1 | var charCodeDef = require('./char-code-definitions');
|
---|
2 | var isDigit = charCodeDef.isDigit;
|
---|
3 | var isHexDigit = charCodeDef.isHexDigit;
|
---|
4 | var isUppercaseLetter = charCodeDef.isUppercaseLetter;
|
---|
5 | var isName = charCodeDef.isName;
|
---|
6 | var isWhiteSpace = charCodeDef.isWhiteSpace;
|
---|
7 | var isValidEscape = charCodeDef.isValidEscape;
|
---|
8 |
|
---|
9 | function getCharCode(source, offset) {
|
---|
10 | return offset < source.length ? source.charCodeAt(offset) : 0;
|
---|
11 | }
|
---|
12 |
|
---|
13 | function getNewlineLength(source, offset, code) {
|
---|
14 | if (code === 13 /* \r */ && getCharCode(source, offset + 1) === 10 /* \n */) {
|
---|
15 | return 2;
|
---|
16 | }
|
---|
17 |
|
---|
18 | return 1;
|
---|
19 | }
|
---|
20 |
|
---|
21 | function cmpChar(testStr, offset, referenceCode) {
|
---|
22 | var code = testStr.charCodeAt(offset);
|
---|
23 |
|
---|
24 | // code.toLowerCase() for A..Z
|
---|
25 | if (isUppercaseLetter(code)) {
|
---|
26 | code = code | 32;
|
---|
27 | }
|
---|
28 |
|
---|
29 | return code === referenceCode;
|
---|
30 | }
|
---|
31 |
|
---|
32 | function cmpStr(testStr, start, end, referenceStr) {
|
---|
33 | if (end - start !== referenceStr.length) {
|
---|
34 | return false;
|
---|
35 | }
|
---|
36 |
|
---|
37 | if (start < 0 || end > testStr.length) {
|
---|
38 | return false;
|
---|
39 | }
|
---|
40 |
|
---|
41 | for (var i = start; i < end; i++) {
|
---|
42 | var testCode = testStr.charCodeAt(i);
|
---|
43 | var referenceCode = referenceStr.charCodeAt(i - start);
|
---|
44 |
|
---|
45 | // testCode.toLowerCase() for A..Z
|
---|
46 | if (isUppercaseLetter(testCode)) {
|
---|
47 | testCode = testCode | 32;
|
---|
48 | }
|
---|
49 |
|
---|
50 | if (testCode !== referenceCode) {
|
---|
51 | return false;
|
---|
52 | }
|
---|
53 | }
|
---|
54 |
|
---|
55 | return true;
|
---|
56 | }
|
---|
57 |
|
---|
58 | function findWhiteSpaceStart(source, offset) {
|
---|
59 | for (; offset >= 0; offset--) {
|
---|
60 | if (!isWhiteSpace(source.charCodeAt(offset))) {
|
---|
61 | break;
|
---|
62 | }
|
---|
63 | }
|
---|
64 |
|
---|
65 | return offset + 1;
|
---|
66 | }
|
---|
67 |
|
---|
68 | function findWhiteSpaceEnd(source, offset) {
|
---|
69 | for (; offset < source.length; offset++) {
|
---|
70 | if (!isWhiteSpace(source.charCodeAt(offset))) {
|
---|
71 | break;
|
---|
72 | }
|
---|
73 | }
|
---|
74 |
|
---|
75 | return offset;
|
---|
76 | }
|
---|
77 |
|
---|
78 | function findDecimalNumberEnd(source, offset) {
|
---|
79 | for (; offset < source.length; offset++) {
|
---|
80 | if (!isDigit(source.charCodeAt(offset))) {
|
---|
81 | break;
|
---|
82 | }
|
---|
83 | }
|
---|
84 |
|
---|
85 | return offset;
|
---|
86 | }
|
---|
87 |
|
---|
88 | // § 4.3.7. Consume an escaped code point
|
---|
89 | function consumeEscaped(source, offset) {
|
---|
90 | // It assumes that the U+005C REVERSE SOLIDUS (\) has already been consumed and
|
---|
91 | // that the next input code point has already been verified to be part of a valid escape.
