Mercurial
comparison .cms/lib/codemirror/mode/xquery/xquery.js @ 0:78edf6b517a0 draft
24.10
author | Coffee CMS <info@coffee-cms.ru> |
---|---|
date | Fri, 11 Oct 2024 22:40:23 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
-1:000000000000 | 0:78edf6b517a0 |
---|---|
1 // CodeMirror, copyright (c) by Marijn Haverbeke and others | |
2 // Distributed under an MIT license: https://codemirror.net/5/LICENSE | |
3 | |
4 (function(mod) { | |
5 if (typeof exports == "object" && typeof module == "object") // CommonJS | |
6 mod(require("../../lib/codemirror")); | |
7 else if (typeof define == "function" && define.amd) // AMD | |
8 define(["../../lib/codemirror"], mod); | |
9 else // Plain browser env | |
10 mod(CodeMirror); | |
11 })(function(CodeMirror) { | |
12 "use strict"; | |
13 | |
14 CodeMirror.defineMode("xquery", function() { | |
15 | |
16 // The keywords object is set to the result of this self executing | |
17 // function. Each keyword is a property of the keywords object whose | |
18 // value is {type: atype, style: astyle} | |
19 var keywords = function(){ | |
20 // convenience functions used to build keywords object | |
21 function kw(type) {return {type: type, style: "keyword"};} | |
22 var operator = kw("operator") | |
23 , atom = {type: "atom", style: "atom"} | |
24 , punctuation = {type: "punctuation", style: null} | |
25 , qualifier = {type: "axis_specifier", style: "qualifier"}; | |
26 | |
27 // kwObj is what is return from this function at the end | |
28 var kwObj = { | |
29 ',': punctuation | |
30 }; | |
31 | |
32 // a list of 'basic' keywords. For each add a property to kwObj with the value of | |
33 // {type: basic[i], style: "keyword"} e.g. 'after' --> {type: "after", style: "keyword"} | |
34 var basic = ['after', 'all', 'allowing', 'ancestor', 'ancestor-or-self', 'any', 'array', 'as', | |
35 'ascending', 'at', 'attribute', 'base-uri', 'before', 'boundary-space', 'by', 'case', 'cast', | |
36 'castable', 'catch', 'child', 'collation', 'comment', 'construction', 'contains', 'content', | |
37 'context', 'copy', 'copy-namespaces', 'count', 'decimal-format', 'declare', 'default', 'delete', | |
38 'descendant', 'descendant-or-self', 'descending', 'diacritics', 'different', 'distance', | |
39 'document', 'document-node', 'element', 'else', 'empty', 'empty-sequence', 'encoding', 'end', | |
40 'entire', 'every', 'exactly', 'except', 'external', 'first', 'following', 'following-sibling', | |
41 'for', 'from', 'ftand', 'ftnot', 'ft-option', 'ftor', 'function', 'fuzzy', 'greatest', 'group', | |
42 'if', 'import', 'in', 'inherit', 'insensitive', 'insert', 'instance', 'intersect', 'into', | |
43 'invoke', 'is', 'item', 'language', 'last', 'lax', 'least', 'let', 'levels', 'lowercase', 'map', | |
44 'modify', 'module', 'most', 'namespace', 'next', 'no', 'node', 'nodes', 'no-inherit', | |
45 'no-preserve', 'not', 'occurs', 'of', 'only', 'option', 'order', 'ordered', 'ordering', | |
46 'paragraph', 'paragraphs', 'parent', 'phrase', 'preceding', 'preceding-sibling', 'preserve', | |
47 'previous', 'processing-instruction', 'relationship', 'rename', 'replace', 'return', | |
48 'revalidation', 'same', 'satisfies', 'schema', 'schema-attribute', 'schema-element', 'score', | |
49 'self', 'sensitive', 'sentence', 'sentences', 'sequence', 'skip', 'sliding', 'some', 'stable', | |
50 'start', 'stemming', 'stop', 'strict', 'strip', 'switch', 'text', 'then', 'thesaurus', 'times', | |
51 'to', 'transform', 'treat', 'try', 'tumbling', 'type', 'typeswitch', 'union', 'unordered', | |
