From 0bd7ac4e24ce9e81232545b69ce8b98db605984b Mon Sep 17 00:00:00 2001
From: Dirk Koopman <djk@tobit.co.uk>
Date: Mon, 1 Jun 2020 15:28:45 +0100
Subject: [PATCH] final(!) issued version of sh/dx

---
 Changes        |  6 ++++++
 cmd/show/dx.pl | 38 +++++++++++++++++++++-----------------
 perl/Filter.pm | 27 +++++++++++++++++----------
 3 files changed, 44 insertions(+), 27 deletions(-)

diff --git a/Changes b/Changes
index 71bf512a..9b412288 100644
--- a/Changes
+++ b/Changes
@@ -1,3 +1,9 @@
+01Jun20=======================================================================
+1. Fix sh/dx iota and qra. This completes the conversion of sh/dx's limited
+   "old style" parsing to using Filter style parsing with something that 
+   allows expressions include brackets ( and ) as well as the 'not' keyword.
+   NOTE: the precedence rules are the same as perl's '!', '||' and '&&' 
+         operators.  
 31May20=======================================================================
 1. Improve links command layout slightly.
 2. Issue an *accurate* UPGRADE.mojo with all the new packages included! 
diff --git a/cmd/show/dx.pl b/cmd/show/dx.pl
index 07dbed15..c9d85c4b 100644
--- a/cmd/show/dx.pl
+++ b/cmd/show/dx.pl
@@ -71,21 +71,21 @@ sub handle
 		}
 		if (lc $f eq 'iota') {
 			my $doiota;
-			if (@list && $list[0] && (($a, $b) = $list[0] =~ /(AF|AN|NA|SA|EU|AS|OC)-?(\d?\d\d)/oi)) {
+			if (@list && $list[0] && (($a, $b) = $list[0] =~ /(AF|AN|NA|SA|EU|AS|OC)[-\s]?(\d\d?\d?)/i)) {
 				$a = uc $a;
 				$doiota = "\\b$a\[\-\ \]\?$b\\b";
 				shift @list;
 			}
-			$doiota = '\b(IOTA|(AF|AN|NA|SA|EU|AS|OC)[- ]?\d?\d\d)\b' unless $doiota;
-			push @flist, "info {$doiota}";
-			dbg("sh/dx iota") if isdbg('sh/dx');
+			$doiota = '\b(IOTA|(AF|AN|NA|SA|EU|AS|OC)[-\s]?\d?\d\d)\b' unless $doiota;
+			push @flist, 'info', "{$doiota}";
+			dbg("sh/dx iota info {$doiota}") if isdbg('sh/dx');
 			next;
 		}
 		if (lc $f eq 'qra') {
-			my $doqra = uc shift @list if @list && $list[0] =~ /[A-Z][A-Z]\d\d/oi;
+			my $doqra = uc shift @list if @list && $list[0] =~ /[A-Z][A-Z]\d\d/i;
 			$doqra = '\b([A-Z][A-Z]\d\d|[A-Z][A-Z]\d\d[A-Z][A-Z])\b' unless $doqra;
-			push @flist, "info {$doqra}";
-			dbg("sh/dx qra") if isdbg('sh/dx');
+			push @flist, 'info',  "{$doqra}";
+			dbg("sh/dx qra info {$doqra}") if isdbg('sh/dx');
 			next;
 		}
 		if (grep {lc $f eq $_} qw { ( or and not ) }) {
@@ -93,7 +93,7 @@ sub handle
 			dbg("sh/dx operator $f") if isdbg('sh/dx');
 			next;
 		}
-		if (grep {lc $f eq $_} qw(zone byzone by_zone itu byitu by_itu state bystate by_state info on spotter by) ) {
+		if (grep {lc $f eq $_} qw(on freq call info spotter by call_dxcc by_dxcc bydxcc origin call_itu itu call_zone zone  byitu by_itu by_zone byzone call_state state bystate by_state ip) ) {
 			$f =~ s/^by(\w)/by_$1/;
 			push @flist, $f;
 			push @flist, shift @list if @list;
@@ -109,17 +109,21 @@ sub handle
 
