1 | /**
|
---|
2 | *############################################################################
|
---|
3 | * A component of the Greenstone Librarian Interface, part of the Greenstone
|
---|
4 | * digital library suite from the New Zealand Digital Library Project at the
|
---|
5 | * University of Waikato, New Zealand.
|
---|
6 | *
|
---|
7 | * Author: Michael Dewsnip, NZDL Project, University of Waikato, NZ
|
---|
8 | *
|
---|
9 | * Copyright (C) 2004 New Zealand Digital Library Project
|
---|
10 | *
|
---|
11 | * This program is free software; you can redistribute it and/or modify
|
---|
12 | * it under the terms of the GNU General Public License as published by
|
---|
13 | * the Free Software Foundation; either version 2 of the License, or
|
---|
14 | * (at your option) any later version.
|
---|
15 | *
|
---|
16 | * This program is distributed in the hope that it will be useful,
|
---|
17 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
18 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
19 | * GNU General Public License for more details.
|
---|
20 | *
|
---|
21 | * You should have received a copy of the GNU General Public License
|
---|
22 | * along with this program; if not, write to the Free Software
|
---|
23 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
24 | *############################################################################
|
---|
25 | */
|
---|
26 |
|
---|
27 | package org.greenstone.gatherer.metadata;
|
---|
28 |
|
---|
29 |
|
---|
30 | import java.io.*;
|
---|
31 | import java.util.*;
|
---|
32 | import java.net.URLDecoder;
|
---|
33 | import org.greenstone.gatherer.DebugStream;
|
---|
34 | import org.greenstone.gatherer.util.Utility;
|
---|
35 |
|
---|
36 | //import org.greenstone.gatherer.feedback.Base64; // decode() from Base64 didn't work
|
---|
37 | import org.apache.commons.codec.binary.Base64; // decoding from Base64 works
|
---|
38 |
|
---|
39 | /** This class represents one doc.xml file */
|
---|
40 |
|
---|
41 | public abstract class DocXMLFile extends File
|
---|
42 | {
|
---|
43 | protected HashMap source_file_name_to_description_elements_mapping = new HashMap();
|
---|
44 |
|
---|
45 | protected final String MetadataWrap;
|
---|
46 | protected final String MetadataItem;
|
---|
47 |
|
---|
48 | protected final String FILE_RENAME_METHOD_NONE = "none";
|
---|
49 | protected final String FILE_RENAME_METHOD_URL = "url";
|
---|
50 | protected final String FILE_RENAME_METHOD_BASE64 = "base64";
|
---|
51 |
|
---|
52 | public DocXMLFile(String doc_xml_file_path, String metaWrap, String metaItem)
|
---|
53 | {
|
---|
54 | super(doc_xml_file_path);
|
---|
55 | this.MetadataWrap = metaWrap;
|
---|
56 | this.MetadataItem = metaItem;
|
---|
57 | }
|
---|
58 |
|
---|
59 | /**
|
---|
60 | * Checks if various versions of the file object's filename, denoted relatively by file_relative_path,
|
---|
61 | * occur in the source_file_name_to_description_elements_mapping map
|
---|
62 | */
|
---|
63 | private ArrayList findSourceFileMapKeyMatch(File file, String file_relative_path) {
|
---|
64 | ArrayList description_elements_list = null;
|
---|
65 |
|
---|
66 | ///System.err.println("Looking for key " + file_relative_path);
|
---|
67 | description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
|
---|
68 | if(description_elements_list != null) {
|
---|
69 | ///System.err.println(" Found key matching REGULAR filepath: " + file_relative_path);
|
---|
70 | return description_elements_list;
|
---|
71 | }
|
---|
72 | else if(!Utility.isWindows()) { // couldn't find a matching key, we're done
|
---|
73 | ///System.err.println("Unable to find ex.meta for regular file path form " + file_relative_path);
|
---|
74 | return null;
|
---|
75 | }
|
---|
76 |
|
---|
77 | // Now we can try windows short filename as map key
|
---|
78 |
|
---|
79 | String win_short_file_relative_path = "";
|
---|
80 | try{
|
---|
81 | win_short_file_relative_path = Utility.getWindowsShortFileName(file.getAbsolutePath());
|
---|
82 | //System.err.println("@@@ Searching for short file name: " + win_short_file_relative_path);
|
---|
83 | } catch(Exception e) { // we're done trying to find a matching key
|
---|
84 | System.err.println("Failed to convert to windows short file name: " + win_short_file_relative_path);
|
---|
85 | return null;
|
---|
86 | }
|
---|
87 |
|
---|
88 | // Got a windows short file name, lop off import folder again
|
---|
89 | int import_index = win_short_file_relative_path.indexOf("import");
|
---|
90 | if (import_index != -1) {
|
---|
91 | win_short_file_relative_path = win_short_file_relative_path.substring(import_index + "import".length() + 1);
|
---|
92 | }
|
---|
93 |
|
---|
94 | ///System.err.println("### Looking for Windows short file name |" + win_short_file_relative_path + "| in map of sourcefilenames to doc.xml's ex meta.");
|
---|
95 | description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(win_short_file_relative_path);
|
---|
96 | if (description_elements_list != null) {
|
---|
97 | ///System.err.println(" Found key matching FULL win shortfile path: " + win_short_file_relative_path);
|
---|
98 | return description_elements_list; // found
|
---|
99 | }
|
---|
100 |
|
---|
101 | // else, check whether a map key is matched by any REMAINING combination of windows shortfile path and regular path:
|
---|
102 | // - windows shortfilename's rel-dir-path with regular tailname
|
---|
103 | // - and regular rel-dir-path with windows shortfilename's tailname
|
---|
104 |
|
---|
105 | String shortFileTailName = win_short_file_relative_path;
|
---|
106 | String shortFileRelDirPath = "";
|
---|
107 | int lastSep = win_short_file_relative_path.lastIndexOf(File.separator);
|
---|
108 | if(lastSep != -1) {
|
---|
109 | shortFileTailName = win_short_file_relative_path.substring(lastSep+1);
|
---|
110 | shortFileRelDirPath = win_short_file_relative_path.substring(0, lastSep+1); // include the slash
|
---|
111 | }
|
---|
112 |
|
---|
113 | String fileTailName = file_relative_path;
