1 | package eu.clarin.cmdi.vlo.importer; |
---|
2 | |
---|
3 | import eu.clarin.cmdi.vlo.CommonUtils; |
---|
4 | import eu.clarin.cmdi.vlo.FacetConstants; |
---|
5 | import eu.clarin.cmdi.vlo.config.DataRoot; |
---|
6 | import eu.clarin.cmdi.vlo.config.VloConfig; |
---|
7 | import eu.clarin.cmdi.vlo.config.XmlVloConfigFactory; |
---|
8 | |
---|
9 | import java.io.File; |
---|
10 | import java.io.IOException; |
---|
11 | import java.net.MalformedURLException; |
---|
12 | import java.net.URL; |
---|
13 | import java.text.SimpleDateFormat; |
---|
14 | import java.util.ArrayList; |
---|
15 | import java.util.Collection; |
---|
16 | import java.util.Date; |
---|
17 | import java.util.HashMap; |
---|
18 | import java.util.HashSet; |
---|
19 | import java.util.List; |
---|
20 | import java.util.Map; |
---|
21 | import java.util.Set; |
---|
22 | |
---|
23 | import org.apache.commons.cli.CommandLine; |
---|
24 | import org.apache.commons.cli.CommandLineParser; |
---|
25 | import org.apache.commons.cli.Options; |
---|
26 | import org.apache.commons.cli.PosixParser; |
---|
27 | import org.apache.commons.io.FileUtils; |
---|
28 | import org.apache.solr.client.solrj.SolrServerException; |
---|
29 | import org.apache.solr.client.solrj.impl.ConcurrentUpdateSolrServer; |
---|
30 | import org.apache.solr.client.solrj.util.ClientUtils; |
---|
31 | import org.apache.solr.common.SolrInputDocument; |
---|
32 | import org.apache.solr.common.params.MapSolrParams; |
---|
33 | import org.apache.solr.common.params.SolrParams; |
---|
34 | import org.slf4j.Logger; |
---|
35 | import org.slf4j.LoggerFactory; |
---|
36 | |
---|
37 | /** |
---|
38 | * The main metadataImporter class. Also contains the main function. |
---|
39 | * |
---|
40 | * The metadataimporter reads all the config files and then, for each |
---|
41 | * metadatafile in each defined directory structure parses and imports them as |
---|
42 | * defined in the configuration. The startImport function starts the importing |
---|
43 | * and so on. |
---|
44 | */ |
---|
45 | @SuppressWarnings({"serial"}) |
---|
46 | public class MetadataImporter { |
---|
47 | |
---|
48 | /** |
---|
49 | * Defines which files to try and parse. In this case all files ending in |
---|
50 | * "xml" or "cmdi". |
---|
51 | */ |
---|
52 | private static final String[] VALID_CMDI_EXTENSIONS = new String[]{"xml", "cmdi"}; |
---|
53 | |
---|
54 | /** |
---|
55 | * Log log log log |
---|
56 | */ |
---|
57 | protected final static Logger LOG = LoggerFactory.getLogger(MetadataImporter.class); |
---|
58 | /** |
---|
59 | * Some place to store errors. |
---|
60 | */ |
---|
61 | private static Throwable serverError; |
---|
62 | /** |
---|
63 | * the solr server. |
---|
64 | */ |
---|
65 | private ConcurrentUpdateSolrServer solrServer; |
---|
66 | /** |
---|
67 | * Defines the post-processor associations. At import, for each facet value, |
---|
68 | * this map is checked and all postprocessors associated with the facet |
---|
69 | * _type_ are applied to the value before storing the new value in the solr |
---|
70 | * document. |
---|
71 | */ |
---|
72 | final static Map<String, PostProcessor> POST_PROCESSORS = new HashMap<String, PostProcessor>(); |
---|
73 | |
---|
74 | static { |
---|
75 | POST_PROCESSORS.put(FacetConstants.FIELD_ID, new IdPostProcessor()); |
---|
76 | POST_PROCESSORS.put(FacetConstants.FIELD_CONTINENT, new ContinentNamePostProcessor()); |
---|
