X-Git-Url: https://git.exim.org/exim.git/blobdiff_plain/11121d3d3850fe51de5fbd23acbc0bb87fa082d6..cb2243935be366d4605e94a1e55c1dcc933b66cd:/src/src/exipick.src diff --git a/src/src/exipick.src b/src/src/exipick.src index e3967c382..4708ebb4a 100644 --- a/src/src/exipick.src +++ b/src/src/exipick.src @@ -1,17 +1,22 @@ #!PERL_COMMAND -# $Cambridge: exim/src/src/exipick.src,v 1.9 2006/02/16 17:03:16 jetmore Exp $ # This variable should be set by the building process to Exim's spool directory. my $spool = 'SPOOL_DIRECTORY'; +# Need to set this dynamically during build, but it's not used right now anyway. +my $charset = 'ISO-8859-1'; + +# use 'exipick --help' to view documentation for this program. +# Documentation also viewable online at +# http://www.exim.org/eximwiki/ToolExipickManPage use strict; use Getopt::Long; my($p_name) = $0 =~ m|/?([^/]+)$|; -my $p_version = "20060216.1"; +my $p_version = "20100323.0"; my $p_usage = "Usage: $p_name [--help|--version] (see --help for details)"; my $p_cp = < + Copyright (c) 2003-2010 John Jetmore This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -25,7 +30,7 @@ my $p_cp = < \$G::spool, # exim spool dir + 'spool=s' => \$G::spool, # exim spool dir + 'input-dir=s' => \$G::input_dir, # name of the "input" dir + 'finput' => \$G::finput, # same as "--input-dir Finput" 'bp' => \$G::mailq_bp, # List the queue (noop - default) 'bpa' => \$G::mailq_bpa, # ... with generated address as well 'bpc' => \$G::mailq_bpc, # ... but just show a count of messages @@ -43,25 +50,54 @@ GetOptions( 'bpu' => \$G::mailq_bpu, # ... only undelivered addresses 'and' => \$G::and, # 'and' the criteria (default) 'or' => \$G::or, # 'or' the criteria - 'f:s' => \$G::qgrep_f, # from regexp - 'r:s' => \$G::qgrep_r, # recipient regexp - 's:s' => \$G::qgrep_s, # match against size field - 'y:s' => \$G::qgrep_y, # message younger than (secs) - 'o:s' => \$G::qgrep_o, # message older than (secs) + 'f=s' => \$G::qgrep_f, # from regexp + 'r=s' => \$G::qgrep_r, # recipient regexp + 's=s' => \$G::qgrep_s, # match against size field + 'y=s' => \$G::qgrep_y, # message younger than (secs) + 'o=s' => \$G::qgrep_o, # message older than (secs) 'z' => \$G::qgrep_z, # frozen only 'x' => \$G::qgrep_x, # non-frozen only 'c' => \$G::qgrep_c, # display match count 'l' => \$G::qgrep_l, # long format (default) 'i' => \$G::qgrep_i, # message ids only 'b' => \$G::qgrep_b, # brief format + 'size' => \$G::size_only, # sum the size of the matching msgs + 'not' => \$G::negate, # flip every test + 'R|reverse' => \$G::reverse, # reverse output (-R is qgrep option) + 'sort=s' => \@G::sort, # allow you to choose variables to sort by + 'freeze=s' => \$G::freeze, # freeze data in this file + 'thaw=s' => \$G::thaw, # thaw data from this file + 'unsorted' => \$G::unsorted, # unsorted, regardless of output format + 'random' => \$G::random, # (poorly) randomize evaluation order 'flatq' => \$G::flatq, # brief format 'caseful' => \$G::caseful, # in '=' criteria, respect case 'caseless' => \$G::caseless, # ...ignore case (default) - 'show-vars:s' => \$G::show_vars, # display the contents of these vars + 'charset=s' => \$charset, # charset for $bh and $h variables + 'show-vars=s' => \$G::show_vars, # display the contents of these vars + 'just-vars' => \$G::just_vars, # only display vars, no other info 'show-rules' => \$G::show_rules, # display compiled match rules 'show-tests' => \$G::show_tests # display tests as applied to each message ) || exit(1); +# if both freeze and thaw specified, only thaw as it is less desctructive +$G::freeze = undef if ($G::freeze && $G::thaw); +freeze_start() if ($G::freeze); +thaw_start() if ($G::thaw); + +# massage sort options (make '$var,Var:' be 'var','var') +for (my $i = scalar(@G::sort)-1; $i >= 0; $i--) { + $G::sort[$i] = lc($G::sort[$i]); + $G::sort[$i] =~ s/[\$:\s]//g; + if ((my @vars = split(/,/, $G::sort[$i])) > 1) { + $G::sort[$i] = $vars[0]; shift(@vars); # replace current slot w/ first var + splice(@G::sort, $i+1, 0, @vars); # add other vars after current pos + } +} +push(@G::sort, "message_exim_id") if (@G::sort); +die "empty value provided to --sort not allowed, exiting\n" + if (grep /^\s*$/, @G::sort); + +# massage the qgrep options into standard criteria push(@ARGV, "\$sender_address =~ /$G::qgrep_f/") if ($G::qgrep_f); push(@ARGV, "\$recipients =~ /$G::qgrep_r/") if ($G::qgrep_r); push(@ARGV, "\$shown_message_size eq $G::qgrep_s") if ($G::qgrep_s); @@ -69,35 +105,51 @@ push(@ARGV, "\$message_age < $G::qgrep_y") if ($G::qgrep_y); push(@ARGV, "\$message_age > $G::qgrep_o") if ($G::qgrep_o); push(@ARGV, "\$deliver_freeze") if ($G::qgrep_z); push(@ARGV, "!\$deliver_freeze") if ($G::qgrep_x); + $G::mailq_bp = $G::mailq_bp; # shut up -w $G::and = $G::and; # shut up -w $G::msg_ids = {}; # short circuit when crit is only MID $G::caseless = $G::caseful ? 0 : 1; # nocase by default, case if both @G::recipients_crit = (); # holds per-recip criteria $spool = $G::spool if ($G::spool); -my $count_only = 1 if ($G::mailq_bpc || $G::qgrep_c); -my $unsorted = 1 if ($G::mailq_bpr || $G::mailq_bpra || $G::mailq_bpru); -my $msg = get_all_msgs($spool, $unsorted); +my $input_dir = $G::input_dir || ($G::finput ? "Finput" : "input"); +my $count_only = 1 if ($G::mailq_bpc || $G::qgrep_c); +my $unsorted = 1 if ($G::mailq_bpr || $G::mailq_bpra || + $G::mailq_bpru || $G::unsorted); +my $msg = $G::thaw ? thaw_message_list() + : get_all_msgs($spool, $input_dir, $unsorted, + $G::reverse, $G::random); +die "Problem accessing thaw file\n" if ($G::thaw && !$msg); my $crit = process_criteria(\@ARGV); my $e = Exim::SpoolFile->new(); my $tcount = 0 if ($count_only); # holds count of all messages my $mcount = 0 if ($count_only); # holds count of matching messages +my $total_size = 0 if ($G::size_only); $e->set_undelivered_only(1) if ($G::mailq_bpru || $G::mailq_bpu); $e->set_show_generated(1) if ($G::mailq_bpra || $G::mailq_bpa); $e->output_long() if ($G::qgrep_l); $e->output_idonly() if ($G::qgrep_i); $e->output_brief() if ($G::qgrep_b); $e->output_flatq() if ($G::flatq); +$e->output_vars_only() if ($G::just_vars && $G::show_vars); $e->set_show_vars($G::show_vars) if ($G::show_vars); -$e->set_spool($spool); +$e->set_spool($spool, $input_dir); MSG: foreach my $m (@$msg) { next if (scalar(keys(%$G::msg_ids)) && !$G::or && !$G::msg_ids->{$m->{message}}); - if (!$e->parse_message($m->{message})) { - warn "Couldn't parse $m->{message}: ".$e->error()."\n"; - next MSG; + if ($G::thaw) { + my $data = thaw_data(); + if (!$e->restore_state($data)) { + warn "Couldn't thaw $data->{_message}: ".$e->error()."\n"; + next MSG; + } + } else { + if (!$e->parse_message($m->{message}, $m->{path})) { + warn "Couldn't parse $m->{message}: ".$e->error()."\n"; + next MSG; + } } $tcount++; my $match = 0; @@ -133,67 +185,201 @@ foreach my $m (@$msg) { # skip this message if any criteria were supplied and it didn't match next MSG if ((scalar(@$crit) || scalar(@local_crit)) && !$match); - if ($count_only) { + if ($count_only || $G::size_only) { $mcount++; + $total_size += $e->get_var('message_size'); } else { - $e->print_message(\*STDOUT); + if (@G::sort) { + # if we are defining criteria to sort on, save the message here. If + # we don't save here and do the sort later, we have a chicken/egg + # problem + push(@G::to_print, { vars => {}, output => "" }); + foreach my $var (@G::sort) { + # save any values we want to sort on. I don't like doing the internal + # struct access here, but calling get_var a bunch can be _slow_ =( + $G::sort_type{$var} ||= '<=>'; + $G::to_print[-1]{vars}{$var} = $e->{_vars}{$var}; + $G::sort_type{$var} = 'cmp' if ($G::to_print[-1]{vars}{$var} =~ /\D/); + } + $G::to_print[-1]{output} = $e->format_message(); + } else { + print $e->format_message(); + } + } + + if ($G::freeze) { + freeze_data($e->get_state()); + push(@G::frozen_msgs, $m); } } -if ($G::mailq_bpc) { - print "$mcount\n"; -} elsif ($G::qgrep_c) { - print "$mcount matches out of $tcount messages\n"; +if (@G::to_print) { + msg_sort(\@G::to_print, \@G::sort, $G::reverse); + foreach my $msg (@G::to_print) { + print $msg->{output}; + } +} + +if ($G::qgrep_c) { + print "$mcount matches out of $tcount messages" . + ($G::size_only ? " ($total_size)" : "") . "\n"; +} elsif ($G::mailq_bpc) { + print "$mcount" . ($G::size_only ? " ($total_size)" : "") . "\n"; +} elsif ($G::size_only) { + print "$total_size\n"; +} + +if ($G::freeze) { + freeze_message_list(\@G::frozen_msgs); + freeze_end(); +} elsif ($G::thaw) { + thaw_end(); } exit; +# sender_address_domain,shown_message_size +sub msg_sort { + my $msgs = shift; + my $vars = shift; + my $reverse = shift; + + my @pieces = (); + foreach my $v (@G::sort) { + push(@pieces, "\$a->{vars}{\"$v\"} $G::sort_type{$v} \$b->{vars}{\"$v\"}"); + } + my $sort_str = join(" || ", @pieces); + + @$msgs = sort { eval $sort_str } (@$msgs); + @$msgs = reverse(@$msgs) if ($reverse); +} + +sub try_load { + my $mod = shift; + + eval("use $mod"); + return $@ ? 