|
---|
114 | String fileRelDirPath = "";
|
---|
115 | lastSep = file_relative_path.lastIndexOf(File.separator);
|
---|
116 | if(lastSep != -1) {
|
---|
117 | fileTailName = file_relative_path.substring(lastSep+1);
|
---|
118 | fileRelDirPath = file_relative_path.substring(0, lastSep+1); // include the slash
|
---|
119 | }
|
---|
120 |
|
---|
121 | String path = shortFileRelDirPath + fileTailName;
|
---|
122 | ///System.err.println("### Looking for Windows short file name |" + path + "| in map of sourcefilenames to doc.xml's ex meta.");
|
---|
123 | description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(path);
|
---|
124 |
|
---|
125 | if(description_elements_list != null) {
|
---|
126 | ///System.err.println(" Found key matching MIX of win shortfile path and regular path: " + path);
|
---|
127 | return description_elements_list; // found
|
---|
128 | }
|
---|
129 |
|
---|
130 | // try the other combination
|
---|
131 | path = fileRelDirPath + shortFileTailName;
|
---|
132 | ///System.err.println("### Looking for Windows short file name |" + path + "| in map of sourcefilenames to doc.xml's ex meta.");
|
---|
133 | description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(path);
|
---|
134 |
|
---|
135 | if(description_elements_list != null) {
|
---|
136 | ///System.err.println(" Found key matching MIX of regular path and win shortfile path: " + path);
|
---|
137 | return description_elements_list; // found
|
---|
138 | }
|
---|
139 |
|
---|
140 | // could not find gsdlsourcefilename in map
|
---|
141 | ///System.err.println("Unable to find ex.meta for regular file path form " + file_relative_path);
|
---|
142 | ///System.err.println(" Or for windows shortFile path form, or for combinations with regular file path form");
|
---|
143 |
|
---|
144 | return description_elements_list; // returns null at this point
|
---|
145 | }
|
---|
146 |
|
---|
147 |
|
---|
148 | public ArrayList getMetadataExtractedFromFile(File file)
|
---|
149 | {
|
---|
150 | // Build up a list of metadata extracted from this file
|
---|
151 | ArrayList metadata_values = new ArrayList();
|
---|
152 |
|
---|
153 | String file_relative_path = file.getAbsolutePath();
|
---|
154 | int import_index = file_relative_path.indexOf("import");
|
---|
155 | if (import_index != -1) {
|
---|
156 | file_relative_path = file_relative_path.substring(import_index + "import".length() + 1);
|
---|
157 | }
|
---|
158 |
|
---|
159 | ///for (Object relFilename : source_file_name_to_description_elements_mapping.keySet()) {
|
---|
160 | /// System.err.println("\n@@@ relFilename: " + relFilename);
|
---|
161 | ///}
|
---|
162 |
|
---|
163 | // Check whether this file (i.e. doc.xml or docmets.xml on inheritance) file contains extracted metadata for the specified file
|
---|
164 | //ArrayList description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
|
---|
165 | ArrayList description_elements_list = findSourceFileMapKeyMatch(file, file_relative_path);
|
---|
166 | if (description_elements_list == null) {
|
---|
167 | // ...it doesn't
|
---|
168 | return metadata_values; // we're done
|
---|
169 | }
|
---|
170 |
|
---|
171 | MetadataSet extracted_metadata_set = MetadataSetManager.getMetadataSet(MetadataSetManager.EXTRACTED_METADATA_NAMESPACE);
|
---|
172 |
|
---|
173 | // Parse the file
|
---|
174 | DebugStream.println("Applicable file: " + this);
|
---|
175 | try {
|
---|
176 | BufferedReader buffered_reader = new BufferedReader(new InputStreamReader(new FileInputStream(this), "UTF-8"));
|
---|
177 |
|
---|
178 | int description_element_num = 0;
|
---|
179 | int next_description_element_start = ((Integer) description_elements_list.get(description_element_num)).intValue();
|
---|
180 | boolean in_relevant_description_element = false;
|
---|
181 |
|
---|
182 | String line = null;
|
---|
183 | for (int line_num = 0; (line = buffered_reader.readLine()) != null; line_num++) {
|
---|
184 | // Check if this line contains the start of a relevant "Description" element
|
---|
185 | // (mets:xmlData in METS parlance, Description in GreenstoneArchive format)
|
---|
186 | if (line_num == next_description_element_start) {
|
---|
187 | in_relevant_description_element = true;
|
---|
188 | continue;
|
---|
189 | }
|
---|
190 |
|
---|
191 | // If we're not in a relevant Description element we don't care about anything
|
---|
192 | if (in_relevant_description_element == false) {
|
---|
193 | continue;
|
---|
194 | }
|
---|
195 |
|
---|
196 | // Check if this line contains the end of the relevant Description element
|
---|
197 | if (line.indexOf("</"+MetadataWrap+">") != -1) {
|
---|
198 | description_element_num++;
|
---|
199 | if (description_element_num == description_elements_list.size()) {
|
---|
200 | break;
|
---|
201 | }
|
---|
202 |
|
---|
203 | next_description_element_start = ((Integer) description_elements_list.get(description_element_num)).intValue();
|
---|
204 | in_relevant_description_element = false;
|
---|
205 | continue;
|
---|
206 | }
|
---|
207 |
|
---|
208 | // If this line doesn't contain a complete Metadata element, we're not interested
|
---|
209 | if (line.indexOf("<"+MetadataItem+" ") == -1 || line.indexOf("</"+MetadataItem+">") == -1) {
|
---|
210 | continue;
|
---|
211 | }
|
---|
212 |
|
---|
213 | // Extract the metadata element name
|
---|
214 | int name_index = line.indexOf(" name=\"") + " name=\"".length();
|
---|
215 | String metadata_element_name_full = line.substring(name_index, line.indexOf("\"", name_index));
|
---|
216 |
|
---|
217 | // If the metadata has a namespace it isn't extracted metadata, so we're not interested
|
---|
218 | // Actually, if it is ex. then we are interested
|
---|
219 | String metadata_set_namespace = MetadataTools.getMetadataSetNamespace(metadata_element_name_full);
|
---|
220 |
|
---|
221 | if (!metadata_set_namespace.equals("") && !metadata_set_namespace.equals("ex")) {
|
---|
222 | continue;
|
---|
223 | }
|
---|
224 |
|
---|
225 | // Extracted metadata!