77 | POST_PROCESSORS.put(FacetConstants.FIELD_COUNTRY, new CountryNamePostProcessor()); |
---|
78 | POST_PROCESSORS.put(FacetConstants.FIELD_LANGUAGE, new LanguageNamePostProcessor()); |
---|
79 | POST_PROCESSORS.put(FacetConstants.FIELD_LANGUAGES, new LanguageLinkPostProcessor()); |
---|
80 | POST_PROCESSORS.put(FacetConstants.FIELD_LANGUAGE_CODE, new LanguageCodePostProcessor()); |
---|
81 | POST_PROCESSORS.put(FacetConstants.FIELD_LICENSE, new LicensePostProcessor()); |
---|
82 | POST_PROCESSORS.put(FacetConstants.FIELD_ORGANISATION, new OrganisationPostProcessor()); |
---|
83 | POST_PROCESSORS.put(FacetConstants.FIELD_YEAR, new YearPostProcessor()); |
---|
84 | POST_PROCESSORS.put(FacetConstants.FIELD_NATIONAL_PROJECT, new NationalProjectPostProcessor()); |
---|
85 | POST_PROCESSORS.put(FacetConstants.FIELD_CLARIN_PROFILE, new CMDIComponentProfileNamePostProcessor()); |
---|
86 | POST_PROCESSORS.put(FacetConstants.FIELD_RESOURCE_CLASS, new ResourceClassPostProcessor()); |
---|
87 | } |
---|
88 | |
---|
89 | /** |
---|
90 | * Constructor |
---|
91 | */ |
---|
92 | public MetadataImporter() {} |
---|
93 | |
---|
94 | /** |
---|
95 | * Contains MDSelflinks (usually). Just to know what we have already done. |
---|
96 | */ |
---|
97 | protected final Set<String> processedIds = new HashSet<String>(); |
---|
98 | /** |
---|
99 | * Some caching for solr documents (we are more efficient if we ram a whole |
---|
100 | * bunch to the solr server at once. |
---|
101 | */ |
---|
102 | protected List<SolrInputDocument> docs = new ArrayList<SolrInputDocument>(); |
---|
103 | |
---|
104 | // SOME STATS |
---|
105 | protected int nrOFDocumentsSend; |
---|
106 | protected int nrOfFilesAnalyzed = 0; |
---|
107 | protected int nrOfFilesWithoutId = 0; |
---|
108 | protected int nrOfIgnoredFiles = 0; |
---|
109 | protected int nrOfFilesWithError = 0; |
---|
110 | |
---|
111 | /** |
---|
112 | * Retrieve all files with VALID_CMDI_EXTENSIONS from all DataRoot entries |
---|
113 | * and starts processing for every single file |
---|
114 | * |
---|
115 | * @throws MalformedURLException |
---|
116 | */ |
---|
117 | void startImport() throws MalformedURLException { |
---|
118 | |
---|
119 | initSolrServer(); |
---|
120 | List<DataRoot> dataRoots = checkDataRoots(); |
---|
121 | long start = System.currentTimeMillis(); |
---|
122 | try { |
---|
123 | // Delete the whole Solr db |
---|
124 | if (config.getDeleteAllFirst()) { |
---|
125 | LOG.info("Deleting original data..."); |
---|
126 | solrServer.deleteByQuery("*:*"); |
---|
127 | solrServer.commit(); |
---|
128 | LOG.info("Deleting original data done."); |
---|
129 | } |
---|
130 | |
---|
131 | // Import the specified data roots |
---|
132 | for (DataRoot dataRoot : dataRoots) { |
---|
133 | LOG.info("Start of processing: " + dataRoot.getOriginName()); |
---|
134 | if (dataRoot.deleteFirst()) { |
---|
135 | LOG.info("Deleting data for data provider: " + dataRoot.getOriginName()); |
---|
136 | solrServer.deleteByQuery(FacetConstants.FIELD_DATA_PROVIDER + ":" + ClientUtils.escapeQueryChars(dataRoot.getOriginName())); |
---|
137 | LOG.info("Deleting data of provider done."); |
---|
138 | } |
---|
139 | CMDIDataProcessor processor = new CMDIParserVTDXML(POST_PROCESSORS, false); |
---|
140 | List<File> files = getFilesFromDataRoot(dataRoot.getRootFile()); |
---|
141 | for (File file : files) { |
---|
142 | if (config.getMaxFileSize() > 0 |
---|
143 | && file.length() > config.getMaxFileSize()) { |