0 : 1; +} + +# FREEZE FILE FORMAT: +# message_data_bytes +# message_data +# <...> +# EOM +# message_list +# message_list_bytes <- 10 bytes, zero-packed, plus \n + +sub freeze_start { + eval("use Storable"); + die "Storable module not found: $@\n" if ($@); + open(O, ">$G::freeze") || die "Can't open freeze file $G::freeze: $!\n"; + $G::freeze_handle = \*O; +} + +sub freeze_end { + close($G::freeze_handle); +} + +sub thaw_start { + eval("use Storable"); + die "Storable module not found: $@\n" if ($@); + open(I, "<$G::thaw") || die "Can't open freeze file $G::thaw: $!\n"; + $G::freeze_handle = \*I; +} + +sub thaw_end { + close($G::freeze_handle); +} + +sub freeze_data { + my $h = Storable::freeze($_[0]); + print $G::freeze_handle length($h)+1, "\n$h\n"; +} + +sub freeze_message_list { + my $h = Storable::freeze($_[0]); + my $l = length($h) + 1; + printf $G::freeze_handle "EOM\n$l\n$h\n%010d\n", $l+11+length($l)+1; +} + +sub thaw_message_list { + my $orig_pos = tell($G::freeze_handle); + seek($G::freeze_handle, -11, 2); + chomp(my $bytes = <$G::freeze_handle>); + seek($G::freeze_handle, $bytes * -1, 2); + my $obj = thaw_data(); + seek($G::freeze_handle, 0, $orig_pos); + return($obj); +} + +sub thaw_data { + my $obj; + chomp(my $bytes = <$G::freeze_handle>); + return(undef) if (!$bytes || $bytes eq 'EOM'); + my $read = read(I, $obj, $bytes); + die "Format error in thaw file (expected $bytes bytes, got $read)\n" + if ($bytes != $read); + chomp($obj); + return(Storable::thaw($obj)); +} + sub process_criteria { my $a = shift; my @c = (); my $e = 0; foreach (@$a) { - foreach my $t ('@') { s/$t/\\$t/g; } # '$' + foreach my $t ('@') { s/$t/\\$t/g; } if (/^(.*?)\s+(<=|>=|==|!=|<|>)\s+(.*)$/) { #print STDERR "found as integer\n"; my $v = $1; my $o = $2; my $n = $3; - if ($n =~ /^([\d\.]+)M$/) { $n = $1 * 1024 * 1024; } - elsif ($n =~ /^([\d\.]+)K$/) { $n = $1 * 1024; } - elsif ($n =~ /^([\d\.]+)B?$/) { $n = $1; } - elsif ($n =~ /^([\d\.]+)d$/) { $n = $1 * 60 * 60 * 24; } - elsif ($n =~ /^([\d\.]+)h$/) { $n = $1 * 60 * 60; } - elsif ($n =~ /^([\d\.]+)m$/) { $n = $1 * 60; } - elsif ($n =~ /^([\d\.]+)s?$/) { $n = $1; } + if ($n =~ /^(-?[\d\.]+)M$/) { $n = $1 * 1024 * 1024; } + elsif ($n =~ /^(-?[\d\.]+)K$/) { $n = $1 * 1024; } + elsif ($n =~ /^(-?[\d\.]+)B?$/) { $n = $1; } + elsif ($n =~ /^(-?[\d\.]+)d$/) { $n = $1 * 60 * 60 * 24; } + elsif ($n =~ /^(-?[\d\.]+)h$/) { $n = $1 * 60 * 60; } + elsif ($n =~ /^(-?[\d\.]+)m$/) { $n = $1 * 60; } + elsif ($n =~ /^(-?[\d\.]+)s?$/) { $n = $1; } else { print STDERR "Expression $_ did not parse: numeric comparison with ", "non-number\n"; $e = 1; next; } - push(@c, { var => lc($v), cmp => "(\$var $o $n) ? 1 : 0" }); + push(@c, { var => lc($v), cmp => "(\$var $o $n)" }); } elsif (/^(.*?)\s+(=~|!~)\s+(.*)$/) { #print STDERR "found as string regexp\n"; - push(@c, { var => lc($1), cmp => "(\"\$var\" $2 $3) ? 1 : 0" }); + push(@c, { var => lc($1), cmp => "(\"\$var\" $2 $3)" }); } elsif (/^(.*?)\s+=\s+(.*)$/) { #print STDERR "found as bare string regexp\n"; my $case = $G::caseful ? '' : 'i'; - push(@c, { var => lc($1), cmp => "(\"\$var\" =~ /$2/$case) ? 1 : 0" }); + push(@c, { var => lc($1), cmp => "(\"\$var\" =~ /$2/$case)" }); + # quote special characters in perl text string + #foreach my $t ('@') { $c[-1]{cmp} =~ s/$t/\\$t/g; } } elsif (/^(.*?)\s+(eq|ne)\s+(.*)$/) { #print STDERR "found as string cmp\n"; my $var = lc($1); my $op = $2; my $val = $3; $val =~ s|^(['"])(.*)\1$|$2|; - push(@c, { var => $var, cmp => "(\"\$var\" $op \"$val\") ? 1 : 0" }); + push(@c, { var => $var, cmp => "(\"\$var\" $op \"$val\")" }); if (($var eq 'message_id' || $var eq 'message_exim_id') && $op eq "eq") { #print STDERR "short circuit @c[-1]->{cmp} $val\n"; $G::msg_ids->{$val} = 1; } - } elsif (/^(!)?(\S+)$/) { + #foreach my $t ('@') { $c[-1]{cmp} =~ s/$t/\\$t/g; } + } elsif (/^(\S+)$/) { #print STDERR "found as boolean\n"; - push(@c, { var => lc($2), cmp => "($1\$var) ? 1 : 0" }); + push(@c, { var => lc($1), cmp => "(\$var)" }); } else { print STDERR "Expression $_ did not parse\n"; $e = 1; + next; + } + # assign the results of the cmp test here (handle "!" negation) + # also handle global --not negation + if ($c[-1]{var} =~ s|^!||) { + $c[-1]{cmp} .= $G::negate ? " ? 1 : 0" : " ? 0 : 1"; + } else { + $c[-1]{cmp} .= $G::negate ? " ? 0 : 1" : " ? 1 : 0"; } # support the each_* psuedo variables. Steal the criteria off of the # queue for special processing later @@ -212,25 +398,42 @@ sub process_criteria { } sub get_all_msgs { - my $d = shift() . '/input'; - my $u = shift; + my $d = shift(); + my $i = shift(); + my $u = shift; # don't sort + my $r = shift; # right before returning, reverse order + my $o = shift; # if true, randomize list order before returning my @m = (); + if ($i =~ m|^/|) { $d = $i; } else { $d = $d . '/' . $i; } + opendir(D, "$d") || die "Couldn't opendir $d: $!\n"; foreach my $e (grep !/^\./, readdir(D)) { if ($e =~ /^[a-zA-Z0-9]$/) { opendir(DD, "$d/$e") || next; foreach my $f (grep !/^\./, readdir(DD)) { - push(@m, { message => $1, path => "$e/$1" }) if ($f =~ /^(.{16})-H$/); + push(@m, { message => $1, path => "$d/$e" }) if ($f =~ /^(.{16})-H$/); } closedir(DD); } elsif ($e =~ /^(.{16})-H$/) { - push(@m, { message => $1, path => $1 }); + push(@m, { message => $1, path => $d }); } } closedir(D); - return($u ? \@m : [ sort { $a->{message} cmp $b->{message} } @m ]); + if ($o) { + my $c = scalar(@m); + # loop twice to pretend we're doing a good job of mixing things up + for (my $i = 0; $i < 2 * $c; $i++) { + my $rand = int(rand($c)); + ($m[$i % $c],$m[$rand]) = ($m[$rand],$m[$i % $c]); + } + } elsif (!$u) { + @m = sort { $a->{message} cmp $b->{message} } @m; + } + @m = reverse(@m) if ($r); + + return(\@m); } BEGIN { @@ -248,12 +451,14 @@ sub new { bless($self, $class); $self->{_spool_dir} = ''; + $self->{_input_path} = ''; $self->{_undelivered_only} = 0; $self->{_show_generated} = 0; $self->{_output_long} = 1; $self->{_output_idonly} = 0; $self->{_output_brief} = 0; $self->{_output_flatq} = 0; + $self->{_output_vars_only} = 0; $self->{_show_vars} = []; $self->_reset(); @@ -267,6 +472,7 @@ sub output_long { $self->{_output_idonly} = 0; $self->{_output_brief} = 0; $self->{_output_flatq} = 0; + $self->{_output_vars_only} = 0; } sub output_idonly { @@ -276,6 +482,7 @@ sub output_idonly { $self->{_output_idonly} = 1; $self->{_output_brief} = 0; $self->{_output_flatq} = 0; + $self->{_output_vars_only} = 0; } sub output_brief { @@ -285,6 +492,7 @@ sub output_brief { $self->{_output_idonly} = 0; $self->{_output_brief} = 1; $self->{_output_flatq} = 0; + $self->{_output_vars_only} = 0; } sub output_flatq { @@ -294,6 +502,17 @@ sub output_flatq { $self->{_output_idonly} = 0; $self->{_output_brief} = 0; $self->{_output_flatq} = 1; + $self->{_output_vars_only} = 0; +} + +sub output_vars_only { + my $self = shift; + + $self->{_output_long} = 0; + $self->{_output_idonly} = 0; + $self->{_output_brief} = 0; + $self->{_output_flatq} = 0; + $self->{_output_vars_only} = 1; } sub set_show_vars { @@ -334,6 +553,7 @@ sub _reset { $self->{_message} = ''; $self->{_path} = ''; $self->{_vars} = {}; + $self->{_vars_raw} = {}; $self->{_numrecips} = 0; $self->{_udel_tree} = {}; @@ -348,8 +568,9 @@ sub parse_message { $self->_reset(); $self->{_message} = shift || return(0); - return(0) if (!$self->{_spool_dir}); - if (!$self->_find_path()) { + $self->{_path} = shift; # optional path to message + return(0) if (!$self->{_input_path}); + if (!$self->{_path} && !$self->_find_path()) { # assume the message was delivered from under us and ignore $self->{_delivered} = 1; return(1); @@ -359,15 +580,69 @@ sub parse_message { return(1); } +# take the output of get_state() and set up a message internally like +# parse_message (except from a saved data struct, not by parsing the +# files on disk). +sub restore_state { + my $self = shift; + my $h = shift; + + return(1) if ($h->{_delivered}); + $self->_reset(); + $self->{_message} = $h->{_message} || return(0); + return(0) if (!