1 | package eu.clarin.sru.server.fcs; |
---|
2 | |
---|
3 | import javax.xml.stream.XMLStreamException; |
---|
4 | import javax.xml.stream.XMLStreamWriter; |
---|
5 | |
---|
6 | |
---|
7 | /** |
---|
8 | * This class provides several helper methods for writing records in the |
---|
9 | * CLARIN-FCS record schema. These methods <em>do not</em> cover the full |
---|
10 | * spectrum of all variations of records that are permitted by the CLARIN-FCS |
---|
11 | * specification. |
---|
12 | * |
---|
13 | * @see <a |
---|
14 | * href="https://trac.clarin.eu/wiki/FCS/Specification"> |
---|
15 | * CLARIN FCS specification, section "Operation searchRetrieve"</a> |
---|
16 | */ |
---|
17 | public final class XMLStreamWriterHelper { |
---|
18 | private static final String FCS_NS = "http://clarin.eu/fcs/1.0"; |
---|
19 | private static final String FCS_PREFIX = "fcs"; |
---|
20 | private static final String FCS_KWIC_NS = |
---|
21 | "http://clarin.eu/fcs/1.0/kwic"; |
---|
22 | private static final String FCS_KWIC_PREFIX = "kwic"; |
---|
23 | private static final String FCS_KWIC_MIMETYPE = |
---|
24 | "application/x-clarin-fcs-kwic+xml"; |
---|
25 | private static final String FCS_HITS_NS = |
---|
26 | "http://clarin.eu/fcs/dataview/hits"; |
---|
27 | private static final String FCS_HITS_PREFIX = "hits"; |
---|
28 | private static final String FCS_HITS_MIMETYPE = |
---|
29 | "application/x-clarin-fcs-hits+xml"; |
---|
30 | |
---|
31 | |
---|
32 | /** |
---|
33 | * Write the start of a resource (i.e. the <code><Resource></code> |
---|
34 | * element). Calls to this method need to be balanced with calls to the |
---|
35 | * {@link #writeEndResource(XMLStreamWriter)} method. |
---|
36 | * |
---|
37 | * |
---|
38 | * @param writer |
---|
39 | * the {@link XMLStreamWriter} to be used |
---|
40 | * @param pid |
---|
41 | * the persistent identifier of this resource or |
---|
42 | * <code>null</code>, if not applicable |
---|
43 | * @param ref |
---|
44 | * the reference of this resource or <code>null</code>, if not |
---|
45 | * applicable |
---|
46 | * @throws XMLStreamException |
---|
47 | * if an error occurred |
---|
48 | */ |
---|
49 | public static void writeStartResource(XMLStreamWriter writer, String pid, |
---|
50 | String ref) throws XMLStreamException { |
---|
51 | if (writer == null) { |
---|
52 | throw new NullPointerException("writer == null"); |
---|
53 | } |
---|
54 | |
---|
55 | writer.setPrefix(FCS_PREFIX, FCS_NS); |
---|
56 | writer.writeStartElement(FCS_NS, "Resource"); |
---|
57 | writer.writeNamespace(FCS_PREFIX, FCS_NS); |
---|
58 | if ((pid != null) && !pid.isEmpty()) { |
---|
59 | writer.writeAttribute("pid", pid); |
---|
60 | } |
---|
61 | if ((ref != null) && !ref.isEmpty()) { |
---|
62 | writer.writeAttribute("ref", ref); |
---|
63 | } |
---|
64 | } |
---|
65 | |
---|
66 | |
---|
67 | /** |
---|
68 | * Write the end of a resource (i.e. the <code></Resource></code> |
---|
69 | * element). Calls to this method need to be balanced with calls to the |
---|
70 | * {@link #writeStartResource(XMLStreamWriter, String, String)} method. |
---|
71 | * |
---|
72 | * @param writer |
---|
73 | * the {@link XMLStreamWriter} to be used |
---|
74 | * @throws XMLStreamException |
---|
75 | * if an error occurred |
---|
76 | */ |
---|
77 | public static void writeEndResource(XMLStreamWriter writer) |
---|
78 | throws XMLStreamException { |
---|
79 | if (writer == null) { |
---|
80 | throw new NullPointerException("writer == null"); |
---|
81 | } |
---|
82 | |
---|