52 'update', 'updating', 'uppercase', 'using', 'validate', 'value', 'variable', 'version', | |
53 'weight', 'when', 'where', 'wildcards', 'window', 'with', 'without', 'word', 'words', 'xquery']; | |
54 for(var i=0, l=basic.length; i < l; i++) { kwObj[basic[i]] = kw(basic[i]);}; | |
55 | |
56 // a list of types. For each add a property to kwObj with the value of | |
57 // {type: "atom", style: "atom"} | |
58 var types = ['xs:anyAtomicType', 'xs:anySimpleType', 'xs:anyType', 'xs:anyURI', | |
59 'xs:base64Binary', 'xs:boolean', 'xs:byte', 'xs:date', 'xs:dateTime', 'xs:dateTimeStamp', | |
60 'xs:dayTimeDuration', 'xs:decimal', 'xs:double', 'xs:duration', 'xs:ENTITIES', 'xs:ENTITY', | |
61 'xs:float', 'xs:gDay', 'xs:gMonth', 'xs:gMonthDay', 'xs:gYear', 'xs:gYearMonth', 'xs:hexBinary', | |
62 'xs:ID', 'xs:IDREF', 'xs:IDREFS', 'xs:int', 'xs:integer', 'xs:item', 'xs:java', 'xs:language', | |
63 'xs:long', 'xs:Name', 'xs:NCName', 'xs:negativeInteger', 'xs:NMTOKEN', 'xs:NMTOKENS', | |
64 'xs:nonNegativeInteger', 'xs:nonPositiveInteger', 'xs:normalizedString', 'xs:NOTATION', | |
65 'xs:numeric', 'xs:positiveInteger', 'xs:precisionDecimal', 'xs:QName', 'xs:short', 'xs:string', | |
66 'xs:time', 'xs:token', 'xs:unsignedByte', 'xs:unsignedInt', 'xs:unsignedLong', | |
67 'xs:unsignedShort', 'xs:untyped', 'xs:untypedAtomic', 'xs:yearMonthDuration']; | |
68 for(var i=0, l=types.length; i < l; i++) { kwObj[types[i]] = atom;}; | |
69 | |
70 // each operator will add a property to kwObj with value of {type: "operator", style: "keyword"} | |
71 var operators = ['eq', 'ne', 'lt', 'le', 'gt', 'ge', ':=', '=', '>', '>=', '<', '<=', '.', '|', '?', 'and', 'or', 'div', 'idiv', 'mod', '*', '/', '+', '-']; | |
72 for(var i=0, l=operators.length; i < l; i++) { kwObj[operators[i]] = operator;}; | |
73 | |
74 // each axis_specifiers will add a property to kwObj with value of {type: "axis_specifier", style: "qualifier"} | |
75 var axis_specifiers = ["self::", "attribute::", "child::", "descendant::", "descendant-or-self::", "parent::", | |
76 "ancestor::", "ancestor-or-self::", "following::", "preceding::", "following-sibling::", "preceding-sibling::"]; | |
77 for(var i=0, l=axis_specifiers.length; i < l; i++) { kwObj[axis_specifiers[i]] = qualifier; }; | |
78 | |
79 return kwObj; | |
80 }(); | |
81 | |
82 function chain(stream, state, f) { | |
83 state.tokenize = f; | |
84 return f(stream, state); | |
85 } | |
86 | |
87 // the primary mode tokenizer | |
88 function tokenBase(stream, state) { | |
89 var ch = stream.next(), | |
90 mightBeFunction = false, | |
91 isEQName = isEQNameAhead(stream); | |
92 | |
93 // an XML tag (if not in some sub, chained tokenizer) | |
94 if (ch == "<") { | |
95 if(stream.match("!--", true)) | |
96 return chain(stream, state, tokenXMLComment); | |
97 | |
98 if(stream.match("![CDATA", false)) { | |
99 state.tokenize = tokenCDATA; | |
100 return "tag"; | |
101 } | |
102 | |
103 if(stream.match("?", false)) { | |
104 return chain(stream, state, tokenPreProcessing); | |
105 } | |
106 | |
107 var isclose = stream.eat("/"); | |
108 stream.eatSpace(); | |
109 var tagName = "", c; | |
110 while ((c = stream.eat(/[^\s\u00a0=<>\"\'\/?]/))) tagName += c; | |
111 | |
112 return chain(stream, state, tokenTag(tagName, isclose)); | |
113 } | |
114 // start code block | |
115 else if(ch == "{") { | |
116 pushStateStack(state, { type: "codeblock"}); | |
117 return null; | |
118 } | |
119 // end code block | |
120 else if(ch == "}") { | |