|
---|
226 | // do it like this just in case we have ex.
|
---|
227 | String metadata_element_name = MetadataTools.getMetadataElementName(metadata_element_name_full);
|
---|
228 |
|
---|
229 | // We completely ignore bibliographic data
|
---|
230 | if (metadata_element_name.equals("SourceSegment")) {
|
---|
231 | buffered_reader.close();
|
---|
232 | return new ArrayList();
|
---|
233 | }
|
---|
234 |
|
---|
235 | // Ignore metadata starting with gsdl (used to be lowercase metadata and /metadata)
|
---|
236 | if (metadata_element_name.startsWith("gsdl")) {
|
---|
237 | continue;
|
---|
238 | }
|
---|
239 |
|
---|
240 | MetadataElement metadata_element = extracted_metadata_set.getMetadataElementWithName(metadata_element_name);
|
---|
241 |
|
---|
242 | // Value trees are not stored for extracted metadata, so create a new value tree node now
|
---|
243 | int value_index = line.indexOf(">", name_index) + ">".length();
|
---|
244 | String metadata_element_value = line.substring(value_index, line.lastIndexOf("</"+MetadataItem+">"));
|
---|
245 |
|
---|
246 | metadata_element.addMetadataValue(metadata_element_value);
|
---|
247 | MetadataValueTreeNode metadata_value_tree_node = metadata_element.getMetadataValueTreeNode(metadata_element_value);
|
---|
248 |
|
---|
249 | // Add the new metadata value to the list
|
---|
250 | MetadataValue metadata_value = new MetadataValue(metadata_element, metadata_value_tree_node);
|
---|
251 | metadata_values.add(metadata_value);
|
---|
252 | }
|
---|
253 |
|
---|
254 | buffered_reader.close();
|
---|
255 | }
|
---|
256 | catch (FileNotFoundException exception) {
|
---|
257 | DebugStream.printStackTrace(exception);
|
---|
258 | }
|
---|
259 | catch (IOException exception) {
|
---|
260 | DebugStream.printStackTrace(exception);
|
---|
261 | }
|
---|
262 |
|
---|
263 | return metadata_values;
|
---|
264 | }
|
---|
265 |
|
---|
266 |
|
---|
267 |
|
---|
268 |
|
---|
269 | /**
|
---|
270 | * Every file must be skimmed when a collection is opened, for two reasons:
|
---|
271 | * - To build a mapping from source file to its corresponding doc.xml file
|
---|
272 | * - To get a complete list of all extracted metadata elements
|
---|
273 | */
|
---|
274 | public void skimFile()
|
---|
275 | {
|
---|
276 | String fileRenameMethod = null;
|
---|
277 | String gsdlsourcefilename_value = null;
|
---|
278 | boolean is_unix_path = false;
|
---|
279 |
|
---|
280 | MetadataSet extracted_metadata_set = MetadataSetManager.getMetadataSet(MetadataSetManager.EXTRACTED_METADATA_NAMESPACE);
|
---|
281 |
|
---|
282 | // Skim the file as quickly as possible (don't parse as XML), looking at the Metadata elements
|
---|
283 | DebugStream.println("Skimming " + this + "...");
|
---|
284 | try {
|
---|
285 | BufferedReader buffered_reader = new BufferedReader(new InputStreamReader(new FileInputStream(this), "UTF-8"));
|
---|
286 | int description_element_start = -1;
|
---|
287 |
|
---|
288 | String line = null;
|
---|
289 | for (int line_num = 0; (line = buffered_reader.readLine()) != null; line_num++) {
|
---|
290 | // This line contains the start of a "MetadataWrap" element
|
---|
291 | // (mets:xmlData in METS parlance, Description in GreenstoneArchive format)
|
---|
292 | if (line.indexOf("<"+MetadataWrap+">") != -1) {
|
---|
293 | if (description_element_start != -1) {
|
---|
294 | System.err.println("Parse error: previous " + MetadataWrap + " element unfinished!");
|
---|
295 | }
|
---|
296 | description_element_start = line_num;
|
---|
297 | continue;
|
---|
298 | }
|
---|
299 |
|
---|
300 | // This line contains the end of a "MetadataWrap" element
|
---|
301 | if (line.indexOf("</"+MetadataWrap+">") != -1) {
|
---|
302 | if (description_element_start == -1) {
|
---|
303 | System.err.println("Parse error: "+MetadataWrap+" element unstarted!");
|
---|
304 | }
|
---|
305 | description_element_start = -1;
|
---|
306 | continue;
|
---|
307 | }
|
---|
308 |
|
---|
309 | // If we're not in a"MetadataWrap" element there shouldn't be any Metadata elements
|
---|
310 | if (description_element_start == -1) {
|
---|
311 | continue;
|
---|
312 | }
|
---|
313 |
|
---|
314 | // This line doesn't contain a Metadata element, so we're not interested
