1 | /*
|
---|
2 | * 09/03/2005
|
---|
3 | *
|
---|
4 | * CSSTokenMaker.java - Token maker for CSS 3 files.
|
---|
5 | *
|
---|
6 | * This library is distributed under a modified BSD license. See the included
|
---|
7 | * RSyntaxTextArea.License.txt file for details.
|
---|
8 | */
|
---|
9 | package org.fife.ui.rsyntaxtextarea.modes;
|
---|
10 |
|
---|
11 | import java.io.*;
|
---|
12 | import javax.swing.text.Segment;
|
---|
13 |
|
---|
14 | import org.fife.ui.rsyntaxtextarea.*;
|
---|
15 |
|
---|
16 |
|
---|
17 | /**
|
---|
18 | * This class splits up text into tokens representing a CSS 3 file. It's
|
---|
19 | * written with a few extra internal states so that it can easily be copy
|
---|
20 | * and pasted into HTML, PHP, and JSP TokenMakres when it is updated.<p>
|
---|
21 | *
|
---|
22 | * This implementation was created using
|
---|
23 | * <a href="http://www.jflex.de/">JFlex</a> 1.4.1; however, the generated file
|
---|
24 | * was modified for performance. Memory allocation needs to be almost
|
---|
25 | * completely removed to be competitive with the handwritten lexers (subclasses
|
---|
26 | * of <code>AbstractTokenMaker</code>, so this class has been modified so that
|
---|
27 | * Strings are never allocated (via yytext()), and the scanner never has to
|
---|
28 | * worry about refilling its buffer (needlessly copying chars around).
|
---|
29 | * We can achieve this because RText always scans exactly 1 line of tokens at a
|
---|
30 | * time, and hands the scanner this line as an array of characters (a Segment
|
---|
31 | * really). Since tokens contain pointers to char arrays instead of Strings
|
---|
32 | * holding their contents, there is no need for allocating new memory for
|
---|
33 | * Strings.<p>
|
---|
34 | *
|
---|
35 | * The actual algorithm generated for scanning has, of course, not been
|
---|
36 | * modified.<p>
|
---|
37 | *
|
---|
38 | * If you wish to regenerate this file yourself, keep in mind the following:
|
---|
39 | * <ul>
|
---|
40 | * <li>The generated CSSTokenMaker.java</code> file will contain 2
|
---|
41 | * definitions of both <code>zzRefill</code> and <code>yyreset</code>.
|
---|
42 | * You should hand-delete the second of each definition (the ones
|
---|
43 | * generated by the lexer), as these generated methods modify the input
|
---|
44 | * buffer, which we'll never have to do.</li>
|
---|
45 | * <li>You should also change the declaration/definition of zzBuffer to NOT
|
---|
46 | * be initialized. This is a needless memory allocation for us since we
|
---|
47 | * will be pointing the array somewhere else anyway.</li>
|
---|
48 | * <li>You should NOT call <code>yylex()</code> on the generated scanner
|
---|
49 | * directly; rather, you should use <code>getTokenList</code> as you would
|
---|
50 | * with any other <code>TokenMaker</code> instance.</li>
|
---|
51 | * </ul>
|
---|
52 | *
|
---|
53 | * @author Robert Futrell
|
---|
54 | * @version 0.4
|
---|
55 | *
|
---|
56 | */
|
---|
57 | %%
|
---|
58 |
|
---|
59 | %public
|
---|
60 | %class CSSTokenMaker
|
---|
61 | %extends AbstractJFlexTokenMaker
|
---|
62 | %implements TokenMaker
|
---|
63 | %unicode
|
---|
64 | %type org.fife.ui.rsyntaxtextarea.Token
|
---|
65 |
|
---|
66 |
|
---|
67 | %{
|
---|
68 |
|
---|
69 | /**
|
---|
70 | * Internal type denoting a line ending in a CSS property.
|
---|
71 | */
|
---|
72 | public static final int INTERNAL_CSS_PROPERTY = -1;
|
---|
73 |
|
---|
74 | /**
|
---|
75 | * Internal type denoting a line ending in a CSS property value.