121 popStateStack(state); | |
122 return null; | |
123 } | |
124 // if we're in an XML block | |
125 else if(isInXmlBlock(state)) { | |
126 if(ch == ">") | |
127 return "tag"; | |
128 else if(ch == "/" && stream.eat(">")) { | |
129 popStateStack(state); | |
130 return "tag"; | |
131 } | |
132 else | |
133 return "variable"; | |
134 } | |
135 // if a number | |
136 else if (/\d/.test(ch)) { | |
137 stream.match(/^\d*(?:\.\d*)?(?:E[+\-]?\d+)?/); | |
138 return "atom"; | |
139 } | |
140 // comment start | |
141 else if (ch === "(" && stream.eat(":")) { | |
142 pushStateStack(state, { type: "comment"}); | |
143 return chain(stream, state, tokenComment); | |
144 } | |
145 // quoted string | |
146 else if (!isEQName && (ch === '"' || ch === "'")) | |
147 return chain(stream, state, tokenString(ch)); | |
148 // variable | |
149 else if(ch === "$") { | |
150 return chain(stream, state, tokenVariable); | |
151 } | |
152 // assignment | |
153 else if(ch ===":" && stream.eat("=")) { | |
154 return "keyword"; | |
155 } | |
156 // open paren | |
157 else if(ch === "(") { | |
158 pushStateStack(state, { type: "paren"}); | |
159 return null; | |
160 } | |
161 // close paren | |
162 else if(ch === ")") { | |
163 popStateStack(state); | |
164 return null; | |
165 } | |
166 // open paren | |
167 else if(ch === "[") { | |
168 pushStateStack(state, { type: "bracket"}); | |
169 return null; | |
170 } | |
171 // close paren | |
172 else if(ch === "]") { | |
173 popStateStack(state); | |
174 return null; | |
175 } | |
176 else { | |
177 var known = keywords.propertyIsEnumerable(ch) && keywords[ch]; | |
178 | |
179 // if there's a EQName ahead, consume the rest of the string portion, it's likely a function | |
180 if(isEQName && ch === '\"') while(stream.next() !== '"'){} | |
181 if(isEQName && ch === '\'') while(stream.next() !== '\''){} | |
182 | |
183 // gobble up a word if the character is not known | |
184 if(!known) stream.eatWhile(/[\w\$_-]/); | |
185 | |
186 // gobble a colon in the case that is a lib func type call fn:doc | |
187 var foundColon = stream.eat(":"); | |
188 | |
189 // if there's not a second colon, gobble another word. Otherwise, it's probably an axis specifier | |
190 // which should get matched as a keyword | |
191 if(!stream.eat(":") && foundColon) { | |
192 stream.eatWhile(/[\w\$_-]/); | |
193 } | |
194 // if the next non whitespace character is an open paren, this is probably a function (if not a keyword of other sort) | |
195 if(stream.match(/^[ \t]*\(/, false)) { | |
196 mightBeFunction = true; | |
197 } | |
198 // is the word a keyword? | |
199 var word = stream.current(); | |
200 known = keywords.propertyIsEnumerable(word) && keywords[word]; | |
201 | |
202 // if we think it's a function call but not yet known, | |
203 // set style to variable for now for lack of something better | |
204 if(mightBeFunction && !known) known = {type: "function_call", style: "variable def"}; | |
205 | |
206 // if the previous word was element, attribute, axis specifier, this word should be the name of that | |
207 if(isInXmlConstructor(state)) { | |
208 popStateStack(state); | |
209 return "variable"; | |
210 } | |
211 // as previously checked, if the word is element,attribute, axis specifier, call it an "xmlconstructor" and | |
212 // push the stack so we know to look for it on the next word | |
213 if(word == "element" || word == "attribute" || known.type == "axis_specifier") pushStateStack(state, {type: "xmlconstructor"}); | |
214 | |