83 | writer.writeEndElement(); // "Resource" element |
---|
84 | } |
---|
85 | |
---|
86 | |
---|
87 | /** |
---|
88 | * Write the start of a resource fragment (i.e. the |
---|
89 | * <code><ResourceFragment></code> element). Calls to this method need |
---|
90 | * to be balanced with calls to the |
---|
91 | * {@link #writeEndResourceFragment(XMLStreamWriter)} method. |
---|
92 | * |
---|
93 | * |
---|
94 | * @param writer |
---|
95 | * the {@link XMLStreamWriter} to be used |
---|
96 | * @param pid |
---|
97 | * the persistent identifier of this resource or |
---|
98 | * <code>null</code>, if not applicable |
---|
99 | * @param ref |
---|
100 | * the reference of this resource or <code>null</code>, if not |
---|
101 | * applicable |
---|
102 | * @throws XMLStreamException |
---|
103 | * if an error occurred |
---|
104 | */ |
---|
105 | public static void writeStartResourceFragment(XMLStreamWriter writer, |
---|
106 | String pid, String ref) throws XMLStreamException { |
---|
107 | if (writer == null) { |
---|
108 | throw new NullPointerException("writer == null"); |
---|
109 | } |
---|
110 | |
---|
111 | writer.writeStartElement(FCS_NS, "ResourceFragment"); |
---|
112 | if ((pid != null) && !pid.isEmpty()) { |
---|
113 | writer.writeAttribute("pid", pid); |
---|
114 | } |
---|
115 | if ((ref != null) && !ref.isEmpty()) { |
---|
116 | writer.writeAttribute("ref", ref); |
---|
117 | } |
---|
118 | } |
---|
119 | |
---|
120 | |
---|
121 | /** |
---|
122 | * Write the end of a resource fragment (i.e. the |
---|
123 | * <code></ResourceFragment></code> element). Calls to this method |
---|
124 | * need to be balanced with calls to the |
---|
125 | * {@link #writeStartResourceFragment(XMLStreamWriter, String, String)} |
---|
126 | * method. |
---|
127 | * |
---|
128 | * @param writer |
---|
129 | * the {@link XMLStreamWriter} to be used |
---|
130 | * @throws XMLStreamException |
---|
131 | * if an error occurred |
---|
132 | */ |
---|
133 | public static void writeEndResourceFragment(XMLStreamWriter writer) |
---|
134 | throws XMLStreamException { |
---|
135 | if (writer == null) { |
---|
136 | throw new NullPointerException("writer == null"); |
---|
137 | } |
---|
138 | |
---|
139 | writer.writeEndElement(); // "ResourceFragment" element |
---|
140 | } |
---|
141 | |
---|
142 | |
---|
143 | /** |
---|
144 | * Write the start of a data view (i.e. the |
---|
145 | * <code><DataView></code> element). Calls to this method need |
---|
146 | * to be balanced with calls to the |
---|
147 | * {@link #writeEndResource(XMLStreamWriter)} method. |
---|
148 | * |
---|
149 | * |
---|
150 | * @param writer |
---|
151 | * the {@link XMLStreamWriter} to be used |
---|
152 | * @param mimetype |
---|
153 | * the MIME type of this data view |
---|
154 | * applicable |
---|
155 | * @throws XMLStreamException |
---|
156 | * if an error occurred |
---|
157 | */ |
---|
158 | public static void writeStartDataView(XMLStreamWriter writer, |
---|
159 | String mimetype) throws XMLStreamException { |
---|
160 | if (writer == null) { |
---|
161 | throw new NullPointerException("writer == null"); |
---|
162 | } |
---|
163 | if (mimetype == null) { |
---|
164 | throw new NullPointerException("mimetype == null"); |
---|
165 | } |
---|
166 | if (mimetype.isEmpty()) { |
---|
167 | throw new IllegalArgumentException("mimetype is empty"); |
---|
168 | } |
---|
169 | |
---|
170 | writer.writeStartElement(FCS_NS, "DataView"); |
---|
171 | writer.writeAttribute("type", mimetype); |
---|
172 | } |
---|
173 | |
---|
174 | |
---|