|
---|
315 | if (line.indexOf("<"+MetadataItem+" ") == -1) {
|
---|
316 | DebugStream.println("Warning: "+MetadataWrap+" element line doesn't contain Metadata element.");
|
---|
317 | continue;
|
---|
318 | }
|
---|
319 |
|
---|
320 | // Extract the metadata element name
|
---|
321 | int name_index = line.indexOf(" name=\"") + " name=\"".length();
|
---|
322 | String metadata_element_name_full = line.substring(name_index, line.indexOf("\"", name_index));
|
---|
323 |
|
---|
324 | // If the metadata has a namespace it isn't extracted metadata, so we're not interested
|
---|
325 | String metadata_set_namespace = MetadataTools.getMetadataSetNamespace(metadata_element_name_full);
|
---|
326 | if (!metadata_set_namespace.equals("") && !metadata_set_namespace.equals("ex")) {
|
---|
327 | continue;
|
---|
328 | }
|
---|
329 |
|
---|
330 | // Extracted metadata! May have ex. so make sure we remove that
|
---|
331 | String metadata_element_name = MetadataTools.getMetadataElementName(metadata_element_name_full);
|
---|
332 | if(metadata_element_name.equals("gsdlsourcefilerenamemethod")) {
|
---|
333 | // Extract the element value
|
---|
334 | int value_index = line.indexOf(">", name_index) + ">".length();
|
---|
335 | fileRenameMethod = line.substring(value_index, line.indexOf("<", value_index));
|
---|
336 | }
|
---|
337 |
|
---|
338 | // Note which file this is for
|
---|
339 | else if (metadata_element_name.equals("gsdlsourcefilename")) {
|
---|
340 | // Extract the gsdlsourcefilename element value
|
---|
341 | int value_index = line.indexOf(">", name_index) + ">".length();
|
---|
342 | gsdlsourcefilename_value = line.substring(value_index, line.indexOf("<", value_index));
|
---|
343 |
|
---|
344 | // We're only interested in the path relative to the import folder
|
---|
345 | int import_index = gsdlsourcefilename_value.indexOf("import");
|
---|
346 | if (import_index != -1) {
|
---|
347 | gsdlsourcefilename_value = gsdlsourcefilename_value.substring(import_index + "import".length());
|
---|
348 |
|
---|
349 | is_unix_path = gsdlsourcefilename_value.startsWith("/");
|
---|
350 | gsdlsourcefilename_value = gsdlsourcefilename_value.substring(1);
|
---|
351 |
|
---|
352 | // (Will decode gsdlsourcefilename at end of this method, once we know
|
---|
353 | // for certain the fileRenameMethod that was used to encode it.)
|
---|
354 |
|
---|
355 | // Make sure the path matches the OS that is running
|
---|
356 | if (is_unix_path && Utility.isWindows()) {
|
---|
357 | // Convert path from Unix to Windows
|
---|
358 | gsdlsourcefilename_value = gsdlsourcefilename_value.replaceAll("\\/", "\\\\");
|
---|
359 | }
|
---|
360 | else if (!is_unix_path && !Utility.isWindows()) {
|
---|
361 | // Convert path from Windows to Unix
|
---|
362 | gsdlsourcefilename_value = gsdlsourcefilename_value.replaceAll("\\\\", "/");
|
---|
363 | }
|
---|
364 |
|
---|
365 | ///System.err.println("@@@@ Found gsdlsourcefilename: " + gsdlsourcefilename_value);
|
---|
366 | // Remember this for quick access later
|
---|
367 | if (source_file_name_to_description_elements_mapping.get(gsdlsourcefilename_value) == null) {
|
---|
368 | source_file_name_to_description_elements_mapping.put(gsdlsourcefilename_value, new ArrayList());
|
---|
369 | }
|
---|
370 |
|
---|
371 | ((ArrayList) source_file_name_to_description_elements_mapping.get(gsdlsourcefilename_value)).add(new Integer(description_element_start));
|
---|
372 | }
|
---|
373 |
|
---|
374 | // Warn about an odd gsdlsourcefilename, except if it is the Greenstone "tmp" directory or
|
---|
375 | // (as in the case of using FLI) if it is the etc/collect.cfg or etc/collectionConfig.xml file
|
---|
376 | // which are the gsdlsourcefilenames for the fedora digital object representing a collection.
|
---|
377 | // This (tmp dir) is true when the source files come from a zip file processed by ZIPPlug, for example
|
---|
378 | else if (gsdlsourcefilename_value.indexOf("tmp") == -1
|
---|
379 | && !gsdlsourcefilename_value.endsWith("collect.cfg")
|
---|
380 | && !gsdlsourcefilename_value.endsWith("collectionConfig.xml")) {
|
---|
381 | // We don't really know what is going on...
