X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FConfig.pm;h=f2e5d5f0801e4c0f566f51ae654bdc55a5112c38;hb=1d6ebab439828dce8b7e8cbb99dfbc237e9d5a74;hp=8ecf549d47a42b8bc07c859389effb4adef3722d;hpb=4bffd23fdd592c356c9b3d56b202d178a12fec43;p=public-inbox.git diff --git a/lib/PublicInbox/Config.pm b/lib/PublicInbox/Config.pm index 8ecf549d..f2e5d5f0 100644 --- a/lib/PublicInbox/Config.pm +++ b/lib/PublicInbox/Config.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2014-2019 all contributors +# Copyright (C) 2014-2021 all contributors # License: AGPL-3.0+ # # Used throughout the project for reading configuration @@ -9,30 +9,33 @@ package PublicInbox::Config; use strict; -use warnings; -require PublicInbox::Inbox; +use v5.10.1; +use PublicInbox::Inbox; use PublicInbox::Spawn qw(popen_rd); +our $LD_PRELOAD = $ENV{LD_PRELOAD}; # only valid at startup sub _array ($) { ref($_[0]) eq 'ARRAY' ? $_[0] : [ $_[0] ] } # returns key-value pairs of config directives in a hash # if keys may be multi-value, the value is an array ref containing all values sub new { - my ($class, $file) = @_; - $file = default_file() unless defined($file); + my ($class, $file, $errfh) = @_; + $file //= default_file(); my $self; if (ref($file) eq 'SCALAR') { # used by some tests open my $fh, '<', $file or die; # PerlIO::scalar $self = config_fh_parse($fh, "\n", '='); + bless $self, $class; } else { - $self = git_config_dump($file); + $self = git_config_dump($class, $file, $errfh); + $self->{'-f'} = $file; } - bless $self, $class; # caches $self->{-by_addr} = {}; $self->{-by_list_id} = {}; $self->{-by_name} = {}; $self->{-by_newsgroup} = {}; + $self->{-by_eidx_key} = {}; $self->{-no_obfuscate} = {}; $self->{-limiters} = {}; $self->{-code_repos} = {}; # nick => PublicInbox::Git object @@ -86,16 +89,29 @@ sub lookup_list_id { sub lookup_name ($$) { my ($self, $name) = @_; - $self->{-by_name}->{$name} // _fill($self, "publicinbox.$name"); + $self->{-by_name}->{$name} // _fill_ibx($self, $name); } +sub lookup_ei { + my ($self, $name) = @_; + $self->{-ei_by_name}->{$name} //= _fill_ei($self, $name); +} + +sub lookup_eidx_key { + my ($self, $eidx_key) = @_; + _lookup_fill($self, '-by_eidx_key', $eidx_key); +} + +# special case for [extindex "all"] +sub ALL { lookup_ei($_[0], 'all') } + sub each_inbox { - my ($self, $cb, $arg) = @_; + my ($self, $cb, @arg) = @_; # may auto-vivify if config file is non-existent: foreach my $section (@{$self->{-section_order}}) { next if $section !~ m!\Apublicinbox\.([^/]+)\z!; my $ibx = lookup_name($self, $1) or next; - $cb->($ibx, $arg); + $cb->($ibx, @arg); } } @@ -118,27 +134,21 @@ sub limiter { sub config_dir { $ENV{PI_DIR} // "$ENV{HOME}/.public-inbox" } sub default_file { - my $f = $ENV{PI_CONFIG}; - return $f if defined $f; - config_dir() . '/config'; + $ENV{PI_CONFIG} // (config_dir() . '/config'); } sub config_fh_parse ($$$) { my ($fh, $rs, $fs) = @_; - my %rv; - my (%section_seen, @section_order); + my (%rv, %seen, @section_order, $line, $k, $v, $section, $cur, $i); local $/ = $rs; - while (defined(my $line = <$fh>)) { - chomp $line; - my ($k, $v) = split($fs, $line, 2); - my ($section) = ($k =~ /\A(\S+)\.[^\.]