$self->{_input_path}); + + $self->{_path} = $h->{_path}; + $self->{_vars} = $h->{_vars}; + $self->{_numrecips} = $h->{_numrecips}; + $self->{_udel_tree} = $h->{_udel_tree}; + $self->{_del_tree} = $h->{_del_tree}; + $self->{_recips} = $h->{_recips}; + + $self->{_vars}{message_age} = time() - $self->{_vars}{received_time}; + return(1); +} + +# This returns the state data for a specific message in a format that can +# be later frozen back in to regain state +# +# after calling this function, this specific state is not expect to be +# reused. That's because we're returning direct references to specific +# internal structures. We're also modifying the structure ourselves +# by deleting certain internal message variables. +sub get_state { + my $self = shift; + my $h = {}; # this is the hash ref we'll be returning. + + $h->{_delivered} = $self->{_delivered}; + $h->{_message} = $self->{_message}; + $h->{_path} = $self->{_path}; + $h->{_vars} = $self->{_vars}; + $h->{_numrecips} = $self->{_numrecips}; + $h->{_udel_tree} = $self->{_udel_tree}; + $h->{_del_tree} = $self->{_del_tree}; + $h->{_recips} = $self->{_recips}; + + # delete some internal variables that we will rebuild later if needed + delete($h->{_vars}{message_body}); + delete($h->{_vars}{message_age}); + + return($h); +} + +# keep this sub as a feature if we ever break this module out, but do away +# with its use in exipick (pass it in from caller instead) sub _find_path { my $self = shift; return(0) if (!$self->{_message}); - return(0) if (!$self->{_spool_dir}); + return(0) if (!$self->{_input_path}); - foreach my $f ('', substr($self->{_message}, 5, 1).'/') { - if (-f $self->{_spool_dir} . "/input/$f" . $self->{_message} . '-H') { - $self->{_path} = $self->{_spool_dir} . "/input/$f"; + # test split spool first on the theory that people concerned about + # performance will have split spool set =). + foreach my $f (substr($self->{_message}, 5, 1).'/', '') { + if (-f "$self->{_input_path}/$f$self->{_message}-H") { + $self->{_path} = "$self->{_input_path}}/$f"; return(1); } } @@ -377,25 +652,176 @@ sub _find_path { sub set_spool { my $self = shift; $self->{_spool_dir} = shift; + $self->{_input_path} = shift; + if ($self->{_input_path} !~ m|^/|) { + $self->{_input_path} = $self->{_spool_dir} . '/' . $self->{_input_path}; + } +} + +sub get_matching_vars { + my $self = shift; + my $e = shift; + + if ($e =~ /^\^/) { + my @r = (); + foreach my $v (keys %{$self->{_vars}}) { push(@r, $v) if ($v =~ /$e/); } + return(@r); + } else { + return($e); + } } # accepts a variable with or without leading '$' or trailing ':' sub get_var { my $self = shift; - my $var = lc(shift); + my $var = lc(shift); $var =~ s/^\$//; $var =~ s/:$//; + + if ($var eq 'message_body' && !defined($self->{_vars}{message_body})) { + $self->_parse_body() + } elsif ($var =~ s|^([rb]?h)(eader)?_|${1}eader_| && + exists($self->{_vars}{$var}) && !defined($self->{_vars}{$var})) + { + if ((my $type = $1) eq 'rh') { + $self->{_vars}{$var} = join('', @{$self->{_vars_raw}{$var}{vals}}); + } else { + # both bh_ and h_ build their strings from rh_. Do common work here + my $rh = $var; $rh =~ s|^b?|r|; + my $comma = 1 if ($self->{_vars_raw}{$rh}{type} =~ /^[BCFRST]$/); + foreach (@{$self->{_vars_raw}{$rh}{vals}}) { + my $x = $_; # editing $_ here would change the original, which is bad + $x =~ s|^\s+||; + $x =~ s|\s+$||; + if ($comma) { chomp($x); $self->{_vars}{$var} .= "$x,\n"; } + else { $self->{_vars}{$var} .= $x; } + } + $self->{_vars}{$var} =~ s|[\s\n]*$||; + $self->{_vars}{$var} =~ s|,$|| if ($comma); + # ok, that's the preprocessing, not do specific processing for h type + if ($type eq 'bh') { + $self->{_vars}{$var} = $self->_decode_2047($self->{_vars}{$var}); + } else { + $self->{_vars}{$var} = + $self->_decode_2047($self->{_vars}{$var}, $charset); + } + } + } + elsif ($var eq 'received_count' && !defined($self->{_vars}{received_count})) + { + $self->{_vars}{received_count} = + scalar(@{$self->{_vars_raw}{rheader_received}{vals}}); + } + elsif ($var eq 'message_headers' && !defined($self->{_vars}{message_headers})) + { + $self->{_vars}{$var} = + $self->_decode_2047($self->{_vars}{message_headers_raw}, $charset); + chomp($self->{_vars}{$var}); + } + elsif ($var eq 'reply_address' && !defined($self->{_vars}{reply_address})) + { + $self->{_vars}{reply_address} = exists($self->{_vars}{"header_reply-to"}) + ? $self->get_var("header_reply-to") : $self->get_var("header_from"); + } + + #chomp($self->{_vars}{$var}); # I think this was only for headers, obsolete + return $self->{_vars}{$var}; +} + +sub _decode_2047 { + my $self = shift; + my $s = shift; # string to decode + my $c = shift; # target charset. If empty, just decode, don't convert + my $t = ''; # the translated string + my $e = 0; # set to true if we get an error in here anywhere + + return($s) if ($s !~ /=\?/); # don't even bother to look if there's no sign + + my @p = (); + foreach my $mw (split(/(=\?[^\?]{3,}\?[BQ]\?[^\?]{1,74}\?=)/i, $s)) { + next if ($mw eq ''); + if ($mw =~ /=\?([^\?]{3,})\?([BQ])\?([^\?]{1,74})\?=/i) { + push(@p, { data => $3, encoding => uc($2), charset => uc($1), + is_mime => 1 }); + if ($p[-1]{encoding} eq 'Q') { + my @ow = split('', $p[-1]{data}); + my @nw = (); + for (my $i = 0; $i < @ow; $i++) { + if ($ow[$i] eq '_') { push(@nw, ' '); } + elsif ($ow[$i] eq '=') { + if (scalar(@ow) - ($i+1) < 2) { # ran out of characters + $e = 1; last; + } elsif ($ow[$i+1] !~ /[\dA-F]/i || $ow[$i+2] !~ /[\dA-F]/i) { + $e = 1; last; + } else { + #push(@nw, chr('0x'.$ow[$i+1].$ow[$i+2])); + push(@nw, pack("C", hex($ow[$i+1].$ow[$i+2]))); + $i += 2; + } + } + elsif ($ow[$i] =~ /\s/) { # whitspace is illegal + $e = 1; + last; + } + else { push(@nw, $ow[$i]); } + } + $p[-1]{data} = join('', @nw); + } elsif ($p[-1]{encoding} eq 'B') { + my $x = $p[-1]{data}; + $x =~ tr#A-Za-z0-9+/##cd; + $x =~ s|=+$||; + $x =~ tr#A-Za-z0-9+/# -_#; + my $r = ''; + while ($x =~ s/(.{1,60})//s) { + $r .= unpack("u", chr(32 + int(length($1)*3/4)) . $1); + } + $p[-1]{data} = $r; + } + } else { + push(@p, { data => $mw, is_mime => 0, + is_ws => ($mw =~ m|^[\s\n]+|sm) ? 1 : 0 }); + } + } - $var =~ s/^\$//; - $var =~ s/:$//; + for (my $i = 0; $i < @p; $i++) { + # mark entities we want to skip (whitespace between consecutive mimewords) + if ($p[$i]{is_mime} && $p[$i+1]{is_ws} && $p[$i+2]{is_mime}) { + $p[$i+1]{skip} = 1; + } - $self->_parse_body() - if ($var eq 'message_body' && !$self->{_vars}{message_body}); + # if word is a mimeword and we have access to Encode and charset was + # specified, try to convert text + # XXX _cannot_ get consistent conversion results in perl, can't get them + # to return same conversions that exim performs. Until I can figure this + # out, don't attempt any conversions (header_ will return same value as + # bheader_). + #if ($c && $p[$i]{is_mime} && $self->_try_load('Encode')) { + # # XXX not sure how to catch errors here + # Encode::from_to($p[$i]{data}, $p[$i]{charset}, $c); + #} + + # replace binary zeros w/ '?' in decoded text + if ($p[$i]{is_mime}) { $p[$i]{data} =~ s|\x00|?|g; } + } - return $self->{_vars}{$var}; + if ($e) { + return($s); + } else { + return(join('', map { $_->{data} } grep { !$_->{skip} } @p)); + } +} + +# This isn't a class func but I'm tired +sub _try_load { + my $self = shift; + my $mod = shift; + + eval("use $mod"); + return $@ ? 0 : 1; } sub _parse_body { my $self = shift; my $f = $self->{_path} . '/' . $self->{_message} . '-D'; + $self->{_vars}{message_body} = ""; # define var so we only come here once open(I, "<$f") || return($self->_error("Couldn't open $f: $!")); chomp($_ = ); @@ -411,8 +837,23 @@ sub _parse_body { sub _parse_header { my $self = shift; my $f = $self->{_path} . '/' . $self->{_message} . '-H'; + $self->{_vars}{header_path} = $f; + $self->{_vars}{data_path} = $self->{_path} . '/' . $self->{_message} . '-D'; + + if (!open(I, "<$f")) { + # assume message went away and silently ignore + $self->{_delivered} = 1; + return(1); + } + + # There are a few numeric variables that should explicitly be set to + # zero if they aren't found in the header. Technically an empty value + # works just as well, but might as well be pedantic + $self->{_vars}{body_zerocount} = 0; + $self->{_vars}{host_lookup_deferred} = 0; + $self->{_vars}{host_lookup_failed} = 0; + $self->{_vars}{tls_certificate_verified} = 0; - open(I, "<$f") || return($self->_error("Couldn't open $f: $!")); chomp($_ = ); return(0) if ($self->{_message}.'-H' ne $_); $self->{_vars}{message_id} = $self->{_message}; @@ -457,12 +898,14 @@ sub _parse_header { read(I, $self->{_vars}{$t}, $2+1) || return(0); chomp($self->{_vars}{$t}); } elsif ($tag eq '-aclc') { - return(0) if ($arg !~ /^(\d+)\s(\d+)$/); + #return(0) if ($arg !~ /^(\d+)\s(\d+)$/); + return(0) if ($arg !