215 // if the word is known, return the details of that else just call this a generic 'word' | |
216 return known ? known.style : "variable"; | |
217 } | |
218 } | |
219 | |
220 // handle comments, including nested | |
221 function tokenComment(stream, state) { | |
222 var maybeEnd = false, maybeNested = false, nestedCount = 0, ch; | |
223 while (ch = stream.next()) { | |
224 if (ch == ")" && maybeEnd) { | |
225 if(nestedCount > 0) | |
226 nestedCount--; | |
227 else { | |
228 popStateStack(state); | |
229 break; | |
230 } | |
231 } | |
232 else if(ch == ":" && maybeNested) { | |
233 nestedCount++; | |
234 } | |
235 maybeEnd = (ch == ":"); | |
236 maybeNested = (ch == "("); | |
237 } | |
238 | |
239 return "comment"; | |
240 } | |
241 | |
242 // tokenizer for string literals | |
243 // optionally pass a tokenizer function to set state.tokenize back to when finished | |
244 function tokenString(quote, f) { | |
245 return function(stream, state) { | |
246 var ch; | |
247 | |
248 if(isInString(state) && stream.current() == quote) { | |
249 popStateStack(state); | |
250 if(f) state.tokenize = f; | |
251 return "string"; | |
252 } | |
253 | |
254 pushStateStack(state, { type: "string", name: quote, tokenize: tokenString(quote, f) }); | |
255 | |
256 // if we're in a string and in an XML block, allow an embedded code block | |
257 if(stream.match("{", false) && isInXmlAttributeBlock(state)) { | |
258 state.tokenize = tokenBase; | |
259 return "string"; | |
260 } | |
261 | |
262 | |
263 while (ch = stream.next()) { | |
264 if (ch == quote) { | |
265 popStateStack(state); | |
266 if(f) state.tokenize = f; | |
267 break; | |
268 } | |
269 else { | |
270 // if we're in a string and in an XML block, allow an embedded code block in an attribute | |
271 if(stream.match("{", false) && isInXmlAttributeBlock(state)) { | |
272 state.tokenize = tokenBase; | |
273 return "string"; | |
274 } | |
275 | |
276 } | |
277 } | |
278 | |
279 return "string"; | |
280 }; | |
281 } | |
282 | |
283 // tokenizer for variables | |
284 function tokenVariable(stream, state) { | |
285 var isVariableChar = /[\w\$_-]/; | |
286 | |
287 // a variable may start with a quoted EQName so if the next character is quote, consume to the next quote | |
288 if(stream.eat("\"")) { | |
289 while(stream.next() !== '\"'){}; | |
290 stream.eat(":"); | |
291 } else { | |
292 stream.eatWhile(isVariableChar); | |
293 if(!stream.match(":=", false)) stream.eat(":"); | |
294 } | |
295 stream.eatWhile(isVariableChar); | |
296 state.tokenize = tokenBase; | |
297 return "variable"; | |
298 } | |
299 | |
300 // tokenizer for XML tags | |
301 function tokenTag(name, isclose) { | |
302 return function(stream, state) { | |
303 stream.eatSpace(); | |
304 if(isclose && stream.eat(">")) { | |
305 popStateStack(state); | |
306 state.tokenize = tokenBase; | |
307 return "tag"; | |
308 } | |
309 // self closing tag without attributes? | |
310 if(!stream.eat("/")) | |
311 pushStateStack(state, { type: "tag", name: name, tokenize: tokenBase}); | |
312 if(!stream.eat(">")) { | |
313 state.tokenize = tokenAttribute; | |
314 return "tag"; | |
315 } | |
316 else { | |
317 state.tokenize = tokenBase; | |
318 } | |
319 return "tag"; | |
320 }; | |
321 } | |
322 | |
323 // tokenizer for XML attributes | |
324 function tokenAttribute(stream, state) { | |
325 var ch = stream.next(); | |
326 | |
327 if(ch == "/" && stream.eat(">")) { | |
328 if(isInXmlAttributeBlock(state)) popStateStack(state); | |
329 if(isInXmlBlock(state)) popStateStack(state); | |