---|
144 | LOG.info("Skipping " + file.getAbsolutePath() + " because it is too large."); |
---|
145 | } else { |
---|
146 | LOG.debug("PROCESSING FILE: {}", file.getAbsolutePath()); |
---|
147 | processCmdi(file, dataRoot, processor); |
---|
148 | } |
---|
149 | } |
---|
150 | if (!docs.isEmpty()) { |
---|
151 | sendDocs(); |
---|
152 | } |
---|
153 | LOG.info("End of processing: " + dataRoot.getOriginName()); |
---|
154 | } |
---|
155 | |
---|
156 | // delete outdated entries (based on maxDaysInSolr parameter) |
---|
157 | if (config.getMaxDaysInSolr() > 0 && config.getDeleteAllFirst() == false) { |
---|
158 | LOG.info("Deleting old files that were not seen for more than " + config.getMaxDaysInSolr() + " days..."); |
---|
159 | solrServer.deleteByQuery(FacetConstants.FIELD_LAST_SEEN + ":[* TO NOW-" + config.getMaxDaysInSolr() + "DAYS]"); |
---|
160 | LOG.info("Deleting old files done."); |
---|
161 | } |
---|
162 | } catch (SolrServerException e) { |
---|
163 | LOG.error("error updating files:\n", e); |
---|
164 | LOG.error("Also see vlo_solr server logs for more information"); |
---|
165 | } catch (IOException e) { |
---|
166 | LOG.error("error updating files:\n", e); |
---|
167 | } finally { |
---|
168 | try { |
---|
169 | if (solrServer != null) { |
---|
170 | solrServer.commit(); |
---|
171 | buildSuggesterIndex(); |
---|
172 | } |
---|
173 | } catch (SolrServerException e) { |
---|
174 | LOG.error("cannot commit:\n", e); |
---|
175 | } catch (IOException e) { |
---|
176 | LOG.error("cannot commit:\n", e); |
---|
177 | } |
---|
178 | } |
---|
179 | long took = (System.currentTimeMillis() - start) / 1000; |
---|
180 | LOG.info("Found " + nrOfFilesWithoutId + " file(s) without an id. (id is generated based on fileName but that may not be unique)"); |
---|
181 | LOG.info("Found " + nrOfFilesWithError + " file(s) with errors."); |
---|
182 | LOG.info("Found " + nrOfIgnoredFiles |
---|
183 | + " file(s) that where ignored (files without resources or any link to a search service or landing page are ignored)."); |
---|
184 | LOG.info("Update of " + nrOFDocumentsSend + " took " + took + " secs. Total nr of files analyzed " + nrOfFilesAnalyzed); |
---|
185 | solrServer.shutdown(); |
---|
186 | } |
---|
187 | |
---|
188 | /** |
---|
189 | * Check a List of DataRoots for existence of RootFile (typically parent |
---|
190 | * directory of metadata files) |
---|
191 | * |
---|
192 | * @return |
---|
193 | */ |
---|
194 | protected List<DataRoot> checkDataRoots() { |
---|
195 | List<DataRoot> dataRoots = config.getDataRoots(); |
---|
196 | for (DataRoot dataRoot : dataRoots) { |
---|
197 | if (!dataRoot.getRootFile().exists()) { |
---|
198 | LOG.error("Root file " + dataRoot.getRootFile() + " does not exist. Probable configuration error so stopping import."); |
---|
199 | System.exit(1); |
---|
200 | } |
---|
201 | } |
---|
202 | return dataRoots; |
---|
203 | } |
---|
204 | |
---|
205 | /** |
---|
206 | * Get the rootFile or all files with VALID_CMDI_EXTENSIONS if rootFile is a |
---|
207 | * directory |
---|
208 | * |
---|
209 | * @param rootFile |
---|
210 | * @return List with the rootFile or all contained files if rootFile is a |
---|
211 | * directory |
---|
212 | */ |
---|
213 | protected List<File> getFilesFromDataRoot(File rootFile) { |
---|
214 | List<File> result = new ArrayList<File>(); |
---|
215 | if (rootFile.isFile()) { |
---|
216 | result.add(rootFile); |
---|
217 | } else { |
---|