~ /^(\S+)\s(\d+)$/); my $t = "acl_c$1"; read(I, $self->{_vars}{$t}, $2+1) || return(0); chomp($self->{_vars}{$t}); } elsif ($tag eq '-aclm') { - return(0) if ($arg !~ /^(\d+)\s(\d+)$/); + #return(0) if ($arg !~ /^(\d+)\s(\d+)$/); + return(0) if ($arg !~ /^(\S+)\s(\d+)$/); my $t = "acl_m$1"; read(I, $self->{_vars}{$t}, $2+1) || return(0); chomp($self->{_vars}{$t}); @@ -483,6 +926,8 @@ sub _parse_header { $self->{_vars}{host_lookup_failed} = 1; } elsif ($tag eq '-body_linecount') { $self->{_vars}{body_linecount} = $arg; + } elsif ($tag eq '-max_received_linelength') { + $self->{_vars}{max_received_linelength} = $arg; } elsif ($tag eq '-body_zerocount') { $self->{_vars}{body_zerocount} = $arg; } elsif ($tag eq '-frozen') { @@ -510,12 +955,16 @@ sub _parse_header { $self->{_vars}{tls_cipher} = $arg; } elsif ($tag eq '-tls_peerdn') { $self->{_vars}{tls_peerdn} = $arg; + } elsif ($tag eq '-tls_sni') { + $self->{_vars}{tls_sni} = $arg; } elsif ($tag eq '-host_address') { $self->{_vars}{sender_host_port} = $self->_get_host_and_port(\$arg); $self->{_vars}{sender_host_address} = $arg; } elsif ($tag eq '-interface_address') { - $self->{_vars}{interface_port} = $self->_get_host_and_port(\$arg); - $self->{_vars}{interface_address} = $arg; + $self->{_vars}{received_port} = + $self->{_vars}{interface_port} = $self->_get_host_and_port(\$arg); + $self->{_vars}{received_ip_address} = + $self->{_vars}{interface_address} = $arg; } elsif ($tag eq '-active_hostname') { $self->{_vars}{smtp_active_hostname} = $arg; } elsif ($tag eq '-host_auth') { @@ -571,6 +1020,12 @@ sub _parse_header { return($self->_error("incorrect format: $_")) if (length($2) != $3); $self->{_recips}{$1} = { pno => $4, errors_to => $2 }; $addr = $1; + } elsif (/^(\S*)\s(\S*)\s(\d+),(\d+)\s(\S*)\s(\d+),(-?\d+)#3$/) { + #print STDERR "exim4 new type #3 DSN (untested): $_\n"; + return($self->_error("incorrect format: $_")) + if ((length($2) != $3) || (length($5) != $6)); + $self->{_recips}{$1} = { pno => $7, errors_to => $5 }; + $addr = $1; } elsif (/^.*#(\d+)$/) { #print STDERR "exim4 #$1 style (unimplemented): $_\n"; $self->_error("exim4 #$1 style (unimplemented): $_"); @@ -603,44 +1058,31 @@ sub _parse_header { $_ .= $t; $t = getc(I); } - # ok, right here $t contains the header flag and $_ contains the number of - # bytes to read. If we ever use the header flag, grab it here. - $self->{_vars}{message_size} += $_ if ($t ne '*'); - $t = getc(I); # strip the space out of the file - my $bytes = $_; - return(0) if (read(I, $_, $bytes) != $bytes); - chomp(); # may regret this later - if ($t ne '*') { - # use of this temp variable is a little lame but it prevents a - # -w warning (Use of implicit split to @_ is deprecated) - my @t = split(/\n/); - $self->{_vars}{message_linecount} += scalar(@t); - } - # build the $header_ variable, following exim's rules (sort of) - if (/^([^ :]+):(.*)$/s) { - my $v = "header_" . lc($1); - my $d = $2; - $d =~ s/^\s*//; - $d =~ s/\s*$//; - $self->{_vars}{$v} .= (defined($self->{_vars}{$v}) ? "\n" : '') . $d; - $self->{_vars}{received_count}++ if ($v eq 'header_received'); + my $hdr_flag = $t; + my $hdr_bytes = $_; + $t = getc(I); # strip the space out of the file + return(0) if (read(I, $_, $hdr_bytes) != $hdr_bytes); + if ($hdr_flag ne '*') { + $self->{_vars}{message_linecount} += (tr/\n//); + $self->{_vars}{message_size} += $hdr_bytes; } - # push header onto $message_headers var, following exim's rules - $self->{_vars}{message_headers} .= - (defined($self->{_vars}{message_headers}) ? "\n" : '') . $_; - } - close(I); - if (length($self->{_vars}{"header_reply-to"}) > 0) { - $self->{_vars}{reply_address} = $self->{_vars}{"header_reply-to"}; - } else { - $self->{_vars}{reply_address} = $self->{_vars}{header_from}; + # mark (rb)?header_ vars as existing and store raw value. They'll be + # processed further in get_var() if needed + my($v,$d) = split(/:/, $_, 2); + $v = "header_" . lc($v); + $self->{_vars}{$v} = $self->{_vars}{"b$v"} = $self->{_vars}{"r$v"} = undef; + push(@{$self->{_vars_raw}{"r$v"}{vals}}, $d); + $self->{_vars_raw}{"r$v"}{type} = $hdr_flag; + $self->{_vars}{message_headers_raw} .= $_; } + close(I); $self->{_vars}{message_body_size} = (stat($self->{_path}.'/'.$self->{_message}.'-D'))[7] - 19; if ($self->{_vars}{message_body_size} < 0) { $self->{_vars}{message_size} = 0; + $self->{_vars}{message_body_missing} = 1; } else { $self->{_vars}{message_size} += $self->{_vars}{message_body_size} + 1; } @@ -648,12 +1090,12 @@ sub _parse_header { $self->{_vars}{message_linecount} += $self->{_vars}{body_linecount}; my $i = $self->{_vars}{message_size}; - if ($i == 0) { $i = ""; } - elsif ($i < 1024) { $i = sprintf("%d", $i); } - elsif ($i < 10*1024) { $i = sprintf("%.1fK", $i / 1024); } - elsif ($i < 1024*1024) { $i = sprintf("%dK", ($i+512)/1024); } - elsif ($i < 10*1024*1024) { $i = sprintf("%.1fM", $i/(1024*1024)); } - else { $i = sprintf("%dM", ($i + 512 * 1024)/(1024*1024)); } + if ($i == 0) { $i = ""; } + elsif ($i < 1024) { $i = sprintf("%d", $i); } + elsif ($i < 10240) { $i = sprintf("%.1fK", $i / 1024); } + elsif ($i < 1048576) { $i = sprintf("%dK", ($i+512)/1024); } + elsif ($i < 10485760) { $i = sprintf("%.1fM", $i/1048576); } + else { $i = sprintf("%dM", ($i + 524288)/1048576); } $self->{_vars}{shown_message_size} = $i; return(1); @@ -679,62 +1121,77 @@ sub _get_host_and_port { return(0); } -sub print_message { +# honoring all formatting preferences, return a scalar variable of the +# information for the single message matching what exim -bp would show. +# We can print later if we want. +sub format_message { my $self = shift; - my $fh = shift || \*STDOUT; + my $o = ''; return if ($self->{_delivered}); - if ($self->{_output_idonly}) { - print $fh $self->{_message}; - foreach my $v (@{$self->{_show_vars}}) { - print $fh " $v='", $self->get_var($v), "'"; + # define any vars we want to print out for this message. The requests + # can be regexps, and the defined vars can change for each message, so we + # have to build this list for each message + my @vars = (); + if (@{$self->{_show_vars}}) { + my %t = (); + foreach my $e (@{$self->{_show_vars}}) { + foreach my $v ($self->get_matching_vars($e)) { + next if ($t{$v}); $t{$v}++; push(@vars, $v); + } } - print $fh "\n"; - return; + } + + if ($self->{_output_idonly}) { + $o .= $self->{_message}; + foreach my $v (@vars) { $o .= " $v='" . $self->get_var($v) . "'"; } + $o .= "\n"; + return $o; + } elsif ($self->{_output_vars_only}) { + foreach my $v (@vars) { $o .= $self->get_var($v) . "\n"; } + return $o; } if ($self->{_output_long} || $self->{_output_flatq}) { my $i = int($self->{_vars}{message_age} / 60); if ($i > 90) { $i = int(($i+30)/60); - if ($i > 72) { printf $fh "%2dd ", int(($i+12)/24); } - else { printf $fh "%2dh ", $i; } - } else { printf $fh "%2dm ", $i; } - - if ($self->{_output_flatq} && $self->{_show_vars}) { - print $fh join(';', - map { "$_='".$self->get_var($_)."'" } - (@{$self->{_show_vars}}) - ); + if ($i > 72) { $o .= sprintf "%2dd ", int(($i+12)/24); } + else { $o .= sprintf "%2dh ", $i; } + } else { $o .= sprintf "%2dm ", $i; } + + if ($self->{_output_flatq} && @vars) { + $o .= join(';', map { "$_='".$self->get_var($_)."'" } (@vars) + ); } else { - printf $fh "%5s", $self->{_vars}{shown_message_size}; + $o .= sprintf "%5s", $self->{_vars}{shown_message_size}; } - print $fh " "; + $o .= " "; } - print $fh "$self->{_message} "; - print $fh "From: " if ($self->{_output_brief}); - print $fh "<$self->{_vars}{sender_address}>"; + $o .= "$self->{_message} "; + $o .= "From: " if ($self->{_output_brief}); + $o .= "<$self->{_vars}{sender_address}>"; if ($self->{_output_long}) { - print $fh " ($self->{_vars}{originator_login})" + $o .= " ($self->{_vars}{originator_login})" if ($self->{_vars}{sender_set_untrusted}); # XXX exim contains code here to print spool format errors - print $fh " *** frozen ***" if ($self->{_vars}{deliver_freeze}); - print $fh "\n"; + $o .= " *** frozen ***" if ($self->{_vars}{deliver_freeze}); + $o .= "\n"; - foreach my $v (@{$self->{_show_vars}}) { - printf $fh " %25s = '%s'\n", $v, $self->get_var($v); + foreach my $v (@vars) { + $o .= sprintf " %25s = '%s'\n", $v, $self->get_var($v); } foreach my $r (keys %{$self->{_recips}}) { next if ($self->{_del_tree}{$r} && $self->{_undelivered_only}); - printf $fh " %s %s\n", $self->{_del_tree}{$r} ? "D" : " ", $r; + $o .= sprintf " %s %s\n", $self->{_del_tree}{$r} ? "D" : " ", $r; } if ($self->{_show_generated}) { foreach my $r (keys %{$self->{_del_tree}}) { next if ($self->{_recips}{$r}); - printf $fh " +D %s\n", $r; + $o .