175 | /** |
---|
176 | * Write the end of a data view (i.e. the <code></DataView></code> |
---|
177 | * element). Calls to this method need to be balanced with calls to the |
---|
178 | * {@link #writeStartDataView(XMLStreamWriter, String)} method. |
---|
179 | * |
---|
180 | * @param writer |
---|
181 | * the {@link XMLStreamWriter} to be used |
---|
182 | * @throws XMLStreamException |
---|
183 | * if an error occurred |
---|
184 | */ |
---|
185 | public static void writeEndDataView(XMLStreamWriter writer) |
---|
186 | throws XMLStreamException { |
---|
187 | if (writer == null) { |
---|
188 | throw new NullPointerException("writer == null"); |
---|
189 | } |
---|
190 | |
---|
191 | writer.writeEndElement(); // "DataView" element |
---|
192 | } |
---|
193 | |
---|
194 | |
---|
195 | /** |
---|
196 | * Convince method to write a KWIC data view. It automatically performs the |
---|
197 | * calls to {@link #writeStartDataView(XMLStreamWriter, String)} and |
---|
198 | * {@link #writeEndDataView(XMLStreamWriter)}. |
---|
199 | * |
---|
200 | * @param writer |
---|
201 | * the {@link XMLStreamWriter} to be used |
---|
202 | * @param left |
---|
203 | * the left context of the KWIC or <code>null</code> if not |
---|
204 | * applicable |
---|
205 | * @param keyword |
---|
206 | * the keyword of the KWIC |
---|
207 | * @param right |
---|
208 | * the right context of the KWIC or <code>null</code> if not |
---|
209 | * applicable |
---|
210 | * @throws XMLStreamException |
---|
211 | * if an error occurred |
---|
212 | * @deprecated The the HITS data view instead. |
---|
213 | */ |
---|
214 | @Deprecated |
---|
215 | public static void writeKWICDataView(XMLStreamWriter writer, String left, |
---|
216 | String keyword, String right) throws XMLStreamException { |
---|
217 | if (writer == null) { |
---|
218 | throw new NullPointerException("writer == null"); |
---|
219 | } |
---|
220 | if (keyword == null) { |
---|
221 | throw new NullPointerException("keyword == null"); |
---|
222 | } |
---|
223 | |
---|
224 | writeStartDataView(writer, FCS_KWIC_MIMETYPE); |
---|
225 | |
---|
226 | // actual "kwic" data view |
---|
227 | writer.setPrefix(FCS_KWIC_PREFIX, FCS_KWIC_NS); |
---|
228 | writer.writeStartElement(FCS_KWIC_NS, "kwic"); |
---|
229 | writer.writeNamespace(FCS_KWIC_PREFIX, FCS_KWIC_NS); |
---|
230 | |
---|
231 | writer.writeStartElement(FCS_KWIC_NS, "c"); |
---|
232 | writer.writeAttribute("type", "left"); |
---|
233 | if (left != null) { |
---|
234 | writer.writeCharacters(left); |
---|
235 | } |
---|
236 | writer.writeEndElement(); // "c" element |
---|
237 | |
---|
238 | writer.writeStartElement(FCS_KWIC_NS, "kw"); |
---|
239 | writer.writeCharacters(keyword); |
---|
240 | writer.writeEndElement(); // "kw" element |
---|
241 | |
---|
242 | writer.writeStartElement(FCS_KWIC_NS, "c"); |
---|
243 | writer.writeAttribute("type", "right"); |
---|
244 | if (right != null) { |
---|
245 | writer.writeCharacters(right); |
---|
246 | } |
---|
247 | writer.writeEndElement(); // "c" element |
---|
248 | |
---|
249 | writer.writeEndElement(); // "kwic" element |
---|
250 | |
---|
251 | writeEndDataView(writer); |
---|
252 | } |
---|
253 | |
---|
254 | |
---|
255 | /** |
---|
256 | * Convince method for writing a record with a KWIC data view. The following |
---|
257 | * code (arguments omitted) would accomplish the same result: |
---|
258 | * |
---|
259 | * <pre> |
---|
260 | * ... |
---|
261 | * writeStartResource(...); |
---|
262 | * writeKWICDataView(...); |
---|
263 | * writeEndResource(...); |