|
---|
76 | */
|
---|
77 | public static final int INTERNAL_CSS_VALUE = -2;
|
---|
78 |
|
---|
79 | /**
|
---|
80 | * Internal type denoting line ending in a CSS double-quote string.
|
---|
81 | * The state to return to is embedded in the actual end token type.
|
---|
82 | */
|
---|
83 | public static final int INTERNAL_CSS_STRING = -(1<<11);
|
---|
84 |
|
---|
85 | /**
|
---|
86 | * Internal type denoting line ending in a CSS single-quote string.
|
---|
87 | * The state to return to is embedded in the actual end token type.
|
---|
88 | */
|
---|
89 | public static final int INTERNAL_CSS_CHAR = -(2<<11);
|
---|
90 |
|
---|
91 | /**
|
---|
92 | * Internal type denoting line ending in a CSS multi-line comment.
|
---|
93 | * The state to return to is embedded in the actual end token type.
|
---|
94 | */
|
---|
95 | public static final int INTERNAL_CSS_MLC = -(3<<11);
|
---|
96 |
|
---|
97 | /**
|
---|
98 | * The state previous CSS-related state we were in before going into a CSS
|
---|
99 | * string, multi-line comment, etc.
|
---|
100 | */
|
---|
101 | private int cssPrevState;
|
---|
102 |
|
---|
103 |
|
---|
104 | /**
|
---|
105 | * Constructor. This must be here because JFlex does not generate a
|
---|
106 | * no-parameter constructor.
|
---|
107 | */
|
---|
108 | public CSSTokenMaker() {
|
---|
109 | super();
|
---|
110 | }
|
---|
111 |
|
---|
112 |
|
---|
113 | /**
|
---|
114 | * Adds the token specified to the current linked list of tokens as an
|
---|
115 | * "end token;" that is, at <code>zzMarkedPos</code>.
|
---|
116 | *
|
---|
117 | * @param tokenType The token's type.
|
---|
118 | */
|
---|
119 | private void addEndToken(int tokenType) {
|
---|
120 | addToken(zzMarkedPos,zzMarkedPos, tokenType);
|
---|
121 | }
|
---|
122 |
|
---|
123 |
|
---|
124 | /**
|
---|
125 | * Adds the token specified to the current linked list of tokens.
|
---|
126 | *
|
---|
127 | * @param tokenType The token's type.
|
---|
128 | * @see #addToken(int, int, int)
|
---|
129 | */
|
---|
130 | private void addHyperlinkToken(int start, int end, int tokenType) {
|
---|
131 | int so = start + offsetShift;
|
---|
132 | addToken(zzBuffer, start,end, tokenType, so, true);
|
---|
133 | }
|
---|
134 |
|
---|
135 |
|
---|
136 | /**
|
---|
137 | * Adds the token specified to the current linked list of tokens.
|
---|
138 | *
|
---|
139 | * @param tokenType The token's type.
|
---|
140 | */
|
---|
141 | private void addToken(int tokenType) {
|
---|
142 | addToken(zzStartRead, zzMarkedPos-1, tokenType);
|
---|
143 | }
|
---|
144 |
|
---|
145 |
|
---|
146 | /**
|
---|
147 | * Adds the token specified to the current linked list of tokens.
|
---|
148 | *
|
---|
149 | * @param tokenType The token's type.
|
---|
150 | */
|
---|
151 | private void addToken(int start, int end, int tokenType) {
|
---|
152 | int so = start + offsetShift;
|
---|
153 | addToken(zzBuffer, start,end, tokenType, so);
|
---|
154 | }
|
---|
155 |
|
---|
156 |
|
---|
157 | /**
|
---|
158 | * Adds the token specified to the current linked list of tokens.
|
---|
159 | *
|
---|
160 | * @param array The character array.
|
---|
161 | * @param start The starting offset in the array.
|
---|
162 | * @param end The ending offset in the array.
|
---|
163 | * @param tokenType The token's type.
|
---|
164 | * @param startOffset The offset in the document at which this token
|
---|
165 | * occurs.
