Implament case-insensitive class matching

README.md

@@ -136,7 +136,7 @@ Match exactly one character and return it as a string.
 
 #### [*characters*]
 
-Match one character from a set and return it as a string. The characters in the list can be escaped in exactly the same way as in JavaScript string. The list of characters can also contain ranges (e.g. `[a-z]` means “all lowercase letters”). Preceding the characters with `^` inverts the matched set (e.g. `[^a-z]` means “all character but lowercase letters”).
+Match one character from a set and return it as a string. The characters in the list can be escaped in exactly the same way as in JavaScript string. The list of characters can also contain ranges (e.g. `[a-z]` means “all lowercase letters”). Preceding the characters with `^` inverts the matched set (e.g. `[^a-z]` means “all character but lowercase letters”). Appending `i` right after the right bracket makes the match case-insensitive.
 
 #### *rule*
 

src/emitter.js

@@ -911,7 +911,7 @@ PEG.compiler.emitter = function(ast) {
                   + quoteForRegexpClass(part[1])
                 : quoteForRegexpClass(part);
             }).join('')
-          + ']/';
+          + ']/' + (node.ignoreCase ? 'i' : '');
       } else {
         /*
          * Stupid IE considers regexps /[]/ and /[^]/ syntactically invalid, so

src/parser.js

@@ -2272,7 +2272,7 @@ PEG.parser = (function(){
           return cachedResult.result;
         }
         
-        var result0, result1, result2, result3, result4, result5;
+        var result0, result1, result2, result3, result4, result5, result6;
         var pos0, pos1, pos2;
         
         reportFailures++;
@@ -2322,9 +2322,24 @@ PEG.parser = (function(){
                 }
               }
               if (result3 !== null) {
-                result4 = parse___();
+                if (input.charCodeAt(pos) === 105) {
+                  result4 = "i";
+                  pos += 1;
+                } else {
+                  result4 = null;
+                  if (reportFailures === 0) {
+                    matchFailed("\"i\"");
+                  }
+                }
+                result4 = result4 !== null ? result4 : "";
                 if (result4 !== null) {
-                  result0 = [result0, result1, result2, result3, result4];
+                  result5 = parse___();
+                  if (result5 !== null) {
+                    result0 = [result0, result1, result2, result3, result4, result5];
+                  } else {
+                    result0 = null;
+                    pos = pos1;
+                  }
                 } else {
                   result0 = null;
                   pos = pos1;
@@ -2346,21 +2361,23 @@ PEG.parser = (function(){
           pos = pos1;
         }
         if (result0 !== null) {
-          result0 = (function(inverted, parts) {
+          result0 = (function(inverted, parts, flags) {
               var partsConverted = map(parts, function(part) { return part.data; });
               var rawText = "["
                 + inverted
                 + map(parts, function(part) { return part.rawText; }).join("")
-                + "]";
+                + "]"
+                + flags;
         
               return {
-                type:     "class",
+                type:       "class",
-                inverted: inverted === "^",
+                inverted:   inverted === "^",
-                parts:    partsConverted,
+                ignoreCase: flags === "i",
+                parts:      partsConverted,
                 // FIXME: Get the raw text from the input directly.
-                rawText:  rawText
+                rawText:    rawText
               };
-            })(result0[1], result0[2]);
+            })(result0[1], result0[2], result0[4]);
         }
         if (result0 === null) {
           pos = pos0;

src/parser.pegjs

@@ -235,19 +235,21 @@ simpleSingleQuotedCharacter
   = !("'" / "\\" / eolChar) char_:. { return char_; }
 
 class "character class"
-  = "[" inverted:"^"? parts:(classCharacterRange / classCharacter)* "]" __ {
+  = "[" inverted:"^"? parts:(classCharacterRange / classCharacter)* "]" flags:"i"? __ {
       var partsConverted = map(parts, function(part) { return part.data; });
       var rawText = "["
         + inverted
         + map(parts, function(part) { return part.rawText; }).join("")
-        + "]";
+        + "]"
+        + flags;
 
       return {
-        type:     "class",
+        type:       "class",
-        inverted: inverted === "^",
+        inverted:   inverted === "^",
-        parts:    partsConverted,
+        ignoreCase: flags === "i",
+        parts:      partsConverted,
         // FIXME: Get the raw text from the input directly.
-        rawText:  rawText
+        rawText:    rawText
       };
     }
 

test/compiler-test.js

@@ -246,26 +246,62 @@ test("classes", function() {
   doesNotParse(emptyClassParser, "a");
   doesNotParse(emptyClassParser, "ab");
 
-  var nonEmptyClassParser = PEG.buildParser('start = [ab-d]');
-  parses(nonEmptyClassParser, "a", "a");
-  parses(nonEmptyClassParser, "b", "b");
-  parses(nonEmptyClassParser, "c", "c");
-  parses(nonEmptyClassParser, "d", "d");
-  doesNotParse(nonEmptyClassParser, "");
-  doesNotParse(nonEmptyClassParser, "ab");
-
   var invertedEmptyClassParser = PEG.buildParser('start = [^]');
   doesNotParse(invertedEmptyClassParser, "");
   parses(invertedEmptyClassParser, "a", "a");
   doesNotParse(invertedEmptyClassParser, "ab");
 
