-# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
#
# Represents a public-inbox (which may have multiple mailing addresses)
package PublicInbox::Inbox;
use strict;
+use v5.10.1;
use PublicInbox::Git;
use PublicInbox::MID qw(mid2path);
use PublicInbox::Eml;
use List::Util qw(max);
+use Carp qw(croak);
-# Long-running "git-cat-file --batch" processes won't notice
-# unlinked packs, so we need to restart those processes occasionally.
-# Xapian and SQLite file handles are mostly stable, but sometimes an
-# admin will attempt to replace them atomically after compact/vacuum
-# and we need to be prepared for that.
-my $cleanup_timer;
-my $cleanup_avail = -1; # 0, or 1
-my $have_devel_peek;
-my $CLEANUP = {}; # string(inbox) -> inbox
-
-sub git_cleanup ($) {
- my ($self) = @_;
- my $git = $self->{git} or return;
- $git->cleanup;
-}
-
-sub cleanup_task () {
- $cleanup_timer = undef;
- my $next = {};
- for my $ibx (values %$CLEANUP) {
- my $again;
- if ($have_devel_peek) {
- foreach my $f (qw(search)) {
- # we bump refcnt by assigning tmp, here:
- my $tmp = $ibx->{$f} or next;
- next if Devel::Peek::SvREFCNT($tmp) > 2;
- delete $ibx->{$f};
- # refcnt is zero when tmp is out-of-scope
- }
- }
- git_cleanup($ibx);
- if (my $gits = $ibx->{-repo_objs}) {
- foreach my $git (@$gits) {
- $again = 1 if $git->cleanup;
- }
- }
+# returns true if further checking is required
+sub check_inodes ($) {
+ for (qw(over mm)) { $_[0]->{$_}->check_inodes if $_[0]->{$_} }
+}
+
+sub do_cleanup {
+ my ($ibx) = @_;
+ my $live;
+ if (defined $ibx->{git}) {
+ $live = $ibx->isa(__PACKAGE__) ? $ibx->{git}->cleanup(1)
+ : $ibx->{git}->cleanup_if_unlinked;
+ delete($ibx->{git}) unless $live;
+ }
+ if ($live) {
check_inodes($ibx);
- if ($have_devel_peek) {
- $again ||= !!$ibx->{search};
- }
- $next->{"$ibx"} = $ibx if $again;
+ } else {
+ delete(@$ibx{qw(over mm description cloneurl
+ -imap_url -nntp_url)});
}
- $CLEANUP = $next;
+ my $srch = $ibx->{search} // $ibx;
+ delete @$srch{qw(xdb qp)};
+ for my $git (@{$ibx->{-repo_objs} // []}) {
+ $live = 1 if $git->cleanup(1);
+ }
+ PublicInbox::DS::add_uniq_timer($ibx+0, 5, \&do_cleanup, $ibx) if $live;
}
-sub cleanup_possible () {
+sub _cleanup_later ($) {
# no need to require DS, here, if it were enabled another
# module would've require'd it, already
- eval { PublicInbox::DS::in_loop() } or return 0;
-
- eval {
- require Devel::Peek; # needs separate package in Fedora
- $have_devel_peek = 1;
- };
- 1;
-}
-
-sub _cleanup_later ($) {
- my ($self) = @_;
- $cleanup_avail = cleanup_possible() if $cleanup_avail < 0;
- return if $cleanup_avail != 1;
- $cleanup_timer //= PublicInbox::DS::later(\&cleanup_task);
- $CLEANUP->{"$self"} = $self;
+ eval { PublicInbox::DS::in_loop() } and
+ PublicInbox::DS::add_uniq_timer($_[0]+0, 30, \&do_cleanup, @_)
}
sub _set_limiter ($$$) {
my ($self, $pi_cfg, $pfx) = @_;
my $lkey = "-${pfx}_limiter";
- $self->{$lkey} ||= do {
+ $self->{$lkey} //= do {
# full key is: publicinbox.$NAME.httpbackendmax
my $mkey = $pfx.'max';
my $val = $self->{$mkey} or return;
} else {
delete $opts->{feedmax};
}
- $opts->{nntpserver} ||= $pi_cfg->{'publicinbox.nntpserver'};
-
# allow any combination of multi-line or comma-delimited hide entries
my $hide = {};
if (defined(my $h = $opts->{hide})) {
sub git_epoch {
my ($self, $epoch) = @_; # v2-only, callers always supply $epoch
- $self->{"$epoch.git"} ||= do {
+ $self->{"$epoch.git"} //= do {
my $git_dir = "$self->{inboxdir}/git/$epoch.git";
return unless -d $git_dir;
my $g = PublicInbox::Git->new($git_dir);
- $g->{-httpbackend_limiter} = $self->{-httpbackend_limiter};
+ my $lim = $self->{-httpbackend_limiter};
+ $g->{-httpbackend_limiter} = $lim if $lim;
# caller must manually cleanup when done
$g;
};
sub git {
my ($self) = @_;
- $self->{git} ||= do {
+ $self->{git} //= do {