+\z/); - unless (defined $section_seen{$section}) { - $section_seen{$section} = 1; - push @section_order, $section; - } - - my $cur = $rv{$k}; - if (defined $cur) { + while (defined($line = <$fh>)) { # perf critical with giant configs + $i = index($line, $fs); + $k = substr($line, 0, $i); + $v = substr($line, $i + 1, -1); # chop off $fs + $section = substr($k, 0, rindex($k, '.')); + $seen{$section} //= push(@section_order, $section); + + if (defined($cur = $rv{$k})) { if (ref($cur) eq "ARRAY") { push @$cur, $v; } else { @@ -154,18 +164,17 @@ sub config_fh_parse ($$$) { } sub git_config_dump { - my ($file) = @_; - return {} unless -e $file; - my @cmd = (qw/git config -z -l/, "--file=$file"); - my $cmd = join(' ', @cmd); - my $fh = popen_rd(\@cmd) or die "popen_rd failed for $file: $!\n"; + my ($class, $file, $errfh) = @_; + return bless {}, $class unless -e $file; + my $cmd = [ qw(git config -z -l --includes), "--file=$file" ]; + my $fh = popen_rd($cmd, undef, { 2 => $errfh // 2 }); my $rv = config_fh_parse($fh, "\0", "\n"); - close $fh or die "failed to close ($cmd) pipe: $?"; - $rv; + close $fh or die "@$cmd failed: \$?=$?\n"; + bless $rv, $class; } -sub valid_inbox_name ($) { - my ($name) = @_; +sub valid_foo_name ($;$) { + my ($name, $pfx) = @_; # Similar rules found in git.git/remote.c::valid_remote_nick # and git.git/refs.c::check_refname_component @@ -173,6 +182,7 @@ sub valid_inbox_name ($) { if ($name eq '' || $name =~ /\@\{/ || $name =~ /\.\./ || $name =~ m![/:\?\[\]\^~\s\f[:cntrl:]\*]! || $name =~ /\A\./ || $name =~ /\.\z/) { + warn "invalid $pfx name: `$name'\n" if $pfx; return 0; } @@ -190,6 +200,9 @@ sub cgit_repo_merge ($$$) { return unless -e "$path/$se"; } return if -e "$path/noweb"; + # this comes from the cgit config, and AFAIK cgit only allows + # repos to have one URL, but that's just the PATH_INFO component, + # not the Host: portion # $repo = { url => 'foo.git', dir => '/path/to/foo.git' } my $rel = $repo->{url}; unless (defined $rel) { @@ -207,7 +220,7 @@ sub cgit_repo_merge ($$$) { $rel =~ s!/?\.git\z!!; } $self->{"coderepo.$rel.dir"} //= $path; - $self->{"coderepo.$rel.cgiturl"} //= $rel; + $self->{"coderepo.$rel.cgiturl"} //= _array($rel); } sub is_git_dir ($) { @@ -249,7 +262,7 @@ sub scan_projects_coderepo ($$$) { warn "failed to open cgit projectlist=$list: $!\n"; return; }; - foreach (<$fh>) { + while (<$fh>) { chomp; scan_path_coderepo($self, $path, "$path/$_"); } @@ -274,7 +287,7 @@ sub parse_cgitrc { # FIXME: this doesn't support macro expansion via $VARS, yet my $repo; - foreach (<$fh>) { + while (<$fh>) { chomp; if (m!\Arepo\.url=(.+?)/*\z!) { my $nick = $1; @@ -315,40 +328,27 @@ sub parse_cgitrc { sub _fill_code_repo { my ($self, $nick) = @_; my $pfx = "coderepo.$nick"; - - # TODO: support gitweb and other repository viewers? - if (defined(my $cgitrc = delete $self->{-cgitrc_unparsed})) { - parse_cgitrc($self, $cgitrc, 0); - } - my $dir = $self->{"$pfx.dir"}; # aka "GIT_DIR" - unless (defined $dir) { + my $dir = $self->{"$pfx.dir"} // do { # aka "GIT_DIR" warn "$pfx.dir unset\n"; return; - } - + }; my $git = PublicInbox::Git->new($dir); - foreach my $t (qw(blob commit tree tag)) { - $git->{$t.'_url_format'} = - _array($self->{lc("$pfx.${t}UrlFormat")}); - } - - if (my $cgits = $self->{lc("$pfx.cgitUrl")}) { + if (defined(my $cgits = $self->{"$pfx.cgiturl"})) { $git->{cgit_url} = $cgits = _array($cgits); - - # cgit supports "/blob/?id=%s", but it's only a plain-text - # display and requires an unabbreviated id= - foreach my $t (qw(blob commit tag)) { - $git->{$t.'_url_format'} //= map { - "$_/$t/?id=%s" - } @$cgits; - } + $self->{"$pfx.cgiturl"} = $cgits; } $git; } -sub _git_config_bool ($) { - my ($val) = @_; +sub get_all { + my ($self, $key) = @_; + my $v = $self->{$key} // return; + _array($v); +} + +sub git_bool { + my ($val) = $_[-1]; # $_[0] may be $self, or $val if ($val =~ /\A(?:false|no|off|[\-\+]?