218 | Collection<File> listFiles = FileUtils.listFiles(rootFile, VALID_CMDI_EXTENSIONS, true); |
---|
219 | result.addAll(listFiles); |
---|
220 | } |
---|
221 | return result; |
---|
222 | } |
---|
223 | |
---|
224 | /** |
---|
225 | * Create an interface to the SOLR server. |
---|
226 | * |
---|
227 | * After the interface has been created the importer can send documents to |
---|
228 | * the server. Sending documents involves a queue. The importer adds |
---|
229 | * documents to a queue, and dedicated threads will empty it, and |
---|
230 | * effectively store store the documents. |
---|
231 | * |
---|
232 | * @throws MalformedURLException |
---|
233 | */ |
---|
234 | protected void initSolrServer() throws MalformedURLException { |
---|
235 | String solrUrl = config.getSolrUrl(); |
---|
236 | LOG.info("Initializing Solr Server on " + solrUrl); |
---|
237 | |
---|
238 | /* Specify the number of documents in the queue that will trigger the |
---|
239 | * threads, two of them, emptying it. |
---|
240 | */ |
---|
241 | solrServer = new ConcurrentUpdateSolrServer(solrUrl, |
---|
242 | config.getMinDocsInSolrQueue(), 2) { |
---|
243 | /* |
---|
244 | * Let the super class method handle exceptions. Make the |
---|
245 | * exception available to the importer in the form of the |
---|
246 | * serverError variable. |
---|
247 | */ |
---|
248 | @Override |
---|
249 | public void handleError(Throwable exception) { |
---|
250 | super.handleError(exception); |
---|
251 | serverError = exception; |
---|
252 | } |
---|
253 | }; |
---|
254 | } |
---|
255 | |
---|
256 | /** |
---|
257 | * Process single CMDI file with CMDIDataProcessor |
---|
258 | * |
---|
259 | * @param file CMDI input file |
---|
260 | * @param dataOrigin |
---|
261 | * @param processor |
---|
262 | * @throws SolrServerException |
---|
263 | * @throws IOException |
---|
264 | */ |
---|
265 | protected void processCmdi(File file, DataRoot dataOrigin, CMDIDataProcessor processor) throws SolrServerException, IOException { |
---|
266 | nrOfFilesAnalyzed++; |
---|
267 | CMDIData cmdiData = null; |
---|
268 | try { |
---|
269 | cmdiData = processor.process(file); |
---|
270 | if (!idOk(cmdiData.getId())) { |
---|
271 | cmdiData.setId(dataOrigin.getOriginName() + "/" + file.getName()); //No id found in the metadata file so making one up based on the file name. Not quaranteed to be unique, but we have to set something. |
---|
272 | nrOfFilesWithoutId++; |
---|
273 | } |
---|
274 | } catch (Exception e) { |
---|
275 | LOG.error("error in file: {}", file, e); |
---|
276 | nrOfFilesWithError++; |
---|
277 | } |
---|
278 | if (cmdiData != null) { |
---|
279 | if (processedIds.add(cmdiData.getId())) { |
---|
280 | SolrInputDocument solrDocument = cmdiData.getSolrDocument(); |
---|
281 | if (solrDocument != null) { |
---|
282 | if (!cmdiData.getDataResources().isEmpty() || !cmdiData.getLandingPageResources().isEmpty() |
---|
283 | || !cmdiData.getSearchResources().isEmpty() || !cmdiData.getSearchPageResources().isEmpty() |
---|
284 | || cmdiData.getMetadataResources().isEmpty()) { |
---|
285 | // We only add metadata files that have |
---|
286 | // 1) data resources or |
---|
287 | // 2) a landing page or |
---|
288 | // 3) a search service (like SRU/CQL) or |
---|
289 | // 4) a search page or |
---|
290 | // 5) that have none of the above but also lack any metadata links (e.g. olac files describing a corpus with a link to the original archive). |
