lei q: use git approxidate with d:, dt: and rt: ranges

Instead of having --(sent|received)-(before|after)=s
command-line switches, we'll just try to make sense of argv so
it's usable within parenthesized statements and such.

Given the negligible performance penalty with Inline::C
process spawning, we'll probably wire this up to the
WWW interface, too.

"d:" is for mairix compatibility.  I don't know if "dt:" and
"rt:" will be too useful, but they exist because of IMAP
(and JMAP).

2 files changed, 71 insertions, 8 deletions

diff --git a/lib/PublicInbox/LeiQuery.pm b/lib/PublicInbox/LeiQuery.pm
index 9a6fa718..d637b1ae 100644
--- a/lib/PublicInbox/LeiQuery.pm
+++ b/lib/PublicInbox/LeiQuery.pm
@@ -34,9 +34,10 @@ sub lei_q {
         my @only = @{$opt->{only} // []};
         # --local is enabled by default unless --only is used
         # we'll allow "--only $LOCATION --local"
+        my $sto = $self->_lei_store(1);
+        my $lse = $sto->search;
         if ($opt->{'local'} //= scalar(@only) ? 0 : 1) {
-                my $sto = $self->_lei_store(1);
-                $lxs->prepare_external($sto->search);
+                $lxs->prepare_external($lse);
         }
         if (@only) {
                 for my $loc (@only) {
@@ -107,12 +108,7 @@ no query allowed on command-line with --stdin
                 PublicInbox::InputPipe::consume($self->{0}, \&qstr_add, $self);
                 return;
         }
-        # Consider spaces in argv to be for phrase search in Xapian.
-        # In other words, the users should need only care about
-        # normal shell quotes and not have to learn Xapian quoting.
-        $mset_opt{qstr} = join(' ', map {;
-                /\s/ ? (s/\A(\w+:)// ? qq{$1"$_"} : qq{"$_"}) : $_
-        } @argv);
+        $mset_opt{qstr} = $lse->query_argv_to_string($lse->git, \@argv);
         $lxs->do_query($self);
 }
 
diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index dbae3bc5..f42d70e3 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -8,6 +8,7 @@ use strict;
 use parent qw(Exporter);
 our @EXPORT_OK = qw(retry_reopen int_val get_pct xap_terms);
 use List::Util qw(max);
+use POSIX qw(strftime);
 
 # values for searching, changing the numeric value breaks
 # compatibility with old indices (so don't change them it)
@@ -259,6 +260,72 @@ sub reopen {
         $self; # make chaining easier
 }
 
+# Convert git "approxidate" ranges to something usable with our
+# Xapian indices.  At the moment, Xapian only offers a C++-only API
+# and neither the SWIG nor XS bindings allow us to use custom code
+# to parse dates (and libgit2 doesn't expose git__date_parse, either,
+# so we're running git-rev-parse(1)).
+sub date_range {
+        my ($git, $pfx, $range) = @_;
+        # are we inside a parenthesized statement?
+        my $end = $range =~ s/([\)\s]*)\z// ? $1 : '';
+        my @r = split(/\.\./, $range, 2);
+
+        # expand "d:20101002" => "d:20101002..20101003" and like
+        # n.b. git doesn't do YYYYMMDD w/o '-', it needs YYYY-MM-DD
+        if ($pfx eq 'd') {
+                if (!defined($r[1])) {
+                        $r[0] =~ s/\A([0-9]{4})([0-9]{2})([0-9]{2})\z/$1-$2-$3/;
+                        $r[0] = $git->date_parse($r[0]);
+                        $r[1] = $r[0] + 86400;
+                        for my $x (@r) {
+                                $x = strftime('%Y%m%d', gmtime($x));
+                        }
+                } else {
+                        for my $x (@r) {
+                                next if $x eq '' || $x =~ /\A[0-9]{8}\z/;
+                                $x = strftime('%Y%m%d',
+                                                gmtime($git->date_parse($x)));
+                        }
+                }
+        } elsif ($pfx eq 'dt') {
+                if (!defined($r[1])) { # git needs gaps and not /\d{14}/
+                        $r[0] =~ s/\A([0-9]{4})([0-9]{2})([0-9]{2})
+                                        ([0-9]{2})([0-9]{2})([0-9]{2})\z
+                                /$1-$2-$3 $4:$5:$6/x;
+                        $r[0] = $git->date_parse($r[0]);
+                        $r[1] = $r[0] + 86400;
+                        for my $x (@r) {
+                                $x = strftime('%Y%m%d%H%M%S', gmtime($x));
+                        }
+                } else {
+                        for my $x (@r) {
+                                next if $x eq '' || $x =~ /\A[0-9]{14}\z/;
+                                $x = strftime('%Y%m%d%H%M%S',
+                                                gmtime($git->date_parse($x)));
+                        }
+                }
+        } else { # "rt", let git interpret "YYYY", deal with Y10K later :P
+                for my $x (@r) {
+                        next if $x eq '' || $x =~ /\A[0-9]{5,}\z/;
+                        $x = $git->date_parse($x);
+                }
+                $r[1] //= $r[0] + 86400;
+        }
+        "$pfx:".join('..', @r).$end;
+}
+
+sub query_argv_to_string {
+        my (undef, $git, $argv) = @_;
+        join(' ', map {;
+                if (s!\b(d|rt|dt):(.+)\z!date_range($git, $1, $2)!sge) {
+                        $_;
+                } else {
+                        /\s/ ? (s/\A(\w+:)// ? qq{$1"$_"} : qq{"$_}) : $_
+                }
+        } @$argv);
+}
+
 # read-only
 sub mset {
         my ($self, $query_string, $opts) = @_;