first commit

build/node_modules/parse5/lib/tokenization/location_info_mixin.js

@@ -0,0 +1,80 @@
+'use strict';
+
+exports.assign = function (tokenizer) {
+    //NOTE: obtain Tokenizer proto this way to avoid module circular references
+    var tokenizerProto = Object.getPrototypeOf(tokenizer);
+
+    tokenizer.tokenStartLoc = -1;
+
+    //NOTE: add location info builder method
+    tokenizer._attachLocationInfo = function (token) {
+        token.location = {
+            start: this.tokenStartLoc,
+            end: -1
+        };
+    };
+
+    //NOTE: patch token creation methods and attach location objects
+    tokenizer._createStartTagToken = function (tagNameFirstCh) {
+        tokenizerProto._createStartTagToken.call(this, tagNameFirstCh);
+        this._attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createEndTagToken = function (tagNameFirstCh) {
+        tokenizerProto._createEndTagToken.call(this, tagNameFirstCh);
+        this._attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createCommentToken = function () {
+        tokenizerProto._createCommentToken.call(this);
+        this._attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createDoctypeToken = function (doctypeNameFirstCh) {
+        tokenizerProto._createDoctypeToken.call(this, doctypeNameFirstCh);
+        this._attachLocationInfo(this.currentToken);
+    };
+
+    tokenizer._createCharacterToken = function (type, ch) {
+        tokenizerProto._createCharacterToken.call(this, type, ch);
+        this._attachLocationInfo(this.currentCharacterToken);
+    };
+
+    //NOTE: patch token emission methods to determine end location
+    tokenizer._emitCurrentToken = function () {
+        //NOTE: if we have pending character token make it's end location equal to the
+        //current token's start location.
+        if (this.currentCharacterToken)
+            this.currentCharacterToken.location.end = this.currentToken.location.start;
+
+        this.currentToken.location.end = this.preprocessor.pos + 1;
+        tokenizerProto._emitCurrentToken.call(this);
+    };
+
+    tokenizer._emitCurrentCharacterToken = function () {
+        //NOTE: if we have character token and it's location wasn't set in the _emitCurrentToken(),
+        //then set it's location at the current preprocessor position
+        if (this.currentCharacterToken && this.currentCharacterToken.location.end === -1) {
+            //NOTE: we don't need to increment preprocessor position, since character token
+            //emission is always forced by the start of the next character token here.
+            //So, we already have advanced position.
+            this.currentCharacterToken.location.end = this.preprocessor.pos;
+        }
+
+        tokenizerProto._emitCurrentCharacterToken.call(this);
+    };
+
+    //NOTE: patch initial states for each mode to obtain token start position
+    Object.keys(tokenizerProto.MODE)
+
+        .map(function (modeName) {
+            return tokenizerProto.MODE[modeName];
+        })
+
+        .forEach(function (state) {
+            tokenizer[state] = function (cp) {
+                this.tokenStartLoc = this.preprocessor.pos;
+                tokenizerProto[state].call(this, cp);
+            };
+        });
+};