access
      nntp: use Inbox->over directly
      www: use Inbox->over where appropriate
      lazy load Xapian and make it optional for v2
      searchidx: do not create empty Xapian partitions for basic
      admin: improve warnings and errors for missing modules

 INSTALL                       |   2 +-
 TODO                          |   2 -
 ci/deps.perl                  |   7 +--
 ci/profiles.sh                |   3 +-
 lib/PublicInbox/Admin.pm      |  60 ++++++++++++++++++++
 lib/PublicInbox/Feed.pm       |  11 ++--
 lib/PublicInbox/Inbox.pm      |  40 +++++++++-----
 lib/PublicInbox/Mbox.pm       |  25 +++++----
 lib/PublicInbox/NNTP.pm       |  40 +++++++-------
 lib/PublicInbox/OverIdx.pm    |  13 ++++-
 lib/PublicInbox/Search.pm     | 125 +++++++++++++++++-------------------------
 lib/PublicInbox/SearchIdx.pm  |  57 +++++++++++++------
 lib/PublicInbox/SearchMsg.pm  |  14 +++++
 lib/PublicInbox/SearchView.pm |  13 +++--
 lib/PublicInbox/View.pm       |  53 ++++++++----------
 lib/PublicInbox/WWW.pm        |  39 +++++--------
 script/public-inbox-index     |  43 ++++++++++-----
 script/public-inbox-init      |   8 ++-
 script/public-inbox-purge     |  48 +++++++++++++---
 t/admin.t                     |   2 +-
 t/cgi.t                       |  16 ++++--
 t/indexlevels-mirror.t        |  22 ++++++--
 t/nntp.t                      |   2 +-
 t/nntpd.t                     |  15 +++--
 t/over.t                      |   3 +-
 t/psgi_bad_mids.t             |   3 +-
 t/psgi_scan_all.t             |   4 +-
 t/psgi_search.t               |   4 +-
 t/purge.t                     |   2 +-
 t/search-thr-index.t          |   5 +-
 t/search.t                    |  11 ++--
 t/v1reindex.t                 |   3 +-
 t/v2reindex.t                 |   5 +-
 t/v2writable.t                |   2 +-
 t/view.t                      |   1 +
 t/watch_maildir_v2.t          |   4 +-
 36 files changed, 432 insertions(+), 275 deletions(-)

^ permalink raw reply	[relevance 7%]

Results 1-2 of 2 | reverse | sort options + mbox downloads above
-- links below jump to the message on this page --
2019-05-15  6:33  7% [PATCH 0/6] make Search::Xapian optional for v2, NNTP, etc Eric Wong
2019-05-15  6:33  5% ` [PATCH 6/6] admin: improve warnings and errors for missing modules Eric Wong

Code repositories for project(s) associated with this inbox:

	https://80x24.org/public-inbox.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).