my $git_dir = $self->{inboxdir};
$git_dir .= '/all.git' if $self->version == 2;
my $g = PublicInbox::Git->new($git_dir);
- $g->{-httpbackend_limiter} = $self->{-httpbackend_limiter};
+ my $lim = $self->{-httpbackend_limiter};
+ $g->{-httpbackend_limiter} = $lim if $lim;
_cleanup_later($self);
$g;
};
my $cur = $self->{-max_git_epoch};
my $changed;
if (!defined($cur) || ($changed = git($self)->alternates_changed)) {
- git_cleanup($self) if $changed;
+ $self->{git}->cleanup if $changed;
my $gits = "$self->{inboxdir}/git";
if (opendir my $dh, $gits) {
my $max = max(map {
$cur;
}
-sub mm {
+sub mm_file {
my ($self) = @_;
- $self->{mm} ||= eval {
+ my $d = $self->{inboxdir};
+ ($self->version >= 2 ? $d : "$d/public-inbox").'/msgmap.sqlite3';
+}
+
+sub mm {
+ my ($self, $req) = @_;
+ $self->{mm} //= eval {
require PublicInbox::Msgmap;
- my $dir = $self->{inboxdir};
- if ($self->version >= 2) {
- PublicInbox::Msgmap->new_file("$dir/msgmap.sqlite3");
- } else {
- PublicInbox::Msgmap->new($dir);
- }
- };
+ _cleanup_later($self);
+ PublicInbox::Msgmap->new_file($self);
+ } // ($req ? croak("E: $@") : undef);
}
sub search {
my ($self) = @_;
- my $srch = $self->{search} //= eval {
+ $self->{search} // eval {
_cleanup_later($self);
require PublicInbox::Search;
- PublicInbox::Search->new($self);
+ my $srch = PublicInbox::Search->new($self);
+ (eval { $srch->xdb }) ? ($self->{search} = $srch) : undef;
};
- (eval { $srch->xdb }) ? $srch : undef;
}
# isrch is preferred for read-only interfaces if available since it
sub isrch { $_[0]->{isrch} // search($_[0]) }
sub over {
- $_[0]->{over} //= eval {
- my $srch = $_[0]->{search} //= eval {
- _cleanup_later($_[0]);
+ my ($self, $req) = @_;
+ $self->{over} // eval {
+ my $srch = $self->{search} // do {
require PublicInbox::Search;
- PublicInbox::Search->new($_[0]);
+ PublicInbox::Search->new($self);
};
+ _cleanup_later($self);
my $over = PublicInbox::Over->new("$srch->{xpfx}/over.sqlite3");
$over->dbh; # may fail
- $over;
- };
+ $self->{over} = $over;
+ } // ($req ? croak("E: $@") : undef);
}
-
sub try_cat {
my ($path) = @_;
open(my $fh, '<', $path) or return '';
sub cloneurl {
my ($self) = @_;
- ($self->{cloneurl} //= do {
+ $self->{cloneurl} // do {
my $s = try_cat("$self->{inboxdir}/cloneurl");
my @urls = split(/\s+/s, $s);
- scalar(@urls) ? \@urls : undef
- }) // [];
+ scalar(@urls) ? ($self->{cloneurl} = \@urls) : undef;
+ } // [];
}
sub base_url {
my ($self, $env) = @_; # env - PSGI env
- if ($env) {
+ if ($env && $env->{'psgi.url_scheme'}) {
my $url = PublicInbox::Git::host_prefix_url($env, '');
# for mount in Plack::Builder
$url .= '/' if $url !~ m!/\z!;
return $url .= $self->{name} . '/';
}
# called from a non-PSGI environment (e.g. NNTP/POP3):
- $self->{-base_url} ||= do {
- my $url = $self->{url}->[0] or return undef;
- # expand protocol-relative URLs to HTTPS if we're
- # not inside a web server
- $url = "https:$url" if $url =~ m!\A//!;
- $url .= '/' if $url !~ m!/\z!;
- $url;
- };
-}
-
-sub nntp_url {
- my ($self) = @_;
- $self->{-nntp_url} ||= do {
- # no checking for nntp_usable here, we can point entirely
- # to non-local servers or users run by a different user
- my $ns = $self->{nntpserver};
- my $group = $self->{newsgroup};
- my @urls;
- if ($ns && $group) {
- $ns = [ $ns ] if ref($ns) ne 'ARRAY';
- @urls = map {
- my $u = m!\Anntps?://! ? $_ : "nntp://$_";
- $u .= '/' if $u !~ m!/\z!;
- $u.$group;
- } @$ns;
- }
-
- my $mirrors = $self->{nntpmirror};
- if ($mirrors) {
- my @m;
- foreach (@$mirrors) {
- my $u = m!\Anntps?://! ? $_ : "nntp://$_";
- if ($u =~ m!\Anntps?://[^/]+/?\z!) {
- if ($group) {
- $u .= '/' if $u !~ m!/\z!;
- $u .= $group;
- } else {
- warn
-"publicinbox.$self->{name}.nntpmirror=$_ missing newsgroup name\n";
- }
+ my $url = $self->{url} // return undef;