|
---|
92 | offset += 2;
|
---|
93 |
|
---|
94 | // hex digit
|
---|
95 | if (isHexDigit(getCharCode(source, offset - 1))) {
|
---|
96 | // Consume as many hex digits as possible, but no more than 5.
|
---|
97 | // Note that this means 1-6 hex digits have been consumed in total.
|
---|
98 | for (var maxOffset = Math.min(source.length, offset + 5); offset < maxOffset; offset++) {
|
---|
99 | if (!isHexDigit(getCharCode(source, offset))) {
|
---|
100 | break;
|
---|
101 | }
|
---|
102 | }
|
---|
103 |
|
---|
104 | // If the next input code point is whitespace, consume it as well.
|
---|
105 | var code = getCharCode(source, offset);
|
---|
106 | if (isWhiteSpace(code)) {
|
---|
107 | offset += getNewlineLength(source, offset, code);
|
---|
108 | }
|
---|
109 | }
|
---|
110 |
|
---|
111 | return offset;
|
---|
112 | }
|
---|
113 |
|
---|
114 | // §4.3.11. Consume a name
|
---|
115 | // Note: This algorithm does not do the verification of the first few code points that are necessary
|
---|
116 | // to ensure the returned code points would constitute an <ident-token>. If that is the intended use,
|
---|
117 | // ensure that the stream starts with an identifier before calling this algorithm.
|
---|
118 | function consumeName(source, offset) {
|
---|
119 | // Let result initially be an empty string.
|
---|
120 | // Repeatedly consume the next input code point from the stream:
|
---|
121 | for (; offset < source.length; offset++) {
|
---|
122 | var code = source.charCodeAt(offset);
|
---|
123 |
|
---|
124 | // name code point
|
---|
125 | if (isName(code)) {
|
---|
126 | // Append the code point to result.
|
---|
127 | continue;
|
---|
128 | }
|
---|
129 |
|
---|
130 | // the stream starts with a valid escape
|
---|
131 | if (isValidEscape(code, getCharCode(source, offset + 1))) {
|
---|
132 | // Consume an escaped code point. Append the returned code point to result.
|
---|
133 | offset = consumeEscaped(source, offset) - 1;
|
---|
134 | continue;
|
---|
135 | }
|
---|
136 |
|
---|
137 | // anything else
|
---|
138 | // Reconsume the current input code point. Return result.
|
---|
139 | break;
|
---|
140 | }
|
---|
141 |
|
---|
142 | return offset;
|
---|
143 | }
|
---|
144 |
|
---|
145 | // §4.3.12. Consume a number
|
---|
146 | function consumeNumber(source, offset) {
|
---|
147 | var code = source.charCodeAt(offset);
|
---|
148 |
|
---|
149 | // 2. If the next input code point is U+002B PLUS SIGN (+) or U+002D HYPHEN-MINUS (-),
|
---|
150 | // consume it and append it to repr.
|
---|
151 | if (code === 0x002B || code === 0x002D) {
|
---|
152 | code = source.charCodeAt(offset += 1);
|
---|
153 | }
|
---|
154 |
|
---|
155 | // 3. While the next input code point is a digit, consume it and append it to repr.
|
---|
156 | if (isDigit(code)) {
|
---|
157 | offset = findDecimalNumberEnd(source, offset + 1);
|
---|
158 | code = source.charCodeAt(offset);
|
---|
159 | }
|
---|
160 |
|
---|
161 | // 4. If the next 2 input code points are U+002E FULL STOP (.) followed by a digit, then:
|
---|
162 | if (code === 0x002E && isDigit(source.charCodeAt(offset + 1))) {
|
---|
163 | // 4.1 Consume them.
|
---|
164 | // 4.2 Append them to repr.