= sprintf " +D %s\n", $r; } } } elsif ($self->{_output_brief}) { @@ -743,24 +1200,30 @@ sub print_message { next if ($self->{_del_tree}{$r}); push(@r, $r); } - print $fh " To: ", join(';', @r); - if ($self->{_show_vars} && scalar(@{$self->{_show_vars}})) { - print $fh " Vars: ", join(';', - map { "$_='".$self->get_var($_)."'" } - (@{$self->{_show_vars}}) - ); + $o .= " To: " . join(';', @r); + if (scalar(@vars)) { + $o .= " Vars: ".join(';',map { "$_='".$self->get_var($_)."'" } (@vars)); } } elsif ($self->{_output_flatq}) { - print $fh " *** frozen ***" if ($self->{_vars}{deliver_freeze}); + $o .= " *** frozen ***" if ($self->{_vars}{deliver_freeze}); my @r = (); foreach my $r (keys %{$self->{_recips}}) { next if ($self->{_del_tree}{$r}); push(@r, $r); } - print $fh " ", join(' ', @r); + $o .= " " . join(' ', @r); } - print $fh "\n"; + $o .= "\n"; + return($o); +} + +sub print_message { + my $self = shift; + my $fh = shift || \*STDOUT; + return if ($self->{_delivered}); + + print $fh $self->format_message(); } sub dump { @@ -808,445 +1271,545 @@ __END__ =head1 NAME -exipick - display messages from Exim queue based on a variety of criteria +exipick - selectively display messages from an Exim queue -=head1 USAGE +=head1 SYNOPSIS -exipick [--help|--version] | [-spool ] [-and|-or] [-bp|-bpa|-bpc|-bpr|-bpra|-bpru|-bpu] [ [ ...]] +exipick [] [ [ ...]] =head1 DESCRIPTION -exipick is designed to display the contents of a Exim mail spool based on user-specified criteria. It is designed to mimic the output of 'exim -bp' (or any of the other -bp* options) and Exim's spec.txt should be used to learn more about the exact format of the output. The criteria are formed by creating comparisons against characteristics of the messages, for instance $message_size, $sender_helo_name, or $message_headers. +exipick is a tool to display messages in an Exim queue. It is very similar to exiqgrep and is, in fact, a drop in replacement for exiqgrep. exipick allows you to select messages to be displayed using any piece of data stored in an Exim spool file. Matching messages can be displayed in a variety of formats. + +=head1 QUICK START + +Delete every frozen message from queue: + exipick -zi | xargs exim -Mrm + +Show only messages which have not yet been virus scanned: + exipick '$received_protocol ne virus-scanned' + +Run the queue in a semi-random order: + exipick -i --random | xargs exim -M + +Show the count and total size of all messages which either originated from localhost or have a received protocol of 'local': + exipick --or --size --bpc \ + '$sender_host_address eq 127.0.0.1' \ + '$received_protocol eq local' + +Display all messages received on the MSA port, ordered first by the sender's email domain and then by the size of the emails: + exipick --sort sender_address_domain,message_size \ + '$received_port == 587' + +Display only messages whose every recipient is in the example.com domain, also listing the IP address of the sending host: + exipick --show-vars sender_host_address \ + '$each_recipients = example.com' + +Same as above, but show values for all defined variables starting with sender_ and the number of recipients: + exipick --show-vars ^sender_,recipients_count \ + '$each_recipients = example.com' =head1 OPTIONS =over 4 -=item --spool +=item --and -The path to Exim's spool directory. In general usage you should set the $spool variable in the script to your site's main spool directory (and if exipick was installed from the Exim distribution, this is done by default), but this option is useful for alternate installs, or installs on NFS servers, etc. +Display messages matching all criteria (default) -=item --and +=item -b -A message will be displayed only if it matches all of the specified criteria. This is the default. +Display messages in brief format (exiqgrep) -=item --or +=item -bp + +Display messages in standard mailq format (default) + +=item -bpa + +Same as -bp, show generated addresses also (exim) + +=item -bpc + +Show a count of matching messages (exim) + +=item -bpr + +Same as '-bp --unsorted' (exim) -A message will be displayed if it matches any of the specified criteria. +=item -bpra + +Same as '-bpa --unsorted' (exim) + +=item -bpru + +Same as '-bpu --unsorted' (exim) + +=item -bpu + +Same as -bp, but only show undelivered messages (exim) + +=item -c + +Show a count of matching messages (exiqgrep) =item --caseful -By default criteria using the '=' operator are caseless. Specifying this option make them respect case. +Make operators involving '=' honor case -=item --show-vars [,...] +=item --charset -Cause the value of each specified variable to be displayed for every message dispayed. For instance, the command "exipick --show-vars '$sender_ident' 'sender_host_address eq 127.0.01'" will show the ident string for every message submitted via localhost. How exactly the variable value is diplayed changes according to what output format you specify. +Override the default local character set for $header_ decoding -=item --show-rules +=item -f -If specified the internal representation of each message criteria is shown. This is primarily used for debugging purposes. +Same as '$sender_address =~ //' (exiqgrep). Note that this preserves the default case sensitivity of exiqgrep's interface. -==item --show-tests +=item --finput -If specified, for every message (regardless of matching criteria) the criteria's actual value is shown and the compiled internal eval is shown. This is used primarily for debugging purposes. +Same as '--input-dir Finput'. 'Finput' is where exim copies frozen messages when compiled with SUPPORT_MOVE_FROZEN_MESSAGES. =item --flatq -Change format of output so that every message is on a single line. Useful for parsing with tools such as sed, awk, cut, etc. +Use a single-line output format -=item The -bp* options all control how much information is displayed and in what manner. They all match the functionality of the options of the same name in Exim. Briefly: +=item --freeze -=item -bp display the matching messages in 'mailq' format. +Save queue information in an quickly retrievable format -=item -bpa ... with generated addresses as well. +=item --help -=item -bpc ... just show a count of messages. +Display this output -=item -bpr ... do not sort. +=item -i -=item -bpra ... with generated addresses, unsorted. +Display only the message IDs (exiqgrep) -=item -bpru ... only undelivered addresses, unsorted. +=item --input-dir -=item -bpu ... only undelivered addresses. +Set the name of the directory under the spool directory. By defaut this is "input". If this starts with '/', the value of --spool is ignored. See also --finput. -Please see Exim's spec.txt for details on the format and information displayed with each option. +=item -l -=item The following options are included for compatibility with the 'exiqgrep' utility: +Same as -bp (exiqgrep) -=item -f Same as '$sender_address = ' +=item --not -=item -r Same as '$recipients = ' +Negate all tests. -=item -s Same as '$shown_message_size eq ' +=item -o -=item -y Same as '$message_age < ' +Same as '$message_age > ' (exiqgrep) -=item -o Same as '$message_age > ' +=item --or -=item -z Same as '$deliver_freeze' +Display messages matching any criteria -=item -x Same as '!$deliver_freeze' +=item -R -=item -c Display count of matches only +Same as --reverse (exiqgrep) -=item -l Display in long format (default) +=item -r -=item -i Display message IDs only +Same as '$recipients =~ //' (exiqgrep). Note that this preserves the default case sensitivity of exiqgrep's interface. -=item -b Display brief format only +=item --random -Please see the 'exiqgrep' documentation for more details on the behaviour and output format produced by these options +Display messages in random order -=item +=item --reverse -The criteria are used to determine whether or not a given message should be displayed. The criteria are built using variables containing information about the individual messages (see VARIABLES section for list and descriptions of available variables). Each criterion is evaluated for each message in the spool and if all (by default) criteria match or (if --or option is specified) any criterion matches, the message is displayed. See VARIABLE TYPES for explanation of types of variables and the evaluations that can be performed on them and EXAMPLES section for complete examples. +Display messages in reverse order -The format of a criterion is explained in detail below, but a key point to make is that the variable being compared must always be on the left side of the comparison. +=item -s -If no criteria are provided all messages in the queue are displayed (in this case the output of exipick should be identical to the output of 'exim -bp') +Same as '$shown_message_size eq ' (exiqgrep) -=item --help +=item --spool -This screen. +Set the path to the exim spool to use. This value will have the argument to --input or 'input' appended, or be ignored if --input is a full path. -=item --version +=item --show-rules -Version info. +Show the internal representation of each criterion specified -=back +=item --show-tests -=head1 VARIABLE TYPES +Show the result of each criterion on each message -Although there are variable types defined, they are defined only by the type of data that gets put into them. They are internally typeless. Because of this it is perfectly legal to perform a numeric comparison against a string variable, although the results will probably be meaningless. +=item --show-vars [,...] -=over 4 +Show the value for for each displayed message. will be a regular expression if it begins with a circumflex. -=item NUMERIC +=item --size -Variable of the numeric type can be of integer or float. Valid comparisons are <, <=, >, >=, ==, and !