+ $url = $url->[0] // return undef;
+ # expand protocol-relative URLs to HTTPS if we're
+ # not inside a web server
+ substr($url, 0, 0, 'https:') if substr($url, 0, 2) eq '//';
+ $url .= '/' if substr($url, -1, 1) ne '/';
+ $url;
+}
+
+# imapserver, nntpserver, and pop3server configs are used here:
+sub _x_url ($$$) {
+ my ($self, $x, $ctx) = @_; # $x is "imap", "nntp", or "pop3"
+ # no checking for nntp_usable here, we can point entirely
+ # to non-local servers or users run by a different user
+ my $ns = $self->{"${x}server"} //
+ $ctx->{www}->{pi_cfg}->get_all("publicinbox.${x}server");
+ my $group = $self->{newsgroup};
+ my @urls;
+ if ($ns && $group) {
+ @urls = map {
+ my $u = m!\A${x}s?://! ? $_ : "$x://$_";
+ $u .= '/' if $u !~ m!/\z!;
+ $u.$group;
+ } @$ns;
+ }
+ if (my $mirrors = $self->{"${x}mirror"}) {
+ my @m;
+ for (@$mirrors) {
+ my $u = m!\A${x}s?://! ? $_ : "$x://$_";
+ if ($u =~ m!\A${x}s?://[^/]+/?\z!) {
+ if ($group) {
+ $u .= '/' if $u !~ m!/\z!;
+ $u .= $group;
+ } else { # n.b. IMAP and POP3 use "newsgroup"
+ warn <<EOM;
+publicinbox.$self->{name}.${x}mirror=$_ missing newsgroup name
+EOM
}
- # else: allow full URLs like:
- # nntp://news.example.com/alt.example
- push @m, $u;
}
-
- # List::Util::uniq requires Perl 5.26+, maybe we
- # can use it by 2030 or so
- my %seen;
- @urls = grep { !$seen{$_}++ } (@urls, @m);
+ # else: allow full URLs like:
+ # nntp://news.example.com/alt.example
+ push @m, $u;
}
- \@urls;
- };
+
+ # List::Util::uniq requires Perl 5.26+, maybe we
+ # can use it by 2030 or so
+ my %seen;
+ @urls = grep { !$seen{$_}++ } (@urls, @m);
+ }
+ \@urls;
}
+# my ($self, $ctx) = @_;
+sub imap_url { $_[0]->{-imap_url} //= _x_url($_[0], 'imap', $_[1]) }
+sub nntp_url { $_[0]->{-nntp_url} //= _x_url($_[0], 'nntp', $_[1]) }
+sub pop3_url { $_[0]->{-pop3_url} //= _x_url($_[0], 'pop3', $_[1]) }
+
sub nntp_usable {
my ($self) = @_;
my $ret = mm($self) && over($self);
- $self->{mm} = $self->{over} = $self->{search} = undef;
+ delete @$self{qw(mm over search)};
$ret;
}
# ghosts may have undef smsg (from SearchThread.node) or
# no {blob} field
- return unless defined $smsg;
- defined(my $blob = $smsg->{blob}) or return;
-
- $self->git->cat_file($blob);
+ $smsg // return;
+ $self->git->cat_file($smsg->{blob} // return);
}
sub smsg_eml {
my ($self, $smsg) = @_;
my $bref = msg_by_smsg($self, $smsg) or return;
my $eml = PublicInbox::Eml->new($bref);
- $smsg->populate($eml) unless exists($smsg->{num}); # v1 w/o SQLite
+ $smsg->{num} // $smsg->populate($eml);
$eml;
}
my $smsg;
if (my $mm = $self->mm) {
# favor the Message-ID we used for the NNTP article number:
- defined(my $num = $mm->num_for($mid)) or return;
+ my $num = $mm->num_for($mid) // return;
$smsg = $over->get_art($num);
} else {
my ($id, $prev);
# (pathname is NOT public, but prefix is used for Xapian queries)
sub altid_map ($) {
my ($self) = @_;
- $self->{-altid_map} //= eval {
+ eval {
require PublicInbox::AltId;
my $altid = $self->{altid} or return {};
my %h = map {;
delete $self->{unlock_subs}->{$ident};
}
-sub check_inodes ($) {
- my ($self) = @_;
- for (qw(over mm)) { # TODO: search
- $self->{$_}->check_inodes if $self->{$_};
- }
-}
-
# called by inotify
sub on_unlock {
my ($self) = @_;
sub eidx_key { $_[0]->{newsgroup} // $_[0]->{inboxdir} }
+sub mailboxid { # rfc 8474, 8620, 8621
+ my ($self, $imap_slice) = @_;
+ my $pfx = defined($imap_slice) ? $self->{newsgroup} : $self->{name};
+ utf8::encode($pfx); # to octets
+ # RFC 8620, 1.2 recommends not starting with dash or digits
+ # "A good solution to these issues is to prefix every id with a single
+ # alphabetical character."
+ 'M'.join('', map { sprintf('%02x', ord) } split(//, $pfx)) .
+ (defined($imap_slice) ? sprintf('-%x', $imap_slice) : '') .
+ sprintf('-%x', uidvalidity($self) // 0)
+}
+
1;