---|
264 | * ... |
---|
265 | * </pre> |
---|
266 | * |
---|
267 | * @param writer |
---|
268 | * the {@link XMLStreamWriter} to be used |
---|
269 | * @param pid |
---|
270 | * the persistent identifier of this resource or |
---|
271 | * <code>null</code>, if not applicable |
---|
272 | * @param ref |
---|
273 | * the reference of this resource or <code>null</code>, if not |
---|
274 | * applicable |
---|
275 | * @param left |
---|
276 | * the left context of the KWIC or <code>null</code> if not |
---|
277 | * applicable |
---|
278 | * @param keyword |
---|
279 | * the keyword of the KWIC |
---|
280 | * @param right |
---|
281 | * the right context of the KWIC or <code>null</code> if not |
---|
282 | * applicable |
---|
283 | * @throws XMLStreamException |
---|
284 | * if an error occurred |
---|
285 | * @deprecated The the HITS data view instead. |
---|
286 | */ |
---|
287 | @Deprecated |
---|
288 | public static void writeResourceWithKWICDataView(XMLStreamWriter writer, |
---|
289 | String pid, String ref, String left, String keyword, String right) |
---|
290 | throws XMLStreamException { |
---|
291 | if (writer == null) { |
---|
292 | throw new NullPointerException("writer == null"); |
---|
293 | } |
---|
294 | |
---|
295 | writeStartResource(writer, pid, ref); |
---|
296 | writeKWICDataView(writer, left, keyword, right); |
---|
297 | writeEndResource(writer); |
---|
298 | } |
---|
299 | |
---|
300 | |
---|
301 | |
---|
302 | /** |
---|
303 | * Convince method to write a simple HITS data view. It automatically |
---|
304 | * performs the calls to |
---|
305 | * {@link #writeStartDataView(XMLStreamWriter, String)} and |
---|
306 | * {@link #writeEndDataView(XMLStreamWriter)}. |
---|
307 | * |
---|
308 | * @param writer |
---|
309 | * the {@link XMLStreamWriter} to be used |
---|
310 | * @param left |
---|
311 | * the left context of the hit or <code>null</code> if not |
---|
312 | * applicable |
---|
313 | * @param hit |
---|
314 | * the actual hit, that will be highlighted |
---|
315 | * @param right |
---|
316 | * the right context of the hit or <code>null</code> if not |
---|
317 | * applicable |
---|
318 | * @throws XMLStreamException |
---|
319 | * if an error occurred |
---|
320 | */ |
---|
321 | public static void writeHitsDataView(XMLStreamWriter writer, String left, |
---|
322 | String hit, String right) throws XMLStreamException { |
---|
323 | if (writer == null) { |
---|
324 | throw new NullPointerException("writer == null"); |
---|
325 | } |
---|
326 | if (hit == null) { |
---|
327 | throw new NullPointerException("hit == null"); |
---|
328 | } |
---|
329 | |
---|
330 | writeStartDataView(writer, FCS_HITS_MIMETYPE); |
---|
331 | |
---|
332 | // actual "hits" data view |
---|
333 | writer.setPrefix(FCS_HITS_PREFIX, FCS_HITS_NS); |
---|
334 | writer.writeStartElement(FCS_HITS_NS, "Result"); |
---|
335 | writer.writeNamespace(FCS_HITS_PREFIX, FCS_HITS_NS); |
---|
336 | |
---|
337 | if ((left != null) && !left.isEmpty()) { |
---|
338 | writer.writeCharacters(left); |
---|
339 | } |
---|
340 | |
---|
341 | writer.writeStartElement(FCS_HITS_NS, "Hit"); |
---|
342 | writer.writeCharacters(hit); |
---|
343 | writer.writeEndElement(); // "Hit" element |
---|
344 | |
---|
345 | if ((right != null) && !right.isEmpty()) { |
---|
346 | writer.writeCharacters(right); |
---|
347 | } |
---|
348 | |
---|
349 | writer.writeEndElement(); // "Result" element |
---|
350 | |
---|
351 | writeEndDataView(writer); |
---|
352 | } |
---|
353 | |
---|
354 | |
---|
355 | /** |