|
---|
166 | */
|
---|
167 | public void addToken(char[] array, int start, int end, int tokenType, int startOffset) {
|
---|
168 | super.addToken(array, start,end, tokenType, startOffset);
|
---|
169 | zzStartRead = zzMarkedPos;
|
---|
170 | }
|
---|
171 |
|
---|
172 |
|
---|
173 | /**
|
---|
174 | * Returns <code>true</code> since CSS uses curly braces.
|
---|
175 | *
|
---|
176 | * @return <code>true</code> always.
|
---|
177 | */
|
---|
178 | public boolean getCurlyBracesDenoteCodeBlocks() {
|
---|
179 | return true;
|
---|
180 | }
|
---|
181 |
|
---|
182 |
|
---|
183 | /**
|
---|
184 | * Returns the first token in the linked list of tokens generated
|
---|
185 | * from <code>text</code>. This method must be implemented by
|
---|
186 | * subclasses so they can correctly implement syntax highlighting.
|
---|
187 | *
|
---|
188 | * @param text The text from which to get tokens.
|
---|
189 | * @param initialTokenType The token type we should start with.
|
---|
190 | * @param startOffset The offset into the document at which
|
---|
191 | * <code>text</code> starts.
|
---|
192 | * @return The first <code>Token</code> in a linked list representing
|
---|
193 | * the syntax highlighted text.
|
---|
194 | */
|
---|
195 | public Token getTokenList(Segment text, int initialTokenType, int startOffset) {
|
---|
196 |
|
---|
197 | resetTokenList();
|
---|
198 | this.offsetShift = -text.offset + startOffset;
|
---|
199 | cssPrevState = YYINITIAL; // Shouldn't be necessary
|
---|
200 |
|
---|
201 | // Start off in the proper state.
|
---|
202 | int state = Token.NULL;
|
---|
203 | switch (initialTokenType) {
|
---|
204 | case Token.LITERAL_STRING_DOUBLE_QUOTE:
|
---|
205 | state = CSS_STRING;
|
---|
206 | break;
|
---|
207 | case Token.LITERAL_CHAR:
|
---|
208 | state = CSS_CHAR_LITERAL;
|
---|
209 | break;
|
---|
210 | case Token.COMMENT_MULTILINE:
|
---|
211 | state = CSS_C_STYLE_COMMENT;
|
---|
212 | break;
|
---|
213 | case INTERNAL_CSS_PROPERTY:
|
---|
214 | state = CSS_PROPERTY;
|
---|
215 | break;
|
---|
216 | case INTERNAL_CSS_VALUE:
|
---|
217 | state = CSS_VALUE;
|
---|
218 | break;
|
---|
219 | default:
|
---|
220 | if (initialTokenType<-1024) {
|
---|
221 | int main = -(-initialTokenType & 0xffffff00);
|
---|
222 | switch (main) {
|
---|
223 | default: // Should never happen
|
---|
224 | case INTERNAL_CSS_STRING:
|
---|
225 | state = CSS_STRING;
|
---|
226 | break;
|
---|
227 | case INTERNAL_CSS_CHAR:
|
---|
228 | state = CSS_CHAR_LITERAL;
|
---|
229 | break;
|
---|
230 | case INTERNAL_CSS_MLC:
|
---|
231 | state = CSS_C_STYLE_COMMENT;
|
---|
232 | break;
|
---|
233 | }
|
---|
234 | cssPrevState = -initialTokenType&0xff;
|
---|
235 | }
|
---|
236 | else {
|
---|
237 | state = Token.NULL;
|
---|
238 | }
|
---|
239 | }
|
---|
240 |
|
---|
241 | start = text.offset;
|
---|
242 | s = text;
|
---|
243 | try {
|
---|
244 | yyreset(zzReader);
|
---|
245 | yybegin(state);
|
---|
246 | return yylex();
|
---|
247 | } catch (IOException ioe) {
|
---|
248 | ioe.printStackTrace();
|
---|
249 | return new DefaultToken();
|
---|
250 | }
|
---|
251 |
|
---|
252 | }
|
---|
253 |
|
---|
254 |
|
---|
255 | /**
|
---|
256 | * Refills the input buffer.