 	
 	if ($pre) {
-		$pre .= '*' unless $pre =~ /[\*\?\[]$/o;
-		$pre = shellregex($pre);
-		if ($usesql) {
-			$pre =~ s/\.\*/%/g;
+		# someone (probably me) has forgotten the 'info' keyword
+		if ($pre =~ /^{.*}$/) {
+			push @flist, 'info', $pre;
 		} else {
-			$pre =~ s/\.\*\$$//;
+			$pre .= '*' unless $pre =~ /[\*\?\[]$/o;
+			$pre = shellregex($pre);
+			if ($usesql) {
+				$pre =~ s/\.\*/%/g;
+			} else {
+				$pre =~ s/\.\*\$$//;
+			}
+			$pre .= '$' if $exact;
+			$pre =~ s/\^//;
+			push @flist, 'call', $pre;
 		}
-		$pre .= '$' if $exact;
-		$pre =~ s/\^//;
-		
-		push @flist, 'call', $pre;
 	}
 	
     my $newline = join(' ', @flist);
diff --git a/perl/Filter.pm b/perl/Filter.pm
index fd911182..898b004a 100644
--- a/perl/Filter.pm
+++ b/perl/Filter.pm
@@ -375,12 +375,18 @@ sub parse
 	# check the line for non legal characters
 	dbg("Filter::parse line: '$line'") if isdbg('filter');
 	return ('ill', $dxchan->msg('e19')) if $line !~ /{.*}/ && $line =~ /[^\s\w,_\-\*\/\(\)\$!]/;
+
+	$line = lc $line;
+
+	# disguise regexes
+	$line =~ s/\{(.*)\}/'{'. unpack('H*', $1) . '}'/eg;
+	dbg("Filter parse line after regex check: '$line'") if isdbg('filter');
 	
 	# add some spaces for ease of parsing
 	$line =~ s/([\(\!\)])/ $1 /g;
-	$line = lc $line;
 	
 	my @f = split /\s+/, $line;
+
 	my $conj = ' && ';
 	my $not = "";
 	my $lasttok = '';
@@ -485,14 +491,15 @@ sub parse
 						}
 						if ($fref->[1] eq 'a' || $fref->[1] eq 't') {
 							my @t;
-							for (@val) {
-								s/\*//g;        # remove any trailing *
-								if (/^\{.*\}$/) { # we have a regex 
-									s/^\{//;
-								    s/\}$//;
-									return  ('regex', $dxchan->msg('e38', $_)) unless (qr{$_})
+							foreach my $v (@val) {
+								$v =~ s/\*//g;        # remove any trailing *
+								if (my ($r) = $v =~ /^\{(.*)\}$/) { # we have a regex
+									dbg("Filter::parse regex b: '\{$r\}'") if isdbg('filter'); 
+									$v = pack('H*', $r);
+									dbg("Filter::parse regex a: '$v'") if isdbg('filter'); 
+									return  ('regex', $dxchan->msg('e38', $v)) unless (qr{$v});
 								}
-								push @t, "\$r->[$fref->[2]]=~m{$_}i";
+								push @t, "\$r->[$fref->[2]]=~m{$v}i";
 							}
 							$s .= "(" . join(' || ', @t) . ")";
 						} elsif ($fref->[1] eq 'c') {
@@ -533,9 +540,9 @@ sub parse
 						last;
 					}
 				}
-				return ('unknown', $dxchan->msg('e20', $tok)) unless $found;
+				return (0, $dxchan->msg('e20', $tok)) unless $found;
 			} else {
-				return ('no', $dxchan->msg('filter2', $tok));
+				return (0, $dxchan->msg('filter2', $tok));
 			}
 			$lasttok = $tok;
 		}
-- 
2.43.0