GH#6; fix handling of regex character classes, and / detection.

Inside of a regex, an unescaped `/` may be present inside of a character
class definition.

Author: bleargh45

Changes

@@ -7,6 +7,8 @@ Revision history for Perl extension JavaScript::Minifier::XS.
     - optimize whitespace collapsing
     - GH#3 / RT#108682; fix whitespace reduction at end of preserved line
       comment.  Thanks to Dan Goodliffe
+    - GH#6; fix unescaped slash in character set, inside of a regex, with thanks
+      to @faf
 
 0.13      2020-12-30 21:46:29-08:00 America/Vancouver
     - POD cleanups; spelling, SYNOPSIS

XS.xs

@@ -298,6 +298,7 @@ void _JsExtractLiteral(JsDoc* doc, Node* node) {
     const char* buf = doc->buffer;
     size_t offset   = doc->offset;
     char delimiter  = buf[offset];
+    bool in_char_class = 0;
     /* skip start of literal */
     offset ++;
     /* search for end of literal */
@@ -306,12 +307,24 @@ void _JsExtractLiteral(JsDoc* doc, Node* node) {
             /* escaped character; skip */
             offset ++;
         }
-        else if (buf[offset] == delimiter) {
-            const char* start = buf + doc->offset;
-            size_t length     = offset - doc->offset + 1;
-            JsSetNodeContents(node, start, length);
-            node->type = NODE_LITERAL;
-            return;
+        else {
+            /* if in a regex, track if we're in a character class */
+            if (delimiter == '/') {
+                if ((buf[offset] == '[') && !in_char_class) {
+                    in_char_class = 1;
+                }
+                if ((buf[offset] == ']') && in_char_class) {
+                    in_char_class = 0;
+                }
+            }
+            /* if we have found the end of the literal, store it */
+            if ((buf[offset] == delimiter) && !in_char_class) {
+                const char* start = buf + doc->offset;
+                size_t length     = offset - doc->offset + 1;
+                JsSetNodeContents(node, start, length);
+                node->type = NODE_LITERAL;
+                return;
+            }
         }
         /* move onto next character */
         offset ++;

t/minify.t

@@ -236,6 +236,29 @@ subtest 'regexp escaped "/" is not a comment' => sub {
   is $got, $expect;
 };
 
+###############################################################################
+# GH#6; regexps can contain an unescaped slash inside a character set
+subtest 'regexp with unescaped slash in character set' => sub {
+  my $given  = q|var a = ""; var re = /[/"]/i; console.log(re.test(a));|;
+  my $expect = q|var a="";var re=/[/"]/i;console.log(re.test(a));|;
+  my $got    = minify($given);
+  is $got, $expect;
+};
+
+subtest 'regexp with escaped slash in character set' => sub {
+  my $given  = q|var a = ""; var re = /[\/"]/i; console.log(re.test(a));|;
+  my $expect = q|var a="";var re=/[\/"]/i;console.log(re.test(a));|;
+  my $got    = minify($given);
+  is $got, $expect;
+};
+
+subtest 'regexp with not really a character set' => sub {
+  my $given  = q|var a = ""; var re = /\[/i; console.log(re.test(a));|;
+  my $expect = q|var a="";var re=/\[/i;console.log(re.test(a));|;
+  my $got    = minify($given);
+  is $got, $expect;
+};
+
 ###############################################################################
 # should be able to return a regex from a function
 subtest 'return regex from function' => sub {