---|
291 | // Other files will have only metadata resources and are considered 'collection' metadata files they |
---|
292 | // are usually not very interesting (think imdi corpus files) and will not be included. |
---|
293 | updateDocument(solrDocument, cmdiData, file, dataOrigin); |
---|
294 | } else { |
---|
295 | nrOfIgnoredFiles++; |
---|
296 | } |
---|
297 | } |
---|
298 | } else { |
---|
299 | LOG.warn("Skipping {}, already processed id: {}", file, cmdiData.getId()); |
---|
300 | } |
---|
301 | } |
---|
302 | } |
---|
303 | |
---|
304 | /** |
---|
305 | * Check id for validness |
---|
306 | * |
---|
307 | * @param id |
---|
308 | * @return true if id is acceptable, false otherwise |
---|
309 | */ |
---|
310 | protected boolean idOk(String id) { |
---|
311 | return id != null && !id.isEmpty(); |
---|
312 | } |
---|
313 | |
---|
314 | /** |
---|
315 | * Adds some additional information from DataRoot to solrDocument, add |
---|
316 | * solrDocument to document list, submits list to SolrServer every 1000 |
---|
317 | * files |
---|
318 | * |
---|
319 | * @param solrDocument |
---|
320 | * @param cmdiData |
---|
321 | * @param file |
---|
322 | * @param dataOrigin |
---|
323 | * @throws SolrServerException |
---|
324 | * @throws IOException |
---|
325 | */ |
---|
326 | protected void updateDocument(SolrInputDocument solrDocument, CMDIData cmdiData, File file, DataRoot dataOrigin) throws SolrServerException, |
---|
327 | IOException { |
---|
328 | if (!solrDocument.containsKey(FacetConstants.FIELD_COLLECTION)) { |
---|
329 | solrDocument.addField(FacetConstants.FIELD_COLLECTION, dataOrigin.getOriginName()); |
---|
330 | } |
---|
331 | solrDocument.addField(FacetConstants.FIELD_DATA_PROVIDER, dataOrigin.getOriginName()); |
---|
332 | solrDocument.addField(FacetConstants.FIELD_ID, cmdiData.getId()); |
---|
333 | solrDocument.addField(FacetConstants.FIELD_FILENAME, file.getAbsolutePath()); |
---|
334 | |
---|
335 | String metadataSourceUrl = dataOrigin.getPrefix(); |
---|
336 | metadataSourceUrl += file.getAbsolutePath().substring(dataOrigin.getToStrip().length()); |
---|
337 | |
---|
338 | solrDocument.addField(FacetConstants.FIELD_COMPLETE_METADATA, metadataSourceUrl); |
---|
339 | |
---|
340 | // add SearchServices (should be CQL endpoint) |
---|
341 | for (Resource resource : cmdiData.getSearchResources()) { |
---|
342 | solrDocument.addField(FacetConstants.FIELD_SEARCH_SERVICE, resource.getResourceName()); |
---|
343 | } |
---|
344 | |
---|
345 | // add landing page resource |
---|
346 | for (Resource resource : cmdiData.getLandingPageResources()) { |
---|
347 | solrDocument.addField(FacetConstants.FIELD_LANDINGPAGE, resource.getResourceName()); |
---|
348 | } |
---|
349 | |
---|
350 | // add search page resource |
---|
351 | for (Resource resource : cmdiData.getSearchPageResources()) { |
---|
352 | solrDocument.addField(FacetConstants.FIELD_SEARCHPAGE, resource.getResourceName()); |
---|
353 | } |
---|
354 | |
---|
355 | // add timestamp |
---|
356 | Date dt = new Date(); |
---|
357 | SimpleDateFormat df = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss'Z'"); |
---|
358 | solrDocument.addField(FacetConstants.FIELD_LAST_SEEN, df.format(dt)); |
---|
359 | |
---|
360 | // add resource proxys |
---|
361 | addResourceData(solrDocument, cmdiData); |
---|
362 | |
---|
363 | LOG.debug("Adding document for submission to SOLR: {}", file); |
---|
364 | docs.add(solrDocument); |
---|