|
---|
382 | System.err.println("Warning: Could not understand gsdlsourcefilename " + gsdlsourcefilename_value);
|
---|
383 | }
|
---|
384 | }
|
---|
385 |
|
---|
386 | // Ignore metadata starting with gsdl (used to be lowercase metadata and /metadata)
|
---|
387 | if (metadata_element_name.startsWith("gsdl")) {
|
---|
388 | continue;
|
---|
389 | }
|
---|
390 |
|
---|
391 | MetadataElement metadata_element = extracted_metadata_set.getMetadataElementWithName(metadata_element_name);
|
---|
392 | if (metadata_element == null) {
|
---|
393 | // This element isn't defined in ex.mds, so create it for this session
|
---|
394 | DebugStream.println("Extracted metadata element not defined: " + metadata_element_name);
|
---|
395 | extracted_metadata_set.addMetadataElementForThisSession(metadata_element_name);
|
---|
396 | }
|
---|
397 | }
|
---|
398 |
|
---|
399 | buffered_reader.close();
|
---|
400 |
|
---|
401 | // Now that we're done skimming, we actually need to decode gsdlsourcefilename
|
---|
402 | // based on whatever fileRenameMethod was used to encode it, so that we can
|
---|
403 | // at last properly compare properly against filenames on the file system
|
---|
404 | // in order to load the correct ex.meta for the file.
|
---|
405 | // Now that we should have both gsdlsourcefilename AND fileRenameMethod set,
|
---|
406 | // we can finally perform the decoding of gsdlsourcefilename.
|
---|
407 | if(fileRenameMethod == null) {
|
---|
408 | fileRenameMethod = FILE_RENAME_METHOD_URL; // default for building
|
---|
409 | }
|
---|
410 | // If gsdlsourcefilename was encoded, we remove it from the map under its encoded
|
---|
411 | // filename, decode it and add it back into map using its decoded filename.
|
---|
412 | if(!fileRenameMethod.equals(FILE_RENAME_METHOD_NONE)) {
|
---|
413 | ArrayList value_list = (ArrayList) source_file_name_to_description_elements_mapping.remove(gsdlsourcefilename_value);
|
---|
414 | gsdlsourcefilename_value = decodeSourceFilename(gsdlsourcefilename_value, fileRenameMethod, is_unix_path);
|
---|
415 | source_file_name_to_description_elements_mapping.put(gsdlsourcefilename_value, value_list);
|
---|
416 | }
|
---|
417 | }
|
---|
418 | catch (FileNotFoundException exception) {
|
---|
419 | DebugStream.printStackTrace(exception);
|
---|
420 | }
|
---|
421 | catch (IOException exception) {
|
---|
422 | DebugStream.printStackTrace(exception);
|
---|
423 | } catch (Exception exception) { // e.g. exception decoding gsdlsourcefilename
|
---|
424 | DebugStream.printStackTrace(exception);
|
---|
425 | }
|
---|
426 | }
|
---|
427 |
|
---|
428 | protected String decodeSourceFilename(String relative_sourcefile_path,
|
---|
429 | String encodingMethod, boolean is_unix_path)
|
---|
430 | throws Exception
|
---|
431 | {
|
---|
432 |
|
---|
433 | ///System.err.println("*** relative_sourcefile_path: " + relative_sourcefile_path);
|
---|
434 |
|
---|
435 | // First get the file extension. Both in Base64 and URL encoded strings,
|
---|
436 | // the full-stop character (.) doesn't get encoded.
|
---|
437 | // That means getting the file extension is straightforward.
|
---|
438 |
|
---|
439 | // Valid base64: "The 64 characters (hence the name Base64) are 10 digits,
|
---|
440 | // 26 lowercase characters, 26 uppercase characters as well as the
|
---|
441 | // Plus sign (+) and the Forward Slash (/).
|
---|
442 | int fullstop = relative_sourcefile_path.indexOf(".");
|
---|
443 | String file_ext = "";
|
---|
444 | if(fullstop != -1) {
|
---|
445 | file_ext = relative_sourcefile_path.substring(fullstop);
|
---|
446 | relative_sourcefile_path = relative_sourcefile_path.substring(0, fullstop);
|
---|
447 | }
|
---|
448 |
|
---|
449 | String[] importFilePathParts = DocXMLFile.getFilePathParts(relative_sourcefile_path, is_unix_path);
|
---|
450 |
|
---|
451 | String decoded_gsdlsourcefilename = "";
|
---|
452 |
|
---|
453 | String separator = is_unix_path ? "/" : "\\";
|
---|
454 | for(int i = 0; i < importFilePathParts.length; i++) {
|
---|
455 | String decoded_filePathPart = "";
|
---|
456 | if(encodingMethod.equals(FILE_RENAME_METHOD_URL)) {
|
---|
457 | // URL decode each part of gsdlsourcefilename.
|
---|
458 | // Need to set the decoder to use the default system encoding
|
---|
459 | // This is stored in the System's file.encoding property.