---|
356 | * Convince method for writing a record with a KWIC data view. The following |
---|
357 | * code (arguments omitted) would accomplish the same result: |
---|
358 | * |
---|
359 | * <pre> |
---|
360 | * ... |
---|
361 | * writeStartResource(...); |
---|
362 | * writeHitsDataView(...); |
---|
363 | * writeEndResource(...); |
---|
364 | * ... |
---|
365 | * </pre> |
---|
366 | * |
---|
367 | * @param writer |
---|
368 | * the {@link XMLStreamWriter} to be used |
---|
369 | * @param pid |
---|
370 | * the persistent identifier of this resource or |
---|
371 | * <code>null</code>, if not applicable |
---|
372 | * @param ref |
---|
373 | * the reference of this resource or <code>null</code>, if not |
---|
374 | * applicable |
---|
375 | * @param left |
---|
376 | * the left context of the hit or <code>null</code> if not |
---|
377 | * applicable |
---|
378 | * @param hit |
---|
379 | * the actual hit, that will be highlighted |
---|
380 | * @param right |
---|
381 | * the right context of the hit or <code>null</code> if not |
---|
382 | * applicable |
---|
383 | * @throws XMLStreamException |
---|
384 | * if an error occurred |
---|
385 | */ |
---|
386 | public static void writeResourceWithHitsDataView(XMLStreamWriter writer, |
---|
387 | String pid, String ref, String left, String hit, String right) |
---|
388 | throws XMLStreamException { |
---|
389 | if (writer == null) { |
---|
390 | throw new NullPointerException("writer == null"); |
---|
391 | } |
---|
392 | |
---|
393 | writeStartResource(writer, pid, ref); |
---|
394 | writeHitsDataView(writer, left, hit, right); |
---|
395 | writeEndResource(writer); |
---|
396 | } |
---|
397 | |
---|
398 | |
---|
399 | /** |
---|
400 | * Convince method to write a simple HITS data view. It automatically |
---|
401 | * performs the calls to |
---|
402 | * {@link #writeStartDataView(XMLStreamWriter, String)} and |
---|
403 | * {@link #writeEndDataView(XMLStreamWriter)}. |
---|
404 | * |
---|
405 | * @param writer |
---|
406 | * the {@link XMLStreamWriter} to be used |
---|
407 | * @param text |
---|
408 | * the text content of the hit |
---|
409 | * @param hits |
---|
410 | * an even-element array containing tuples for the hit markers in |
---|
411 | * the text content |
---|
412 | * @param secondIsLength |
---|
413 | * if <code>true</code> the second element of each tuple in this |
---|
414 | * <code>hits</code> array is interpreted as an length; if |
---|
415 | * <code>false</code> it is interpreted as an end-offset |
---|
416 | * @throws XMLStreamException |
---|
417 | * if an error occurred |
---|
418 | */ |
---|
419 | public static void writeHitsDataView(XMLStreamWriter writer, String text, |
---|
420 | int[] hits, boolean secondIsLength) throws XMLStreamException { |
---|
421 | if (writer == null) { |
---|
422 | throw new NullPointerException("writer == null"); |
---|
423 | } |
---|
424 | if (text == null) { |
---|
425 | throw new NullPointerException("text == null"); |
---|
426 | } |
---|
427 | if (hits == null) { |
---|
428 | throw new NullPointerException("text == null"); |
---|
429 | } |
---|
430 | if ((hits.length == 0) || ((hits.length % 2) != 0)) { |
---|
431 | throw new NullPointerException("length of hits array must " + |
---|
432 | "contain an even number of elements"); |
---|
433 | } |
---|
434 | |
---|
435 | writeStartDataView(writer, FCS_HITS_MIMETYPE); |
---|
436 | |
---|
437 | // actual "hits" data view |
---|
438 | writer.setPrefix(FCS_HITS_PREFIX, FCS_HITS_NS); |
---|
439 | writer.writeStartElement(FCS_HITS_NS, "Result"); |