|
---|
165 | code = source.charCodeAt(offset += 2);
|
---|
166 |
|
---|
167 | // 4.3 Set type to "number".
|
---|
168 | // TODO
|
---|
169 |
|
---|
170 | // 4.4 While the next input code point is a digit, consume it and append it to repr.
|
---|
171 |
|
---|
172 | offset = findDecimalNumberEnd(source, offset);
|
---|
173 | }
|
---|
174 |
|
---|
175 | // 5. If the next 2 or 3 input code points are U+0045 LATIN CAPITAL LETTER E (E)
|
---|
176 | // or U+0065 LATIN SMALL LETTER E (e), ... , followed by a digit, then:
|
---|
177 | if (cmpChar(source, offset, 101 /* e */)) {
|
---|
178 | var sign = 0;
|
---|
179 | code = source.charCodeAt(offset + 1);
|
---|
180 |
|
---|
181 | // ... optionally followed by U+002D HYPHEN-MINUS (-) or U+002B PLUS SIGN (+) ...
|
---|
182 | if (code === 0x002D || code === 0x002B) {
|
---|
183 | sign = 1;
|
---|
184 | code = source.charCodeAt(offset + 2);
|
---|
185 | }
|
---|
186 |
|
---|
187 | // ... followed by a digit
|
---|
188 | if (isDigit(code)) {
|
---|
189 | // 5.1 Consume them.
|
---|
190 | // 5.2 Append them to repr.
|
---|
191 |
|
---|
192 | // 5.3 Set type to "number".
|
---|
193 | // TODO
|
---|
194 |
|
---|
195 | // 5.4 While the next input code point is a digit, consume it and append it to repr.
|
---|
196 | offset = findDecimalNumberEnd(source, offset + 1 + sign + 1);
|
---|
197 | }
|
---|
198 | }
|
---|
199 |
|
---|
200 | return offset;
|
---|
201 | }
|
---|
202 |
|
---|
203 | // § 4.3.14. Consume the remnants of a bad url
|
---|
204 | // ... its sole use is to consume enough of the input stream to reach a recovery point
|
---|
205 | // where normal tokenizing can resume.
|
---|
206 | function consumeBadUrlRemnants(source, offset) {
|
---|
207 | // Repeatedly consume the next input code point from the stream:
|
---|
208 | for (; offset < source.length; offset++) {
|
---|
209 | var code = source.charCodeAt(offset);
|
---|
210 |
|
---|
211 | // U+0029 RIGHT PARENTHESIS ())
|
---|
212 | // EOF
|
---|
213 | if (code === 0x0029) {
|
---|
214 | // Return.
|
---|
215 | offset++;
|
---|
216 | break;
|
---|
217 | }
|
---|
218 |
|
---|
219 | if (isValidEscape(code, getCharCode(source, offset + 1))) {
|
---|
220 | // Consume an escaped code point.
|
---|
221 | // Note: This allows an escaped right parenthesis ("\)") to be encountered
|
---|
222 | // without ending the <bad-url-token>. This is otherwise identical to
|
---|
223 | // the "anything else" clause.
|
---|
224 | offset = consumeEscaped(source, offset);
|
---|
225 | }
|
---|
226 | }
|
---|
227 |
|
---|
228 | return offset;
|
---|
229 | }
|
---|
230 |
|
---|
231 | module.exports = {
|
---|
232 | consumeEscaped: consumeEscaped,
|
---|
233 | consumeName: consumeName,
|
---|
234 | consumeNumber: consumeNumber,
|
---|
235 | consumeBadUrlRemnants: consumeBadUrlRemnants,
|
---|
236 |
|
---|
237 | cmpChar: cmpChar,
|
---|
238 | cmpStr: cmpStr,
|
---|
239 |
|
---|
240 | getNewlineLength: getNewlineLength,
|
---|
241 | findWhiteSpaceStart: findWhiteSpaceStart,
|
---|
242 | findWhiteSpaceEnd: findWhiteSpaceEnd
|
---|
243 | };
|
---|