|
---|
257 | *
|
---|
258 | * @return <code>true</code> if EOF was reached, otherwise
|
---|
259 | * <code>false</code>.
|
---|
260 | */
|
---|
261 | private boolean zzRefill() {
|
---|
262 | return zzCurrentPos>=s.offset+s.count;
|
---|
263 | }
|
---|
264 |
|
---|
265 |
|
---|
266 | /**
|
---|
267 | * Resets the scanner to read from a new input stream.
|
---|
268 | * Does not close the old reader.
|
---|
269 | *
|
---|
270 | * All internal variables are reset, the old input stream
|
---|
271 | * <b>cannot</b> be reused (internal buffer is discarded and lost).
|
---|
272 | * Lexical state is set to <tt>YY_INITIAL</tt>.
|
---|
273 | *
|
---|
274 | * @param reader the new input stream
|
---|
275 | */
|
---|
276 | public final void yyreset(java.io.Reader reader) {
|
---|
277 | // 's' has been updated.
|
---|
278 | zzBuffer = s.array;
|
---|
279 | /*
|
---|
280 | * We replaced the line below with the two below it because zzRefill
|
---|
281 | * no longer "refills" the buffer (since the way we do it, it's always
|
---|
282 | * "full" the first time through, since it points to the segment's
|
---|
283 | * array). So, we assign zzEndRead here.
|
---|
284 | */
|
---|
285 | //zzStartRead = zzEndRead = s.offset;
|
---|
286 | zzStartRead = s.offset;
|
---|
287 | zzEndRead = zzStartRead + s.count - 1;
|
---|
288 | zzCurrentPos = zzMarkedPos = zzPushbackPos = s.offset;
|
---|
289 | zzLexicalState = YYINITIAL;
|
---|
290 | zzReader = reader;
|
---|
291 | zzAtBOL = true;
|
---|
292 | zzAtEOF = false;
|
---|
293 | }
|
---|
294 |
|
---|
295 |
|
---|
296 | %}
|
---|
297 |
|
---|
298 | Digit = ([0-9])
|
---|
299 | Letter = ([A-Za-z])
|
---|
300 | LetterOrUnderscore = ({Letter}|[_])
|
---|
301 | LetterOrUnderscoreOrDash = ({LetterOrUnderscore}|[\-])
|
---|
302 |
|
---|
303 | CSS_SelectorPiece = (("*"|"."|{LetterOrUnderscoreOrDash})({LetterOrUnderscoreOrDash}|"."|{Digit})*)
|
---|
304 | CSS_PseudoClass = (":"("root"|"nth-child"|"nth-last-child"|"nth-of-type"|"nth-last-of-type"|"first-child"|"last-child"|"first-of-type"|"last-of-type"|"only-child"|"only-of-type"|"empty"|"link"|"visited"|"active"|"hover"|"focus"|"target"|"lang"|"enabled"|"disabled"|"checked"|":first-line"|":first-letter"|":before"|":after"|"not"))
|
---|
305 | CSS_AtKeyword = ("@"{CSS_SelectorPiece})
|
---|
306 | CSS_Id = ("#"{CSS_SelectorPiece})
|
---|
307 | CSS_Separator = ([;\(\)\[\]])
|
---|
308 | WhiteSpace = ([ \t]+)
|
---|
309 | MlcStart = ("/*")
|
---|
310 | MlcEnd = ("*/")
|
---|
311 |
|
---|
312 | CSS_Property = ([\*]?{LetterOrUnderscoreOrDash}({LetterOrUnderscoreOrDash}|{Digit})*)
|
---|
313 | CSS_ValueChar = ({LetterOrUnderscoreOrDash}|[\\/])
|
---|
314 | CSS_Value = ({CSS_ValueChar}*)
|
---|
315 | CSS_Function = ({CSS_Value}\()
|
---|
316 | CSS_Digits = ([\-]?{Digit}+([0-9\.]+)?(pt|pc|in|mm|cm|em|ex|px|ms|s|%)?)