(?:0x)?0+)\z/i) { 0; } elsif ($val =~ /\A(?:true|yes|on|[\-\+]?(?:0x)?[0-9]+)\z/i) { @@ -358,24 +358,75 @@ sub _git_config_bool ($) { } } -sub _fill { - my ($self, $pfx) = @_; - my $ibx = {}; +# abs_path resolves symlinks, so we want to avoid it if rel2abs +# is sufficient and doesn't leave "/.." or "/../" +sub rel2abs_collapsed { + require File::Spec; + my $p = File::Spec->rel2abs($_[-1]); + return $p if substr($p, -3, 3) ne '/..' && index($p, '/../') < 0; + require Cwd; + Cwd::abs_path($p); +} + +sub get_1 { + my ($self, $key) = @_; + my $v = $self->{$key}; + return $v if !ref($v); + warn "W: $key has multiple values, only using `$v->[-1]'\n"; + $v->[-1]; +} + +sub repo_objs { + my ($self, $ibxish) = @_; + my $ibx_code_repos = $ibxish->{coderepo} or return; + $ibxish->{-repo_objs} //= do { + my $code_repos = $self->{-code_repos}; + my @repo_objs; + for my $nick (@$ibx_code_repos) { + my @parts = split(m!/!, $nick); + for (@parts) { + @parts = () unless valid_foo_name($_); + } + unless (@parts) { + warn "invalid coderepo name: `$nick'\n"; + next; + } + my $repo = $code_repos->{$nick} //= + _fill_code_repo($self, $nick); + push @repo_objs, $repo if $repo; + } + if (scalar @repo_objs) { + \@repo_objs; + } else { + delete $ibxish->{coderepo}; + undef; + } + } +} - foreach my $k (qw(inboxdir filter url newsgroup - infourl watch watchheader httpbackendmax - replyto feedmax nntpserver indexlevel)) { +sub _fill_ibx { + my ($self, $name) = @_; + my $pfx = "publicinbox.$name"; + my $ibx = {}; + for my $k (qw(watch)) { my $v = $self->{"$pfx.$k"}; $ibx->{$k} = $v if defined $v; } + for my $k (qw(filter inboxdir newsgroup replyto httpbackendmax feedmax + indexlevel indexsequentialshard boost)) { + my $v = get_1($self, "$pfx.$k") // next; + $ibx->{$k} = $v; + } - # backwards compatibility: - $ibx->{inboxdir} //= $self->{"$pfx.mainrepo"}; - - foreach my $k (qw(obfuscate)) { - my $v = $self->{"$pfx.$k"}; - defined $v or next; - if (defined(my $bval = _git_config_bool($v))) { + # "mainrepo" is backwards compatibility: + my $dir = $ibx->{inboxdir} //= $self->{"$pfx.mainrepo"} // return; + if (index($dir, "\n") >= 0) { + warn "E: `$dir' must not contain `\\n'\n"; + return; + } + for my $k (qw(obfuscate)) { + my $v = $self->{"$pfx.$k"} // next; + if (defined(my $bval = git_bool($v))) { $ibx->{$k} = $bval; } else { warn "Ignoring $pfx.$k=$v in config, not boolean\n"; @@ -383,23 +434,16 @@ sub _fill { } # TODO: more arrays, we should support multi-value for # more things to encourage decentralization - foreach my $k (qw(address altid nntpmirror coderepo hide listid)) { - if (defined(my $v = $self->{"$pfx.$k"})) { - $ibx->{$k} = _array($v); - } - } - - return unless defined($ibx->{inboxdir}); - my $name = $pfx; - $name =~ s/\Apublicinbox\.//; - - if (!valid_inbox_name($name)) { - warn "invalid inbox name: '$name'\n"; - return; + for my $k (qw(address altid nntpmirror imapmirror + coderepo hide listid url + infourl watchheader nntpserver imapserver)) { + my $v = $self->{"$pfx.