=. +Show the total bytes used by each displayed message -The numbers specified in the criteria can have a suffix of d, h, m, s, M, K, or B, in which case the number will be mulitplied by 86400, 3600, 60, 1, 1048576, 1024, or 1 respectively. These suffixes are case sensitive. While these are obviously designed to aid in date and size calculations, they are not restricted to variables of their respective types. That is, though it's odd it's legal to create a criterion of a message being around for 3 kiloseconds: '$message_age >= 3K'. +=item --thaw -=item BOOLEAN +Read queue information cached from a previous --freeze run -Variables of the boolean type are very easy to use in criteria. The format is either the variable by itself or the variable negated with a ! sign. For instance, '$deliver_freeze' matches if the message in question is frozen, '!$deliver_freeze' matches if message is not frozen. +=item --sort [,...] -=item STRING +Display matching messages sorted according to + +=item --unsorted + +Do not apply any sorting to output + +=item --version + +Display the version of this command + +=item -x + +Same as '!$deliver_freeze' (exiqgrep) -String variables are basically defined as those that are neither numeric nor boolean and can contain any data. The string operators are =, eq, ne, =~, and !~. With the exception of '=', the operators all match the functionality of the like-named perl operators. +=item -y -The simplest form is a bare string regular expression, represented by the operator '='. The value used for the comparison will be evaluated as a regular expression and can be as simple or as complex as desired. For instance '$sender_helo_name = example' on the simple end or '$sender_helo_name = ^aol\.com$' on the more complex end. This comparison is caseless by default, but see the --caseful option to change this. +Same as '$message_age < ' (exiqgrep) -Slightly more complex is the string comparison with the operators 'eq' and 'ne' for equal and not equal, respectively. '$sender_helo_name eq hotmail.com' is true for messages with the exact helo string "hotmail.com", while '$sender_helo_name ne hotmail.com' is true for any message with a helo string other than "hotmail.com". +=item -z -The most complex and the most flexible format are straight regular expressions with the operators '=~' and '!~'. The value in the criteria is expected to be a correctly formatted perl regular expression B. The criterion '$sender_helo_name !~ /^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$/' matches for any message which does not have an IP address for its helo string. +Same as '$deliver_freeze' (exiqgrep) =back -=head1 VARIABLES +=head1 CRITERIA -With a few exceptions the available variables match Exim's internal expansion variables in both name and exact contents. There are a few notable additions and format deviations which are noted below. Although a brief explanation is offered below, Exim's spec.txt should be consulted for full details. It is important to remember that not every variable will be defined for every message. For example, $sender_host_port is not defined for messages not received from a remote host. +Exipick decides which messages to display by applying a test against each message. The rules take the general form of 'VARIABLE OPERATOR VALUE'. For example, '$message_age > 60'. When exipick is deciding which messages to display, it checks the $message_age variable for each message. If a message's age is greater than 60, the message will be displayed. If the message's age is 60 or less seconds, it will not be displayed. + +Multiple criteria can be used. The order they are specified does not matter. By default all criteria must evaluate to true for a message to be displayed. If the --or option is used, a message is displayed as long as any of the criteria evaluate to true. -In the list below, '.' denotes standard messages with contents matching Exim's variable, '#' denotes standard variables with non-standard contents, and '+' denotes a non-standard variable. +See the VARIABLES and OPERATORS sections below for more details -=head2 Boolean variables +=head1 OPERATORS =over 4 -=item + $allow_unqualified_recipient +=item BOOLEAN -TRUE if unqualified recipient addresses are permitted in header lines. +Boolean variables are checked simply by being true or false. There is no real operator except negation. Examples of valid boolean tests: + '$deliver_freeze' + '!$deliver_freeze' -=item + $allow_unqualified_sender +=item NUMERIC -TRUE if unqualified sender addresses are permitted in header lines. +Valid comparisons are <, <=, >, >=, ==, and !=. Numbers can be integers or floats. Any number in a test suffixed with d, h, m, s, M, K, or B will be multiplied by 86400, 3600, 60, 1, 1048576, 1024, or 1 respectively. Examples of valid numeric tests: + '$message_age >= 3d' + '$local_interface == 587' + '$message_size < 30K' -=item + $deliver_freeze +=item STRING -TRUE if the message is currently frozen. +The string operators are =, eq, ne, =~, and !~. With the exception of '=', the operators all match the functionality of the like-named perl operators. eq and ne match a string exactly. !~, =~, and = apply a perl regular expression to a string. The '=' operator behaves just like =~ but you are not required to place // around the regular expression. Examples of valid string tests: + '$received_protocol eq esmtp' + '$sender_address = example.com' + '$each_recipients =~ /^a[a-z]{2,3}@example.com$/' -=item . $first_delivery +=item NEGATION -TRUE if the message has never been deferred. +There are many ways to negate tests, each having a reason for existing. Many tests can be negated using native operators. For instance, >1 is the opposite of <=1 and eq and ne are opposites. In addition, each individual test can be negated by adding a ! at the beginning of the test. For instance, '!$acl_m1 =~ /^DENY$/' is the same as '$acl_m1 !~ /^DENY$/'. Finally, every test can be specified by using the command line argument --not. This is functionally equivalent to adding a ! to the beginning of every test. -=item . $manually_thawed +=back -TRUE when the message has been manually thawed. +=head1 VARIABLES -=item + $dont_deliver +With a few exceptions the available variables match Exim's internal expansion variables in both name and exact contents. There are a few notable additions and format deviations which are noted below. Although a brief explanation is offered below, Exim's spec.txt should be consulted for full details. It is important to remember that not every variable will be defined for every message. For example, $sender_host_port is not defined for messages not received from a remote host. -TRUE if, under normal circumstances, Exim will not try to deliver the message. +Internally, all variables are represented as strings, meaning any operator will work on any variable. This means that '$sender_host_name > 4' is a legal criterion, even if it does not produce meaningful results. Variables in the list below are marked with a 'type' to help in choosing which types of operators make sense to use. -=item . $host_lookup_deferred + Identifiers + B - Boolean variables + S - String variables + N - Numeric variables + . - Standard variable matching Exim's content definition + # - Standard variable, contents differ from Exim's definition + + - Non-standard variable -TRUE if there was an attempt to look up the host's name from its IP address, but an error occurred that during the attempt. +=over 4 -=item . $host_lookup_failed +=item S . $acl_c0-$acl_c9, $acl_m0-$acl_m9 -TRUE if there was an attempt to look up the host's name from its IP address, but the attempt returned a negative result. +User definable variables. -=item + $local_error_message +=item B + $allow_unqualified_recipient -TRUE if the message is a locally-generated error message. +TRUE if unqualified recipient addresses are permitted in header lines. -=item + $sender_local +=item B + $allow_unqualified_sender -TRUE if the message was locally generated. +TRUE if unqualified sender addresses are permitted in header lines. -=item + $sender_set_untrusted +=item S . $authenticated_id -TRUE if the envelope sender of this message was set by an untrusted local caller. +Optional saved information from authenticators, or the login name of the calling process for locally submitted messages. -=item . $tls_certificate_verified +=item S . $authenticated_sender -TRUE if a TLS certificate was verified when the message was received. +The value of AUTH= param for smtp messages, or a generated value from the calling processes login and qualify domain for locally submitted messages. -=back +=item S . $bheader_*, $bh_* -=head2 Numeric variables +Value of the header(s) with the same name with any RFC2047 words decoded if present. See section 11.5 of Exim's spec.txt for full details. -=over 4 +=item S + $bmi_verdicts + +The verdict string provided by a Brightmail content scan -=item . $body_linecount +=item N . $body_linecount The number of lines in the message's body. -=item . $body_zerocount +=item N . $body_zerocount The number of binary zero bytes in the message's body. -=item + $deliver_frozen_at +=item S + $data_path -The epoch time at which message was frozen. +The path to the body file's location in the filesystem. -=item . $interface_port +=item B + $deliver_freeze -The local port number if network-originated messages. +TRUE if the message is currently frozen. -=item . $message_age +=item N + $deliver_frozen_at -The number of seconds since the message was received. +The epoch time at which message was frozen. -=item . $message_body_size +=item B + $dont_deliver -The size of the body in bytes. +TRUE if, under normal circumstances, Exim will not try to deliver the message. -=item . $message_linecount +=item S + $each_recipients -The number of lines in the entire message (body and headers). +This is a psuedo variable which allows you to apply a test against each address in $recipients individually. Whereas '$recipients =~ /@aol.com/' will match if any recipient address contains aol.com, '$each_recipients =~ /@aol.com$/' will only be true if every recipient matches that pattern. Note that this obeys --and or --or being set. Using it with --or is very similar to just matching against $recipients, but with the added benefit of being able to use anchors at the beginning and end of each recipient address. -=item . $message_size +=item S + $each_recipients_del -The size of the message in bytes. +Like $each_recipients, but for $recipients_del -=item . $originator_gid +=item S + $each_recipients_undel -The group id under which the process that called Exim was running as when the message was received. +Like $each_recipients, but for $recipients_undel -=item . $originator_uid +=item B . $first_delivery -The user id under which the process that called Exim was running as when the message was received. +TRUE if the message has never been deferred. -=item . $received_count +=item S . $header_*, $h_* -The number of Received: header lines in the message. +This will always match the contents of the corresponding $bheader_* variable currently (the same behaviour Exim displays when iconv is not installed). -=item . $received_time +=item S + $header_path -The epoch time at which the message was received. +The path to the header file's location in the filesystem. -=item . $recipients_count +=item B . $host_lookup_deferred -The number of envelope recipients for the message. +TRUE if there was an attempt to look up the host's name from its IP address, but an error occurred that during the attempt. -=item + $recipients_del_count +=item B . $host_lookup_failed -The number of envelope recipients for the message which have already been delivered. Note that this is the count of original recipients to which the message has been delivered. It does not include generated addresses so it is possible that this number will be less than the number of addresses in the recipients_del string. +TRUE if there was an attempt to look up the host's name from its IP address, but the attempt returned a negative result. -=item + $recipients_undel_count +=item B + $local_error_message -The number of envelope recipients for the message which have not yet been delivered. +TRUE if the message is a locally-generated error message. -=item . $sender_host_port +=item S . $local_scan_data -The port number that was used on the remote host for network-originated messages. +The text returned by the local_scan() function when a message is received. -=item + $warning_count +=item B . $manually_thawed -The number of delay warnings which have been sent for this message. +TRUE when the message has been manually thawed. -=back +=item N . $max_received_linelength -=head2 String variables +The number of bytes in the longest line that was received as part of the message, not counting line termination characters. -=over 4 +=item N . $message_age -=item . $acl_c0-$acl_c9, $acl_m0-$acl_m9 +The number of seconds since the message was received. -User definable variables. +=item S # $message_body -=item . $authenticated_id +The message's body. Unlike Exim's variable of the same name, this variable contains the entire message body. Newlines and nulls are replaced by spaces. -Optional saved information from authenticators, or the login name of the calling process for locally submitted messages. +=item B + $message_body_missing -=item . $authenticated_sender +TRUE is a message's spool data file (-D file) is missing or unreadable. -The value of AUTH= param for smtp messages, or a generated value from the calling processes login and qualify domain for locally submitted messages. +=item N . $message_body_size -=item + $bmi_verdicts +The size of the body in bytes. -I honestly don't know what the format of this variable is. It only exists if you have Exim compiled with WITH_CONTENT_SCAN and EXPERIMENTAL_BRIGHTMAIL (and, you know, pay Symantec/Brightmail a bunch of money for the client libs and a server to use them with). +=item S . $message_exim_id, $message_id -=item + $each_recipients +The unique message id that is used by Exim to identify the message. $message_id is deprecated as of Exim 4.53. + +=item S . $message_headers -This is a psuedo variable which allows you to apply a criterion against each address in $recipients individually. This allows you to create criteria against which every individual recipient is tested. For instance, '$recipients =~ /aol.com/' will match if any of the recipient addresses contain the string "aol.com". However, with the criterion '$each_recipients =~ /@aol.com$/', a message will only match if B recipient matches that pattern. Note that this obeys --and or --or being set. Using it with --or is very similar to just matching against $recipients, but with the added benefit of being able to use anchors at the beginning and end of each recipient address. +A concatenation of all the header lines except for lines added by routers or transports. RFC2047 decoding is performed -=item + $each_recipients_del +=item S . $message_headers_raw -Like $each_recipients, but for the $recipients_del variable. +A concatenation of all the header lines except for lines added by routers or transports. No decoding or translation is performed. -=item + $each_recipients_undel +=item N . $message_linecount -Like $each_recipients, but for the $recipients_undel variable. +The number of lines in the entire message (body and headers). -=item # $header_* +=item N . $message_size -The value of the same named message header, for example header_to or header_reply-to. These variables are really closer to Exim's rheader_* variables, with the exception that leading and trailing space is removed. +The size of the message in bytes. -=item . $interface_address +=item N . $originator_gid -The address of the local IP interface for network-originated messages. +The group id under which the process that called Exim was running as when the message was received. -=item . $local_scan_data +=item S + $originator_login -The text returned by the local_scan() function when a message is received. +The login of the process which called Exim. -=item # $message_body +=item N . $originator_uid -The message's body. Unlike Exim's variable of the same name, this variable contains the entire message body. The logic behind this is that the message body is not read unless it is specifically referenced, so under normal circumstances it is not a penalty, but when you need the entire body you need the entire body. Like Exim's copy, newlines and nulls are replaced by spaces. +The user id under which the process that called Exim was running as when the message was received. -=item . $message_headers +=item S . $received_ip_address, $interface_address -A concatenation of all the header lines except for lines added by routers or transports. +The address of the local IP interface for network-originated messages. $interface_address is deprecated as of Exim 4.64 -=item . $message_exim_id, $message_id +=item N . $received_port, $interface_port -The unique message id that is used by Exim to identify the message. $message_id is deprecated as of Exim 4.53. +The local port number if network-originated messages. $interface_port is deprecated as of Exim 4.64 -=item + $originator_login +=item N . $received_count -The login of the process which called Exim. +The number of Received: header lines in the message. -=item . $received_protocol +=item S . $received_protocol The name of the protocol by which the message was received. -=item # $recipients +=item N . $received_time + +The epoch time at which the message was received. + +=item S # $recipients + +The list of envelope recipients for a message. Unlike Exim's version, this variable always contains every recipient of the message. The recipients are separated by a comma and a space. See also $each_recipients. + +=item N . $recipients_count + +The number of envelope recipients for the message. -The list of envelope recipients for a message. Unlike Exim's version, this variable always contains every envelope recipient of the message. The recipients are separated by a comma and a space. +=item S + $recipients_del -=item + $recipients_del +The list of delivered envelope recipients for a message. This non-standard variable is in the same format as $recipients and contains the list of already-delivered recipients including any generated addresses. See also $each_recipients_del. -The list of delivered envelope recipients for a message. This non-standard variable is in the same format as recipients and contains the list of already-delivered recipients including any generated addresses. +=item N + $recipients_del_count -=item + $recipients_undel +The number of envelope recipients for the message which have already been delivered. Note that this is the count of original recipients to which the message has been delivered. It does not include generated addresses so it is possible that this number will be less than the number of addresses in the $recipients_del string. -The list of undelivered envelope recipients for a message. This non-standard variable is in the same format as recipients and contains the list of undelivered recipients. +=item S + $recipients_undel + +The list of undelivered envelope recipients for a message. This non-standard variable is in the same format as $recipients and contains the list of undelivered recipients. See also $each_recipients_undel. + +=item N + $recipients_undel_count + +The number of envelope recipients for the message which have not yet been delivered. -=item . $reply_address +=item S . $reply_address The contents of the Reply-To: header line if one exists and it is not empty, or otherwise the contents of the From: header line. -=item . $sender_address +=item S . $rheader_*, $rh_* + +The value of the message's header(s) with the same name. See section 11.5 of Exim's spec.txt for full description. + +=item S . $sender_address The sender's address that was received in the message's envelope. For bounce messages, the value of this variable is the empty string. -=item . $sender_address_domain +=item S . $sender_address_domain The domain part of $sender_address. -=item . $sender_address_local_part +=item S . $sender_address_local_part The local part of $sender_address. -=item . $sender_helo_name +=item S . $sender_helo_name The HELO or EHLO value supplied for smtp or bsmtp messages. -=item . $sender_host_address +=item S . $sender_host_address The remote host's IP address. -=item . $sender_host_authenticated +=item S . $sender_host_authenticated The name of the authenticator driver which successfully authenticated the client from which the message was received. -=item . $sender_host_name +=item S . $sender_host_name The remote host's name as obtained by looking up its IP address. -=item . $sender_ident - -The identification received in response to an RFC 1413 request for remote messages, the login name of the user that called Exim for locally generated messages. - -=item + $shown_message_size +=item N . $sender_host_port -This non-standard variable contains the formatted size string. That is, for a message whose $message_size is 66566 bytes, $shown_message_size is 65K. - -=item . $smtp_active_hostname - -The value of the active host name when the message was received, as specified by the "smtp_active_hostname" option. +The port number that was used on the remote host for network-originated messages. -=item . $spam_score +=item S . $sender_ident -The spam score of the message, for example '3.4' or '30.5'. (Requires exiscan or WITH_CONTENT_SCAN) +The identification received in response to an RFC 1413 request for remote messages, the login name of the user that called Exim for locally generated messages. -=item . $spam_score_int +=item B + $sender_local -The spam score of the message, multiplied by ten, as an integer value. For instance '34' or '305'. (Requires exiscan or WITH_CONTENT_SCAN) +TRUE if the message was locally generated. -=item . $tls_cipher +=item B + $sender_set_untrusted -The cipher suite that was negotiated for encrypted SMTP connections. +TRUE if the envelope sender of this message was set by an untrusted local caller. -=item . $tls_peerdn +=item S + $shown_message_size -The value of the Distinguished Name of the certificate if Exim is configured to request one. +This non-standard variable contains the formatted size string. That is, for a message whose $message_size is 66566 bytes, $shown_message_size is 65K. -=back +=item S . $smtp_active_hostname -=head1 EXAMPLES +The value of the active host name when the message was received, as specified by the "smtp_active_hostname" option. -=over 4 +=item S . $spam_score -=item exipick '$deliver_freeze' +The spam score of the message, for example '3.4' or '30.5'. (Requires exiscan or WITH_CONTENT_SCAN) -Display only frozen messages. +=item S . $spam_score_int -=item exipick '$received_protocol eq asmtp' '$message_age < 20m' +The spam score of the message, multiplied by ten, as an integer value. For instance '34' or '305'. (Requires exiscan or WITH_CONTENT_SCAN) -Display only messages which were delivered over an authenticated smtp session in the last 20 minutes. +=item B . $tls_certificate_verified -=item exipick -bpc '$message_size > 200K' +TRUE if a TLS certificate was verified when the message was received. -Display a count of messages in the queue which are over 200 kilobytes in size. +=item S . $tls_cipher -=item exipick -or '$sender_helo_name =~ /^\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}$/' '$sender_helo_name = _' +The cipher suite that was negotiated for encrypted SMTP connections. -Display message which have a HELO string which either is an IP address or contains an underscore. +=item S . $tls_peerdn -=back +The value of the Distinguished Name of the certificate if Exim is configured to request one -=head1 REQUIREMENTS +=item S . $tls_sni -None that I know of, except an Exim installation. Your life will also be a lot easier if you set $spool at the top of the script to your install's spool directory (assuming this was not done automatically by the Exim install process). +The value of the Server Name Indication TLS extension sent by a client, if one was sent. -=head1 ACKNOWLEDGEMENTS +=item N + $warning_count -Although I conceived of the concept for this program independently, the name 'exipick' was taken from the Exim WishList and was suggested by Jeffrey Goldberg. +The number of delay warnings which have been sent for this message. -Thank you to Philip Hazel for writing Exim. Of course this program exists because of Exim, but more specifically the message parsing code is based on Exim's and some of this documentation was copy/pasted from Exim's. +=back =head1 CONTACT