---|
440 | writer.writeNamespace(FCS_HITS_PREFIX, FCS_HITS_NS); |
---|
441 | |
---|
442 | int pos = 0; |
---|
443 | for (int i = 0; i < hits.length; i += 2) { |
---|
444 | int start = hits[i]; |
---|
445 | int end = hits[i + 1]; |
---|
446 | |
---|
447 | if ((start < 0) && (start > text.length())) { |
---|
448 | throw new IllegalArgumentException("start index out of " + |
---|
449 | "bounds: start=" + start); |
---|
450 | } |
---|
451 | if (secondIsLength) { |
---|
452 | if (end < 1) { |
---|
453 | throw new IllegalArgumentException( |
---|
454 | "length must be larger than 0: length = " + end); |
---|
455 | } |
---|
456 | end += start; |
---|
457 | } |
---|
458 | if (start >= end) { |
---|
459 | throw new IllegalArgumentException("end offset must be " + |
---|
460 | "larger then start offset: start=" + start + |
---|
461 | ", end="+ end); |
---|
462 | } |
---|
463 | |
---|
464 | if (start > pos) { |
---|
465 | String s = text.substring(pos, start); |
---|
466 | writer.writeCharacters(s); |
---|
467 | } |
---|
468 | writer.writeStartElement(FCS_HITS_NS, "Hit"); |
---|
469 | writer.writeCharacters(text.substring(start, end)); |
---|
470 | writer.writeEndElement(); // "Hits" element |
---|
471 | pos = end; |
---|
472 | } |
---|
473 | if (pos < text.length() - 1) { |
---|
474 | writer.writeCharacters(text.substring(pos)); |
---|
475 | } |
---|
476 | |
---|
477 | writer.writeEndElement(); // "Result" element |
---|
478 | |
---|
479 | writeEndDataView(writer); |
---|
480 | } |
---|
481 | |
---|
482 | |
---|
483 | /** |
---|
484 | * Convince method to write a simple HITS data view. It automatically |
---|
485 | * performs the calls to |
---|
486 | * {@link #writeStartDataView(XMLStreamWriter, String)} and |
---|
487 | * {@link #writeEndDataView(XMLStreamWriter)}. |
---|
488 | * |
---|
489 | * <pre> |
---|
490 | * ... |
---|
491 | * writeStartResource(...); |
---|
492 | * writeHitsDataView(...); |
---|
493 | * writeEndResource(...); |
---|
494 | * ... |
---|
495 | * </pre> |
---|
496 | * |
---|
497 | * @param writer |
---|
498 | * the {@link XMLStreamWriter} to be used |
---|
499 | * @param pid |
---|
500 | * the persistent identifier of this resource or |
---|
501 | * <code>null</code>, if not applicable |
---|
502 | * @param ref |
---|
503 | * the reference of this resource or <code>null</code>, if not |
---|
504 | * applicable |
---|
505 | * @param text |
---|
506 | * the text content of the hit |
---|
507 | * @param hits |
---|
508 | * an even-element array containing tuples for the hit markers in |
---|
509 | * the text content |
---|
510 | * @param secondIsLength |
---|
511 | * if <code>true</code> the second element of each tuple in this |
---|
512 | * <code>hits</code> array is interpreted as an length; if |
---|
513 | * <code>false</code> it is interpreted as an end-offset |
---|
514 | * @throws XMLStreamException |
---|
515 | * if an error occurred |
---|
516 | */ |
---|
517 | public static void writeResourceWithHitsDataView(XMLStreamWriter writer, |
---|
518 | String pid, String ref, String text, |
---|
519 | int[] hits, boolean secondIsLength) |
---|
520 | throws XMLStreamException { |
---|
521 | if (writer == null) { |
---|
522 | throw new NullPointerException("writer == null"); |
---|
523 | } |
---|
524 | |
---|
525 | writeStartResource(writer, pid, ref); |
---|
526 | writeHitsDataView(writer, text, hits, secondIsLength); |
---|
527 | writeEndResource(writer); |
---|
528 | } |
---|
529 | |
---|
530 | } // class XMLStreamWriterHelper |
---|