365 | if (docs.size() == config.getMaxDocsInList()) { |
---|
366 | sendDocs(); |
---|
367 | } |
---|
368 | } |
---|
369 | |
---|
370 | /** |
---|
371 | * Adds two fields FIELD_FORMAT and FIELD_RESOURCE. The Type can be |
---|
372 | * specified in the "ResourceType" element of an imdi file or possibly |
---|
373 | * overwritten by some more specific xpath (as in the LRT cmdi files). So if |
---|
374 | * a type is overwritten and already in the solrDocument we take that type. |
---|
375 | * |
---|
376 | * @param solrDocument |
---|
377 | * @param cmdiData |
---|
378 | */ |
---|
379 | protected void addResourceData(SolrInputDocument solrDocument, CMDIData cmdiData) { |
---|
380 | List<Object> fieldValues = solrDocument.containsKey(FacetConstants.FIELD_FORMAT) ? new ArrayList<Object>(solrDocument |
---|
381 | .getFieldValues(FacetConstants.FIELD_FORMAT)) : null; |
---|
382 | solrDocument.removeField(FacetConstants.FIELD_FORMAT); //Remove old values they might be overwritten. |
---|
383 | List<Resource> resources = cmdiData.getDataResources(); |
---|
384 | for (int i = 0; i < resources.size(); i++) { |
---|
385 | Resource resource = resources.get(i); |
---|
386 | String mimeType = resource.getMimeType(); |
---|
387 | if (mimeType == null) { |
---|
388 | if (fieldValues != null && i < fieldValues.size()) { |
---|
389 | mimeType = CommonUtils.normalizeMimeType(fieldValues.get(i).toString()); |
---|
390 | } else { |
---|
391 | mimeType = CommonUtils.normalizeMimeType(""); |
---|
392 | } |
---|
393 | } |
---|
394 | |
---|
395 | FormatPostProcessor processor = new FormatPostProcessor(); |
---|
396 | mimeType = processor.process(mimeType).get(0); |
---|
397 | |
---|
398 | // TODO check should probably be moved into Solr (by using some minimum length filter) |
---|
399 | if (!mimeType.equals("")) { |
---|
400 | solrDocument.addField(FacetConstants.FIELD_FORMAT, mimeType); |
---|
401 | } |
---|
402 | solrDocument.addField(FacetConstants.FIELD_RESOURCE, mimeType + FacetConstants.FIELD_RESOURCE_SPLIT_CHAR |
---|
403 | + resource.getResourceName()); |
---|
404 | } |
---|
405 | } |
---|
406 | |
---|
407 | /** |
---|
408 | * Send current list of SolrImputDocuments to SolrServer and clears list |
---|
409 | * afterwards |
---|
410 | * |
---|
411 | * @throws SolrServerException |
---|
412 | * @throws IOException |
---|
413 | */ |
---|
414 | protected void sendDocs() throws SolrServerException, IOException { |
---|
415 | LOG.info("Sending " + docs.size() + " docs to solr server. Total number of docs updated till now: " + nrOFDocumentsSend); |
---|
416 | nrOFDocumentsSend += docs.size(); |
---|
417 | solrServer.add(docs); |
---|
418 | if (serverError != null) { |
---|
419 | throw new SolrServerException(serverError); |
---|
420 | } |
---|
421 | docs = new ArrayList<SolrInputDocument>(); |
---|
422 | } |
---|
423 | |
---|
424 | /** |
---|
425 | * Builds suggester index for autocompletion |
---|
426 | * |
---|
427 | * @throws SolrServerException |
---|
428 | * @throws MalformedURLException |
---|
429 | */ |
---|
430 | private void buildSuggesterIndex() throws SolrServerException, MalformedURLException { |
---|
431 | LOG.info("Building index for autocompletion."); |
---|
432 | HashMap<String, String> paramMap = new HashMap<String, String>(); |
---|
433 | paramMap.put("qt", "/suggest"); |
---|
434 | paramMap.put("spellcheck.build", "true"); |
---|
435 | SolrParams params = new MapSolrParams(paramMap); |
---|