$k"} // next; + $ibx->{$k} = _array($v); } + return unless valid_foo_name($name, 'publicinbox'); $ibx->{name} = $name; - $ibx->{-pi_config} = $self; + $ibx->{-pi_cfg} = $self; $ibx = PublicInbox::Inbox->new($ibx); foreach (@{$ibx->{address}}) { my $lc_addr = lc($_); @@ -407,12 +451,36 @@ sub _fill { $self->{-no_obfuscate}->{$lc_addr} = 1; } if (my $listids = $ibx->{listid}) { + # RFC2919 section 6 stipulates "case insensitive equality" foreach my $list_id (@$listids) { - $self->{-by_list_id}->{$list_id} = $ibx; + $self->{-by_list_id}->{lc($list_id)} = $ibx; } } - if (my $ng = $ibx->{newsgroup}) { - $self->{-by_newsgroup}->{$ng} = $ibx; + if (defined(my $ngname = $ibx->{newsgroup})) { + if (ref($ngname)) { + delete $ibx->{newsgroup}; + warn 'multiple newsgroups not supported: '. + join(', ', @$ngname). "\n"; + # Newsgroup name needs to be compatible with RFC 3977 + # wildmat-exact and RFC 3501 (IMAP) ATOM-CHAR. + # Leave out a few chars likely to cause problems or conflicts: + # '|', '<', '>', ';', '#', '$', '&', + } elsif ($ngname =~ m![^A-Za-z0-9/_\.\-\~\@\+\=:]! || + $ngname eq '') { + delete $ibx->{newsgroup}; + warn "newsgroup name invalid: `$ngname'\n"; + } else { + # PublicInbox::NNTPD does stricter ->nntp_usable + # checks, keep this lean for startup speed + $self->{-by_newsgroup}->{$ngname} = $ibx; + } + } + unless (defined $ibx->{newsgroup}) { # for ->eidx_key + my $abs = rel2abs_collapsed($dir); + if ($abs ne $dir) { + warn "W: `$dir' canonicalized to `$abs'\n"; + $ibx->{inboxdir} = $abs; + } } $self->{-by_name}->{$name} = $ibx; if ($ibx->{obfuscate}) { @@ -420,23 +488,75 @@ sub _fill { $ibx->{-no_obfuscate_re} = $self->{-no_obfuscate_re}; fill_all($self); # noop to populate -no_obfuscate } + if (my $es = ALL($self)) { + require PublicInbox::Isearch; + $ibx->{isrch} = PublicInbox::Isearch->new($ibx, $es); + } + $self->{-by_eidx_key}->{$ibx->eidx_key} = $ibx; +} - if (my $ibx_code_repos = $ibx->{coderepo}) { - my $code_repos = $self->{-code_repos}; - my $repo_objs = $ibx->{-repo_objs} = []; - foreach my $nick (@$ibx_code_repos) { - my @parts = split(m!/!, $nick); - my $valid = 0; - $valid += valid_inbox_name($_) foreach (@parts); - $valid == scalar(@parts) or next; +sub _fill_ei ($$) { + my ($self, $name) = @_; + eval { require PublicInbox::ExtSearch } or return; + my $pfx = "extindex.$name"; + my $d = $self->{"$pfx.topdir"} // return; + -d $d or return; + if (index($d, "\n") >= 0) { + warn "E: `$d' must not contain `\\n'\n"; + return; + } + my $es = PublicInbox::ExtSearch->new($d); + for my $k (qw(indexlevel indexsequentialshard)) { + my $v = get_1($self, "$pfx.$k") // next; + $es->{$k} = $v; + } + for my $k (qw(coderepo hide url infourl)) { + my $v = $self->{"$pfx.$k"} // next; + $es->{$k} = _array($v); + } + return unless valid_foo_name($name, 'extindex'); + $es->{name} = $name; + $es; +} - my $repo = $code_repos->{$nick} //= - _fill_code_repo($self, $nick); - push @$repo_objs, $repo if $repo; - } +sub urlmatch { + my ($self, $key, $url) = @_; + state $urlmatch_broken; # requires git 1.8.5 + return if $urlmatch_broken; + my $file = $self->{'-f'} // default_file(); + my $cmd = [qw/git config -z --includes --get-urlmatch/, + "--file=$file", $key, $url ]; + my $fh = popen_rd($cmd); + local $/ = "\0"; + my $val = <$fh>; + if (close($fh)) { + chomp($val); + $val; + } else { + $urlmatch_broken = 1 if (($? >> 8) != 1); + undef; } +} - $ibx +sub json { + state $json; + $json //= do { + for my $mod (qw(Cpanel::JSON::XS JSON::MaybeXS JSON JSON::PP)) { + eval "require $mod" or next; + # ->ascii encodes non-ASCII to "\uXXXX" + $json = $mod->new->ascii(1) and last; + } + $json; + }; +} + +sub squote_maybe ($) { + my ($val) = @_; + if ($val =~ m{([^\w@\./,\%\+\-])}) { + $val =~ s/(['!])/'\\$1'/g; # '!' for csh + return "'$val'"; + } + $val; } 1;