-  var invertedNonEmptyClassParser = PEG.buildParser('start = [^ab-d]');
+  var nonEmptyCaseSensitiveClassParser = PEG.buildParser('start = [ab-d]');
-  doesNotParse(invertedNonEmptyClassParser, "a", "a");
+  parses(nonEmptyCaseSensitiveClassParser, "a", "a");
-  doesNotParse(invertedNonEmptyClassParser, "b", "b");
+  parses(nonEmptyCaseSensitiveClassParser, "b", "b");
-  doesNotParse(invertedNonEmptyClassParser, "c", "c");
+  parses(nonEmptyCaseSensitiveClassParser, "c", "c");
-  doesNotParse(invertedNonEmptyClassParser, "d", "d");
+  parses(nonEmptyCaseSensitiveClassParser, "d", "d");
-  doesNotParse(invertedNonEmptyClassParser, "");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "");
-  doesNotParse(invertedNonEmptyClassParser, "ab");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "A");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "B");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "C");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "D");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "e");
+  doesNotParse(nonEmptyCaseSensitiveClassParser, "ab");
+
+  var invertedNonEmptyCaseSensitiveClassParser = PEG.buildParser('start = [^ab-d]');
+  parses(invertedNonEmptyCaseSensitiveClassParser, "A", "A");
+  parses(invertedNonEmptyCaseSensitiveClassParser, "B", "B");
+  parses(invertedNonEmptyCaseSensitiveClassParser, "C", "C");
+  parses(invertedNonEmptyCaseSensitiveClassParser, "D", "D");
+  parses(invertedNonEmptyCaseSensitiveClassParser, "e", "e");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "a", "a");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "b", "b");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "c", "c");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "d", "d");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "");
+  doesNotParse(invertedNonEmptyCaseSensitiveClassParser, "ab");
+
+  var nonEmptyCaseInsensitiveClassParser = PEG.buildParser('start = [ab-d]i');
+  parses(nonEmptyCaseInsensitiveClassParser, "a", "a");
+  parses(nonEmptyCaseInsensitiveClassParser, "b", "b");
+  parses(nonEmptyCaseInsensitiveClassParser, "c", "c");
+  parses(nonEmptyCaseInsensitiveClassParser, "d", "d");
+  parses(nonEmptyCaseInsensitiveClassParser, "A", "A");
+  parses(nonEmptyCaseInsensitiveClassParser, "B", "B");
+  parses(nonEmptyCaseInsensitiveClassParser, "C", "C");
+  parses(nonEmptyCaseInsensitiveClassParser, "D", "D");
+  doesNotParse(nonEmptyCaseInsensitiveClassParser, "");
+  doesNotParse(nonEmptyCaseInsensitiveClassParser, "e");
+  doesNotParse(nonEmptyCaseInsensitiveClassParser, "ab");
+
+  var invertedNonEmptyCaseInsensitiveClassParser = PEG.buildParser('start = [^ab-d]i');
+  parses(invertedNonEmptyCaseInsensitiveClassParser, "e", "e");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "a", "a");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "b", "b");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "c", "c");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "d", "d");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "A", "A");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "B", "B");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "C", "C");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "D", "D");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "");
+  doesNotParse(invertedNonEmptyCaseInsensitiveClassParser, "ab");
 
   /*
    * Test that the parsing position moves forward after successful parsing of

test/parser-test.js

@@ -95,12 +95,13 @@ function any() {
   return { type: "any" };
 }
 
-function klass(inverted, parts, rawText) {
+function klass(inverted, ignoreCase, parts, rawText) {
   return {
-    type:     "class",
+    type:       "class",
-    inverted: inverted,
+    inverted:   inverted,
-    parts:    parts,
+    ignoreCase: ignoreCase,
-    rawText:  rawText
+    parts:      parts,
+    rawText:    rawText
   };
 }
 
@@ -141,7 +142,7 @@ function literalGrammar(literal) {
 }
 
 function classGrammar(inverted, parts, rawText) {
-  return oneRuleGrammar(klass(inverted, parts, rawText));
+  return oneRuleGrammar(klass(inverted, false, parts, rawText));
 }
 
 var anyGrammar = oneRuleGrammar(any());
@@ -417,6 +418,10 @@ test("parses class", function() {
     "start = [a-de-hi-l]",
     classGrammar(false, [["a", "d"], ["e", "h"], ["i", "l"]], "[a-de-hi-l]")
   );
+  parserParses(
+    "start = [a-d]i",
+    oneRuleGrammar(klass(false, true, [["a", "d"]], "[a-d]i"))
+  );
 
   parserParses("start = [a-d]\n", classGrammar(false, [["a", "d"]], "[a-d]"));
 });