|
---|
317 | CSS_Hex = ("#"[0-9a-fA-F]+)
|
---|
318 | CSS_Number = ({CSS_Digits}|{CSS_Hex})
|
---|
319 |
|
---|
320 | URLGenDelim = ([:\/\?#\[\]@])
|
---|
321 | URLSubDelim = ([\!\$&'\(\)\*\+,;=])
|
---|
322 | URLUnreserved = ({LetterOrUnderscore}|{Digit}|[\-\.\~])
|
---|
323 | URLCharacter = ({URLGenDelim}|{URLSubDelim}|{URLUnreserved}|[%])
|
---|
324 | URLCharacters = ({URLCharacter}*)
|
---|
325 | URLEndCharacter = ([\/\$]|{Letter}|{Digit})
|
---|
326 | URL = (((https?|f(tp|ile))"://"|"www.")({URLCharacters}{URLEndCharacter})?)
|
---|
327 |
|
---|
328 |
|
---|
329 | %state CSS_PROPERTY
|
---|
330 | %state CSS_VALUE
|
---|
331 | %state CSS_STRING
|
---|
332 | %state CSS_CHAR_LITERAL
|
---|
333 | %state CSS_C_STYLE_COMMENT
|
---|
334 |
|
---|
335 |
|
---|
336 | %%
|
---|
337 |
|
---|
338 | <YYINITIAL> {
|
---|
339 | {CSS_SelectorPiece} { addToken(Token.DATA_TYPE); }
|
---|
340 | {CSS_PseudoClass} { addToken(Token.RESERVED_WORD); }
|
---|
341 | ":" { /* Unknown pseudo class */ addToken(Token.DATA_TYPE); }
|
---|
342 | {CSS_AtKeyword} { addToken(Token.REGEX); }
|
---|
343 | {CSS_Id} { addToken(Token.VARIABLE); }
|
---|
344 | "{" { addToken(Token.SEPARATOR); yybegin(CSS_PROPERTY); }
|
---|
345 | [,] { addToken(Token.IDENTIFIER); }
|
---|
346 | \" { start = zzMarkedPos-1; cssPrevState = zzLexicalState; yybegin(CSS_STRING); }
|
---|
347 | \' { start = zzMarkedPos-1; cssPrevState = zzLexicalState; yybegin(CSS_CHAR_LITERAL); }
|
---|
348 | [+>~\^$\|=] { addToken(Token.OPERATOR); }
|
---|
349 | {CSS_Separator} { addToken(Token.SEPARATOR); }
|
---|
350 | {WhiteSpace} { addToken(Token.WHITESPACE); }
|
---|
351 | {MlcStart} { start = zzMarkedPos-2; cssPrevState = zzLexicalState; yybegin(CSS_C_STYLE_COMMENT); }
|
---|
352 | . { /*System.out.println("yyinitial: " + yytext());*/ addToken(Token.IDENTIFIER); }
|
---|
353 | "\n" |
|
---|
354 | <<EOF>> { addNullToken(); return firstToken; }
|
---|
355 | }
|
---|
356 |
|
---|
357 | <CSS_PROPERTY> {
|
---|
358 | {CSS_Property} { addToken(Token.RESERVED_WORD); }
|
---|
359 | "}" { addToken(Token.SEPARATOR); yybegin(YYINITIAL); }
|
---|
360 | ":" { addToken(Token.OPERATOR); yybegin(CSS_VALUE); }
|
---|
361 | {WhiteSpace} { addToken(Token.WHITESPACE); }
|
---|
362 | {MlcStart} { start = zzMarkedPos-2; cssPrevState = zzLexicalState; yybegin(CSS_C_STYLE_COMMENT); }
|
---|
363 | . { /*System.out.println("css_property: " + yytext());*/ addToken(Token.IDENTIFIER); }
|
---|
364 | "\n" |
|
---|
365 | <<EOF>> { addEndToken(INTERNAL_CSS_PROPERTY); return firstToken; }
|
---|
366 | }
|
---|
367 |
|
---|
368 | <CSS_VALUE> {
|
---|
369 | {CSS_Value} { addToken(Token.IDENTIFIER); }
