www: improve visibility of coderepos

[public-inbox.git] / lib / PublicInbox / Search.pm

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index aa737d63b5f64b7d30df764b6b34ceaf2e3044b0..c7d52dafc3f4f9e37491aea50c6ac13880ee8b62 100644 (file)
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -13,9 +13,9 @@ use POSIX qw(strftime);
 # values for searching, changing the numeric value breaks
 # compatibility with old indices (so don't change them it)
 use constant {
-       TS => 0, # Received: header in Unix time (IMAP INTERNALDATE)
+       TS => 0, # Received: in Unix time (IMAP INTERNALDATE, JMAP receivedAt)
        YYYYMMDD => 1, # Date: header for searching in the WWW UI
-       DT => 2, # Date: YYYYMMDDHHMMSS
+       DT => 2, # Date: YYYYMMDDHHMMSS (IMAP SENT*, JMAP sentAt)
 
        # added for public-inbox 1.6.0+
        BYTES => 3, # IMAP RFC822.SIZE
@@ -321,30 +321,51 @@ sub date_parse_prepare {
        "$pfx:".join('..', @r).$end;
 }
 
+sub date_parse_finalize {
+       my ($git, $to_parse) = @_;
+       # git-rev-parse can handle any number of args up to system
+       # limits (around (4096*32) bytes on Linux).
+       my @r = $git->date_parse(@$to_parse);
+       my $i;
+       $_[2] =~ s/\0(%[%YmdHMSs]+)([0-9\+]+)\0/strftime($1,
+               gmtime($2 eq '+' ? ($r[$i]+86400) : $r[$i=$2+0]))/sge;
+}
+
 # n.b. argv never has NUL, though we'll need to filter it out
 # if this $argv isn't from a command execution
 sub query_argv_to_string {
        my (undef, $git, $argv) = @_;
        my $to_parse;
        my $tmp = join(' ', map {;
-               if (s!\b(d|rt|dt):([[:print:]]+)\z!date_parse_prepare(
+               if (s!\b(d|rt|dt):(\S+)\z!date_parse_prepare(
                                                $to_parse //= [], $1, $2)!sge) {
                        $_;
+               } elsif (/\s/) {
+                       s/(.*?)\b(\w+:)// ? qq{$1$2"$_"} : qq{"$_"};
                } else {
-                       /\s/ ? (s/\A(\w+:)// ? qq{$1"$_"} : qq{"$_}) : $_
+                       $_
                }
        } @$argv);
-       # git-rev-parse can handle any number of args up to system
-       # limits (around (4096*32) bytes on Linux).
-       if ($to_parse) {
-               my @r = $git->date_parse(@$to_parse);
-               my $i;
-               $tmp =~ s/\0(%[%YmdHMSs]+)([0-9\+]+)\0/strftime($1,
-                       gmtime($2 eq '+' ? ($r[$i]+86400) : $r[$i=$2+0]))/sge;
-       }
+       date_parse_finalize($git, $to_parse, $tmp) if $to_parse;
        $tmp
 }
 
+# this is for the WWW "q=" query parameter and "lei q --stdin"
+# it can't do d:"5 days ago", but it will do d:5.days.ago
+sub query_approxidate {
+       my (undef, $git) = @_; # $_[2] = $query_string (modified in-place)
+       my $DQ = qq<"\x{201c}\x{201d}>; # Xapian can use curly quotes
+       $_[2] =~ tr/\x00/ /; # Xapian doesn't do NUL, we use it as a placeholder
+       my ($terms, $phrase, $to_parse);
+       $_[2] =~ s{([^$DQ]*)([$DQ][^$DQ]*[$DQ])?}{
+               ($terms, $phrase) = ($1, $2);
+               $terms =~ s!\b(d|rt|dt):(\S+)!
+                       date_parse_prepare($to_parse //= [], $1, $2)!sge;
+               $terms.($phrase // '');
+               }sge;
+       date_parse_finalize($git, $to_parse, $_[2]) if $to_parse;
+}
+
 # read-only
 sub mset {
        my ($self, $query_string, $opts) = @_;
@@ -407,18 +428,17 @@ sub _enquire_once { # retry_reopen callback
        my $enquire = $X{Enquire}->new($xdb);
        $enquire->set_query($query);
        $opts ||= {};
-        my $desc = !$opts->{asc};
        my $rel = $opts->{relevance} // 0;
-       if ($rel == -1) { # ORDER BY docid/UID
+       if ($rel == -2) { # ORDER BY docid/UID (highest first)
+               $enquire->set_weighting_scheme($X{BoolWeight}->new);
+               $enquire->set_docid_order($ENQ_DESCENDING);
+       } elsif ($rel == -1) { # ORDER BY docid/UID (lowest first)
                $enquire->set_weighting_scheme($X{BoolWeight}->new);
                $enquire->set_docid_order($ENQ_ASCENDING);
        } elsif ($rel == 0) {
-               $enquire->set_sort_by_value_then_relevance(TS, $desc);
-       } elsif ($rel == -2) {
-               $enquire->set_weighting_scheme($X{BoolWeight}->new);
-               $enquire->set_docid_order($ENQ_DESCENDING);
+               $enquire->set_sort_by_value_then_relevance(TS, !$opts->{asc});
        } else { # rel > 0
-               $enquire->set_sort_by_relevance_then_value(TS, $desc);
+               $enquire->set_sort_by_relevance_then_value(TS, !$opts->{asc});
        }
 
        # `mairix -t / --threads' or JMAP collapseThreads
@@ -461,7 +481,7 @@ sub qparse_new ($) {
        $cb->($qp, $NVRP->new(DT, 'dt:'));
 
        # for IMAP, undocumented for WWW and may be split off go away
-       $cb->($qp, $NVRP->new(BYTES, 'bytes:'));
+       $cb->($qp, $NVRP->new(BYTES, 'z:'));
        $cb->($qp, $NVRP->new(TS, 'rt:'));
        $cb->($qp, $NVRP->new(UID, 'uid:'));