|
---|
460 | decoded_filePathPart = URLDecoder.decode(importFilePathParts[i], System.getProperty("file.encoding"));
|
---|
461 | }
|
---|
462 | else{ // if(encodingMethod.equals(FILE_RENAME_METHOD_BASE64)) {
|
---|
463 | // Decoding with org.greenstone.gatherer.feedback.Base64 didn't work
|
---|
464 | //byte[] bytes = org.greenstone.gatherer.feedback.Base64.decode(importFilePathParts[i]);
|
---|
465 | // Using org.apache.commons.codec.binary.Base64 instead
|
---|
466 | // https://commons.apache.org/proper/commons-codec/archives/1.7/apidocs/org/apache/commons/codec/binary/Base64.html
|
---|
467 | // General info: https://stackoverflow.com/questions/43089541/difference-between-basic-and-url-base64-encoding-in-java-8
|
---|
468 | byte[] bytes = Base64.decodeBase64(importFilePathParts[i].getBytes());
|
---|
469 | ///System.err.println("Got base64 string: " + importFilePathParts[i]);
|
---|
470 | ///System.err.println("Decoded from base64 to bytes: " + new String(bytes, System.getProperty("file.encoding")));
|
---|
471 | // Using system file.encoding to interpret the resulting bytestring as a String,
|
---|
472 | // just as we always did with URL decoding method
|
---|
473 | decoded_filePathPart = (bytes == null) ? importFilePathParts[i] : new String(bytes, System.getProperty("file.encoding"));
|
---|
474 | }
|
---|
475 |
|
---|
476 | if(i == 0) {
|
---|
477 | decoded_gsdlsourcefilename = decoded_filePathPart;
|
---|
478 | } else {
|
---|
479 | decoded_gsdlsourcefilename = decoded_gsdlsourcefilename + separator + decoded_filePathPart;
|
---|
480 | }
|
---|
481 | ///System.err.println("Built up: " + decoded_gsdlsourcefilename);
|
---|
482 | }
|
---|
483 |
|
---|
484 | // add the file extension back in
|
---|
485 | decoded_gsdlsourcefilename += file_ext;
|
---|
486 |
|
---|
487 | ///System.err.println("@@@@ decoded_gsdlsourcefilename: " + Utility.debugUnicodeString(decoded_gsdlsourcefilename));
|
---|
488 |
|
---|
489 | return decoded_gsdlsourcefilename;
|
---|
490 | }
|
---|
491 |
|
---|
492 | /**
|
---|
493 | * Given a filepath, returns the parts between each file separator as an array.
|
---|
494 | * For example, "/Users/me/pinky.txt" should return {"Users", "me", "pinky.txt"};
|
---|
495 | */
|
---|
496 | private static String[] getFilePathParts(String filepath, boolean is_unix_path) {
|
---|
497 | StringTokenizer tok;
|
---|
498 | if(is_unix_path) {
|
---|
499 | tok = new StringTokenizer(filepath, "/");
|
---|
500 | } else {
|
---|
501 | tok = new StringTokenizer(filepath, "\\");
|
---|
502 | }
|
---|
503 | String[] parts;
|
---|
504 | int count = tok.countTokens();
|
---|
505 | if(count <= 0) {
|
---|
506 | parts = new String[]{filepath};
|
---|
507 | } else {
|
---|
508 | int i = 0;
|
---|
509 | parts = new String[count];
|
---|
510 | while(tok.hasMoreTokens()) {
|
---|
511 | parts[i] = tok.nextToken();
|
---|
512 | //System.err.println("Next part: " + parts[i]);
|
---|
513 | i++;
|
---|
514 | }
|
---|
515 | }
|
---|
516 | return parts;
|
---|
517 | }
|
---|
518 |
|
---|
519 | /*
|
---|
520 | public ArrayList getMetadataExtractedFromFile(File file)
|
---|
521 | {
|
---|
522 | // Build up a list of metadata extracted from this file
|
---|
523 | ArrayList metadata_values = new ArrayList();
|
---|
524 |
|
---|
525 | String file_relative_path = file.getAbsolutePath();
|
---|
526 | int import_index = file_relative_path.indexOf("import");
|
---|
527 | if (import_index != -1) {
|
---|
528 | file_relative_path = file_relative_path.substring(import_index + "import".length() + 1);
|
---|
529 | }
|
---|
530 |
|
---|
531 | // Check whether this doc.xml file contains extracted metadata for the specified file
|
---|
532 | ArrayList description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
|
---|
533 | if (description_elements_list == null) {
|
---|
534 | // ...it doesn't
|
---|
535 | return metadata_values;
|
---|
536 | }
|
---|
537 |
|
---|
538 | MetadataSet extracted_metadata_set = MetadataSetManager.getMetadataSet(MetadataSetManager.EXTRACTED_METADATA_NAMESPACE);
|
---|
539 |
|
---|
540 | // Parse the doc.xml file
|
---|
541 | DebugStream.println("Applicable doc.xml file: " + this);
|
---|
542 | try {
|
---|
543 | BufferedReader buffered_reader = new BufferedReader(new FileReader(this));
|
---|
544 |
|
---|
545 | int description_element_num = 0;
|
---|
546 | int next_description_element_start = ((Integer) description_elements_list.get(description_element_num)).intValue();
|
---|
547 | boolean in_relevant_description_element = false;
|
---|
548 |
|
---|
549 | String line = null;
|
---|
550 | for (int line_num = 0; (line = buffered_reader.readLine()) != null; line_num++) {
|
---|
551 | // Check if this line contains the start of a relevant Description element
|
---|
552 | if (line_num == next_description_element_start) {
|
---|
553 | in_relevant_description_element = true;
|
---|
554 | continue;
|
---|
555 | }
|
---|
556 |
|
---|
557 | // If we're not in a relevant Description element we don't care about anything
|
---|
558 | if (in_relevant_description_element == false) {
|
---|
559 | continue;
|
---|
560 | }
|
---|
561 |
|
---|
562 | // Check if this line contains the end of the relevant Description element
|
---|
563 | if (line.indexOf("</Description>") != -1) {
|
---|
564 | description_element_num++;
|
---|
565 | if (description_element_num == description_elements_list.size()) {
|
---|
566 | break;
|
---|
567 | }
|
---|
568 |
|
---|
569 | next_description_element_start = ((Integer) description_elements_list.get(description_element_num)).intValue();
|
---|
570 | in_relevant_description_element = false;
|
---|
571 | continue;
|
---|
572 | }
|
---|
573 |
|
---|
574 | // If this line doesn't contain a complete Metadata element, we're not interested
|
---|
575 | if (line.indexOf("<Metadata ") == -1 || line.indexOf("</Metadata>") == -1) {
|
---|
576 | continue;
|
---|
577 | }
|
---|
578 |
|
---|
579 | // Extract the metadata element name
|
---|
580 | int name_index = line.indexOf(" name=\"") + " name=\"".length();
|
---|
581 | String metadata_element_name_full = line.substring(name_index, line.indexOf("\"", name_index));
|
---|
582 |
|
---|
583 | // If the metadata has a namespace it isn't extracted metadata, so we're not interested
|
---|
584 | String metadata_set_namespace = MetadataTools.getMetadataSetNamespace(metadata_element_name_full);
|
---|
585 | if (!metadata_set_namespace.equals("")) {
|
---|
586 | continue;
|
---|
587 | }
|
---|
588 |
|
---|
589 | // Extracted metadata!