330 return "tag"; | |
331 } | |
332 if(ch == ">") { | |
333 if(isInXmlAttributeBlock(state)) popStateStack(state); | |
334 return "tag"; | |
335 } | |
336 if(ch == "=") | |
337 return null; | |
338 // quoted string | |
339 if (ch == '"' || ch == "'") | |
340 return chain(stream, state, tokenString(ch, tokenAttribute)); | |
341 | |
342 if(!isInXmlAttributeBlock(state)) | |
343 pushStateStack(state, { type: "attribute", tokenize: tokenAttribute}); | |
344 | |
345 stream.eat(/[a-zA-Z_:]/); | |
346 stream.eatWhile(/[-a-zA-Z0-9_:.]/); | |
347 stream.eatSpace(); | |
348 | |
349 // the case where the attribute has not value and the tag was closed | |
350 if(stream.match(">", false) || stream.match("/", false)) { | |
351 popStateStack(state); | |
352 state.tokenize = tokenBase; | |
353 } | |
354 | |
355 return "attribute"; | |
356 } | |
357 | |
358 // handle comments, including nested | |
359 function tokenXMLComment(stream, state) { | |
360 var ch; | |
361 while (ch = stream.next()) { | |
362 if (ch == "-" && stream.match("->", true)) { | |
363 state.tokenize = tokenBase; | |
364 return "comment"; | |
365 } | |
366 } | |
367 } | |
368 | |
369 | |
370 // handle CDATA | |
371 function tokenCDATA(stream, state) { | |
372 var ch; | |
373 while (ch = stream.next()) { | |
374 if (ch == "]" && stream.match("]", true)) { | |
375 state.tokenize = tokenBase; | |
376 return "comment"; | |
377 } | |
378 } | |
379 } | |
380 | |
381 // handle preprocessing instructions | |
382 function tokenPreProcessing(stream, state) { | |
383 var ch; | |
384 while (ch = stream.next()) { | |
385 if (ch == "?" && stream.match(">", true)) { | |
386 state.tokenize = tokenBase; | |
387 return "comment meta"; | |
388 } | |
389 } | |
390 } | |
391 | |
392 | |
393 // functions to test the current context of the state | |
394 function isInXmlBlock(state) { return isIn(state, "tag"); } | |
395 function isInXmlAttributeBlock(state) { return isIn(state, "attribute"); } | |
396 function isInXmlConstructor(state) { return isIn(state, "xmlconstructor"); } | |
397 function isInString(state) { return isIn(state, "string"); } | |
398 | |
399 function isEQNameAhead(stream) { | |
400 // assume we've already eaten a quote (") | |
401 if(stream.current() === '"') | |
402 return stream.match(/^[^\"]+\"\:/, false); | |
403 else if(stream.current() === '\'') | |
404 return stream.match(/^[^\"]+\'\:/, false); | |
405 else | |
406 return false; | |
407 } | |
408 | |
409 function isIn(state, type) { | |
410 return (state.stack.length && state.stack[state.stack.length - 1].type == type); | |
411 } | |
412 | |
413 function pushStateStack(state, newState) { | |
414 state.stack.push(newState); | |
415 } | |
416 | |
417 function popStateStack(state) { | |
418 state.stack.pop(); | |
419 var reinstateTokenize = state.stack.length && state.stack[state.stack.length-1].tokenize; | |
420 state.tokenize = reinstateTokenize || tokenBase; | |
421 } | |
422 | |
423 // the interface for the mode API | |
424 return { | |
425 startState: function() { | |
426 return { | |
427 tokenize: tokenBase, | |
428 cc: [], | |
429 stack: [] | |
430 }; | |
431 }, | |
432 | |
433 token: function(stream, state) { | |
434 if (stream.eatSpace()) return null; | |
435 var style = state.tokenize(stream, state); | |
436 return style; | |
437 }, | |
438 | |
439 blockCommentStart: "(:", | |
440 blockCommentEnd: ":)" | |
441 | |
442 }; | |
443 | |
444 }); | |
445 | |
446 CodeMirror.defineMIME("application/xquery", "xquery"); | |
447 | |
448 }); |