|
---|
370 | "!important" { addToken(Token.ANNOTATION); }
|
---|
371 | {CSS_Function} { int temp = zzMarkedPos - 2;
|
---|
372 | addToken(zzStartRead, temp, Token.FUNCTION);
|
---|
373 | addToken(zzMarkedPos-1, zzMarkedPos-1, Token.SEPARATOR);
|
---|
374 | zzStartRead = zzCurrentPos = zzMarkedPos;
|
---|
375 | }
|
---|
376 | {CSS_Number} { addToken(Token.LITERAL_NUMBER_DECIMAL_INT); }
|
---|
377 | \" { start = zzMarkedPos-1; cssPrevState = zzLexicalState; yybegin(CSS_STRING); }
|
---|
378 | \' { start = zzMarkedPos-1; cssPrevState = zzLexicalState; yybegin(CSS_CHAR_LITERAL); }
|
---|
379 | ")" { /* End of a function */ addToken(Token.SEPARATOR); }
|
---|
380 | [;] { addToken(Token.OPERATOR); yybegin(CSS_PROPERTY); }
|
---|
381 | [,\.] { addToken(Token.IDENTIFIER); }
|
---|
382 | "}" { addToken(Token.SEPARATOR); yybegin(YYINITIAL); }
|
---|
383 | {WhiteSpace} { addToken(Token.WHITESPACE); }
|
---|
384 | {MlcStart} { start = zzMarkedPos-2; cssPrevState = zzLexicalState; yybegin(CSS_C_STYLE_COMMENT); }
|
---|
385 | . { /*System.out.println("css_value: " + yytext());*/ addToken(Token.IDENTIFIER); }
|
---|
386 | "\n" |
|
---|
387 | <<EOF>> { addEndToken(INTERNAL_CSS_VALUE); return firstToken; }
|
---|
388 | }
|
---|
389 |
|
---|
390 | <CSS_STRING> {
|
---|
391 | [^\n\\\"]+ {}
|
---|
392 | \\.? { /* Skip escaped chars. */ }
|
---|
393 | \" { addToken(start,zzStartRead, Token.LITERAL_STRING_DOUBLE_QUOTE); yybegin(cssPrevState); }
|
---|
394 | \n |
|
---|
395 | <<EOF>> { addToken(start,zzStartRead-1, Token.LITERAL_STRING_DOUBLE_QUOTE); addEndToken(INTERNAL_CSS_STRING - cssPrevState); return firstToken; }
|
---|
396 | }
|
---|
397 |
|
---|
398 | <CSS_CHAR_LITERAL> {
|
---|
399 | [^\n\\\']+ {}
|
---|
400 | \\.? { /* Skip escaped chars. */ }
|
---|
401 | \' { addToken(start,zzStartRead, Token.LITERAL_CHAR); yybegin(cssPrevState); }
|
---|
402 | \n |
|
---|
403 | <<EOF>> { addToken(start,zzStartRead-1, Token.LITERAL_CHAR); addEndToken(INTERNAL_CSS_CHAR - cssPrevState); return firstToken; }
|
---|
404 | }
|
---|
405 |
|
---|
406 | <CSS_C_STYLE_COMMENT> {
|
---|
407 | [^hwf\n\*]+ {}
|
---|
408 | {URL} { int temp=zzStartRead; addToken(start,zzStartRead-1, Token.COMMENT_MULTILINE); addHyperlinkToken(temp,zzMarkedPos-1, Token.COMMENT_MULTILINE); start = zzMarkedPos; }
|
---|
409 | [hwf] {}
|
---|
410 | {MlcEnd} { addToken(start,zzStartRead+1, Token.COMMENT_MULTILINE); yybegin(cssPrevState); }
|
---|
411 | \* {}
|
---|
412 | \n |
|
---|
413 | <<EOF>> { addToken(start,zzStartRead-1, Token.COMMENT_MULTILINE); addEndToken(INTERNAL_CSS_MLC - cssPrevState); return firstToken; }
|
---|
414 | }
|
---|