|
---|
590 | String metadata_element_name = metadata_element_name_full;
|
---|
591 |
|
---|
592 | // We completely ignore bibliographic data
|
---|
593 | if (metadata_element_name.equals("SourceSegment")) {
|
---|
594 | buffered_reader.close();
|
---|
595 | return new ArrayList();
|
---|
596 | }
|
---|
597 |
|
---|
598 | // Ignore metadata starting with gsdl (used to be lowercase metadata and /metadata)
|
---|
599 | if (metadata_element_name.startsWith("gsdl")) {
|
---|
600 | continue;
|
---|
601 | }
|
---|
602 |
|
---|
603 | MetadataElement metadata_element = extracted_metadata_set.getMetadataElementWithName(metadata_element_name);
|
---|
604 |
|
---|
605 | // Value trees are not stored for extracted metadata, so create a new value tree node now
|
---|
606 | int value_index = line.indexOf(">", name_index) + ">".length();
|
---|
607 | String metadata_element_value = line.substring(value_index, line.lastIndexOf("</Metadata>"));
|
---|
608 |
|
---|
609 | metadata_element.addMetadataValue(metadata_element_value);
|
---|
610 | MetadataValueTreeNode metadata_value_tree_node = metadata_element.getMetadataValueTreeNode(metadata_element_value);
|
---|
611 |
|
---|
612 | // Add the new metadata value to the list
|
---|
613 | MetadataValue metadata_value = new MetadataValue(metadata_element, metadata_value_tree_node);
|
---|
614 | metadata_values.add(metadata_value);
|
---|
615 | }
|
---|
616 |
|
---|
617 | buffered_reader.close();
|
---|
618 | }
|
---|
619 | catch (FileNotFoundException exception) {
|
---|
620 | DebugStream.printStackTrace(exception);
|
---|
621 | }
|
---|
622 | catch (IOException exception) {
|
---|
623 | DebugStream.printStackTrace(exception);
|
---|
624 | }
|
---|
625 |
|
---|
626 | return metadata_values;
|
---|
627 | }
|
---|
628 |
|
---|
629 | */
|
---|
630 |
|
---|
631 | /**
|
---|
632 | * Every doc.xml file must be skimmed when a collection is opened, for two reasons:
|
---|
633 | * - To build a mapping from source file to its corresponding doc.xml file
|
---|
634 | * - To get a complete list of all extracted metadata elements
|
---|
635 | */
|
---|
636 | /*
|
---|
637 | public void skimFile()
|
---|
638 | {
|
---|
639 | MetadataSet extracted_metadata_set = MetadataSetManager.getMetadataSet(MetadataSetManager.EXTRACTED_METADATA_NAMESPACE);
|
---|
640 |
|
---|
641 | // Skim the doc.xml file as quickly as possible (don't parse as XML), looking at the Metadata elements
|
---|
642 | DebugStream.println("Skimming " + this + "...");
|
---|
643 | try {
|
---|
644 | BufferedReader buffered_reader = new BufferedReader(new FileReader(this));
|
---|
645 | int description_element_start = -1;
|
---|
646 |
|
---|
647 | String line = null;
|
---|
648 | for (int line_num = 0; (line = buffered_reader.readLine()) != null; line_num++) {
|
---|
649 | // This line contains the start of a Description element
|
---|
650 | if (line.indexOf("<Description>") != -1) {
|
---|
651 | if (description_element_start != -1) {
|
---|
652 | System.err.println("Parse error: previous Description element unfinished!");
|
---|
653 | }
|
---|
654 | description_element_start = line_num;
|
---|
655 | continue;
|
---|
656 | }
|
---|
657 |
|
---|
658 | // This line contains the end of a Description element
|
---|
659 | if (line.indexOf("</Description>") != -1) {
|
---|
660 | if (description_element_start == -1) {
|
---|
661 | System.err.println("Parse error: Description element unstarted!");
|
---|
662 | }
|
---|
663 | description_element_start = -1;
|
---|
664 | continue;
|
---|
665 | }
|
---|
666 |
|
---|
667 | // If we're not in a Description element there shouldn't be any Metadata elements
|
---|
668 | if (description_element_start == -1) {
|
---|
669 | continue;
|
---|
670 | }
|
---|
671 |
|
---|
672 | // This line doesn't contain a Metadata element, so we're not interested
|
---|
673 | if (line.indexOf("<Metadata ") == -1) {
|
---|
674 | DebugStream.println("Warning: Description element line doesn't contain Metadata element.");
|
---|
675 | continue;
|
---|
676 | }
|
---|
677 |
|
---|
678 | // Extract the metadata element name
|
---|
679 | int name_index = line.indexOf(" name=\"") + " name=\"".length();
|
---|
680 | String metadata_element_name_full = line.substring(name_index, line.indexOf("\"", name_index));
|
---|
681 |
|
---|
682 | // If the metadata has a namespace it isn't extracted metadata, so we're not interested
|
---|
683 | String metadata_set_namespace = MetadataTools.getMetadataSetNamespace(metadata_element_name_full);
|
---|
684 | if (!metadata_set_namespace.equals("")) {
|
---|
685 | continue;
|
---|
686 | }
|
---|
687 |
|
---|
688 | // Extracted metadata!