436 | solrServer.query(params); |
---|
437 | } |
---|
438 | |
---|
439 | public static VloConfig config; |
---|
440 | |
---|
441 | /** |
---|
442 | * @param args |
---|
443 | * @throws MalformedURLException |
---|
444 | * @throws IOException |
---|
445 | */ |
---|
446 | public static void main(String[] args) throws MalformedURLException, IOException { |
---|
447 | |
---|
448 | // path to the configuration file |
---|
449 | String configFile = null; |
---|
450 | |
---|
451 | // use the Apache cli framework for getting command line parameters |
---|
452 | Options options = new Options(); |
---|
453 | |
---|
454 | /** |
---|
455 | * Add a "c" option, the option indicating the specification of an XML |
---|
456 | * configuration file |
---|
457 | */ |
---|
458 | options.addOption("c", true, "-c <file> : use parameters specified in <file>"); |
---|
459 | |
---|
460 | CommandLineParser parser = new PosixParser(); |
---|
461 | |
---|
462 | try { |
---|
463 | // parse the command line arguments |
---|
464 | CommandLine cmd = parser.parse(options, args); |
---|
465 | if (cmd.hasOption("c")) { |
---|
466 | |
---|
467 | // the "c" option was specified, now get its value |
---|
468 | configFile = cmd.getOptionValue("c"); |
---|
469 | } |
---|
470 | |
---|
471 | } catch (org.apache.commons.cli.ParseException ex) { |
---|
472 | |
---|
473 | /** |
---|
474 | * Caught an exception caused by command line parsing. Try to get |
---|
475 | * the name of the configuration file by querying the system |
---|
476 | * property. |
---|
477 | */ |
---|
478 | String message = "Command line parsing failed. " + ex.getMessage(); |
---|
479 | LOG.error(message); |
---|
480 | System.err.println(message); |
---|
481 | } |
---|
482 | |
---|
483 | if (configFile == null) { |
---|
484 | |
---|
485 | String message; |
---|
486 | |
---|
487 | message = "Could not get config file name via the command line, trying the system properties."; |
---|
488 | LOG.info(message); |
---|
489 | |
---|
490 | String key; |
---|
491 | |
---|
492 | key = "configFile"; |
---|
493 | configFile = System.getProperty(key); |
---|
494 | } |
---|
495 | |
---|
496 | if (configFile == null) { |
---|
497 | |
---|
498 | String message; |
---|
499 | |
---|
500 | message = "Could not get filename as system property either - stopping."; |
---|
501 | LOG.error(message); |
---|
502 | } else { |
---|
503 | // read the configuration from the externally supplied file |
---|
504 | final URL configUrl; |
---|
505 | if (configFile.startsWith("file:")) { |
---|
506 | configUrl = new URL(configFile); |
---|
507 | } else { |
---|
508 | configUrl = new File(configFile).toURI().toURL(); |
---|
509 | } |
---|
510 | System.out.println("Reading configuration from " + configUrl.toString()); |
---|
511 | final XmlVloConfigFactory configFactory = new XmlVloConfigFactory(configUrl); |
---|
512 | MetadataImporter.config = configFactory.newConfig(); |
---|
513 | |
---|
514 | // optionally, modify the configuration here |
---|
515 | // create and start the importer |
---|
516 | MetadataImporter importer = new MetadataImporter(); |
---|
517 | importer.startImport(); |
---|
518 | |
---|
519 | // finished importing |
---|
520 | if (MetadataImporter.config.printMapping()) { |
---|
521 | File file = new File("xsdMapping.txt"); |
---|
522 | FacetMappingFactory.printMapping(file); |
---|
523 | LOG.info("Printed facetMapping in " + file); |
---|
524 | } |
---|
525 | } |
---|
526 | } |
---|
527 | } |
---|