|
---|
689 | String metadata_element_name = metadata_element_name_full;
|
---|
690 |
|
---|
691 | // Note which file this doc.xml is for
|
---|
692 | if (metadata_element_name.equals("gsdlsourcefilename")) {
|
---|
693 | // Extract the gsdlsourcefilename element value
|
---|
694 | int value_index = line.indexOf(">", name_index) + ">".length();
|
---|
695 | String gsdlsourcefilename_value = line.substring(value_index, line.indexOf("<", value_index));
|
---|
696 |
|
---|
697 | // We're only interested in the path relative to the import folder
|
---|
698 | int import_index = gsdlsourcefilename_value.indexOf("import");
|
---|
699 | if (import_index != -1) {
|
---|
700 | gsdlsourcefilename_value = gsdlsourcefilename_value.substring(import_index + "import".length());
|
---|
701 |
|
---|
702 | boolean is_unix_path = gsdlsourcefilename_value.startsWith("/");
|
---|
703 | gsdlsourcefilename_value = gsdlsourcefilename_value.substring(1);
|
---|
704 |
|
---|
705 | // URL decode gsdlsourcefilename. Need to set the decoder to use the default system encoding
|
---|
706 | // This is stored in the System's file.encoding property.
|
---|
707 | gsdlsourcefilename_value = URLDecoder.decode(gsdlsourcefilename_value, System.getProperty("file.encoding"));
|
---|
708 |
|
---|
709 | // Make sure the path matches the OS that is running
|
---|
710 | if (is_unix_path && Utility.isWindows()) {
|
---|
711 | // Convert path from Unix to Windows
|
---|
712 | gsdlsourcefilename_value = gsdlsourcefilename_value.replaceAll("\\/", "\\\\");
|
---|
713 | }
|
---|
714 | else if (!is_unix_path && !Utility.isWindows()) {
|
---|
715 | // Convert path from Windows to Unix
|
---|
716 | gsdlsourcefilename_value = gsdlsourcefilename_value.replaceAll("\\\\", "/");
|
---|
717 | }
|
---|
718 |
|
---|
719 | // Remember this for quick access later
|
---|
720 | if (source_file_name_to_description_elements_mapping.get(gsdlsourcefilename_value) == null) {
|
---|
721 | source_file_name_to_description_elements_mapping.put(gsdlsourcefilename_value, new ArrayList());
|
---|
722 | }
|
---|
723 |
|
---|
724 | ((ArrayList) source_file_name_to_description_elements_mapping.get(gsdlsourcefilename_value)).add(new Integer(description_element_start));
|
---|
725 | }
|
---|
726 |
|
---|
727 | // Warn about an odd gsdlsourcefilename, except if it is the Greenstone "tmp" directory
|
---|
728 | // This is true when the source files come from a zip file processed by ZIPPlug, for example
|
---|
729 | else if (gsdlsourcefilename_value.indexOf("tmp") == -1) {
|
---|
730 | // We don't really know what is going on...
|
---|
731 | System.err.println("Warning: Could not understand gsdlsourcefilename " + gsdlsourcefilename_value);
|
---|
732 | }
|
---|
733 | }
|
---|
734 |
|
---|
735 | // Ignore metadata starting with gsdl (used to be lowercase metadata and /metadata)
|
---|
736 | if (metadata_element_name.startsWith("gsdl")) {
|
---|
737 | continue;
|
---|
738 | }
|
---|
739 |
|
---|
740 | MetadataElement metadata_element = extracted_metadata_set.getMetadataElementWithName(metadata_element_name);
|
---|
741 | if (metadata_element == null) {
|
---|
742 | // This element isn't defined in ex.mds, so create it for this session
|
---|
743 | DebugStream.println("Extracted metadata element not defined: " + metadata_element_name);
|
---|
744 | extracted_metadata_set.addMetadataElementForThisSession(metadata_element_name);
|
---|
745 | }
|
---|
746 | }
|
---|
747 |
|
---|
748 | buffered_reader.close();
|
---|
749 | }
|
---|
750 | catch (FileNotFoundException exception) {
|
---|
751 | DebugStream.printStackTrace(exception);
|
---|
752 | }
|
---|
753 | catch (IOException exception) {
|
---|
754 | DebugStream.printStackTrace(exception);
|
---|
755 | }
|
---|
756 | }
|
---|
757 | */
|
---|
758 |
|
---|
759 | }
|
---|