1 | package org.greenstone.gsdl3.gs3build.doctypes;
|
---|
2 |
|
---|
3 | import java.io.File;
|
---|
4 | import java.net.URL;
|
---|
5 | import java.net.MalformedURLException;
|
---|
6 |
|
---|
7 | import java.util.Iterator;
|
---|
8 | import java.util.List;
|
---|
9 | import java.util.Map;
|
---|
10 | import java.util.ArrayList;
|
---|
11 |
|
---|
12 | import org.greenstone.gsdl3.gs3build.metadata.*;
|
---|
13 | import org.greenstone.gsdl3.gs3build.util.*;
|
---|
14 | import org.greenstone.gsdl3.gs3build.xpointer.XPointer;
|
---|
15 | import org.greenstone.gsdl3.util.XMLConverter;
|
---|
16 |
|
---|
17 | import org.w3c.dom.*;
|
---|
18 |
|
---|
19 | public class HTMLDocument extends AbstractDocument
|
---|
20 | {
|
---|
21 | public static final String HTML_DOCUMENT_TYPE = "HTML";
|
---|
22 | Document domDocument;
|
---|
23 |
|
---|
24 | public HTMLDocument(DocumentID id)
|
---|
25 | { super(id);
|
---|
26 | this.domDocument = null;
|
---|
27 | }
|
---|
28 |
|
---|
29 | /**
|
---|
30 | * Create the HTMLDocument from a given URL - the URL may in fact be a reference
|
---|
31 | * to a local file.
|
---|
32 | *
|
---|
33 | * @param <code>URL</code> The location from which to load the file
|
---|
34 | */
|
---|
35 | public HTMLDocument(URL url)
|
---|
36 | { super(url);
|
---|
37 |
|
---|
38 | this.loadDocument(url);
|
---|
39 |
|
---|
40 | this._extractDocumentFiles();
|
---|
41 | this._extractDocumentMetadata();
|
---|
42 | // extract the files before resolving the images
|
---|
43 | resolveImages();
|
---|
44 | HTMLDocumentTools docTools = new HTMLDocumentTools(this.domDocument);
|
---|
45 | docTools.setMetsDocument(this);
|
---|
46 | docTools.setUrl(this.fileSet.getFile(0).getLocation());
|
---|
47 | docTools.findSections();
|
---|
48 | METSStructure sectionStruct = docTools.getStructure();
|
---|
49 |
|
---|
50 | if (sectionStruct.size() > 0) {
|
---|
51 | METSStructureSet structureSet = this.getDocumentStructure();
|
---|
52 | structureSet.addStructure(sectionStruct);
|
---|
53 | }
|
---|
54 | }
|
---|
55 |
|
---|
56 | private void loadDocument(URL url)
|
---|
57 | {
|
---|
58 | // HTMLDoc htmlDoc;
|
---|
59 |
|
---|
60 | HTMLTidy tidyDoc;
|
---|
61 | if (url.getProtocol().equals("file"))
|
---|
62 | { tidyDoc = new HTMLTidy(new File(url.getPath()));
|
---|
63 | }
|
---|
64 | else
|
---|
65 | { //htmlDoc = new HTMLDoc(url);
|
---|
66 | tidyDoc = new HTMLTidy(url);
|
---|
67 | }
|
---|
68 |
|
---|
69 | // System.out.println(Runtime.getRuntime().freeMemory() + " " + Runtime.getRuntime().totalMemory());
|
---|
70 |
|
---|
71 | this.domDocument = tidyDoc.getDocument();
|
---|
72 | Runtime.getRuntime().gc();
|
---|
73 | // System.out.println(Runtime.getRuntime().freeMemory() + " " + Runtime.getRuntime().totalMemory());
|
---|
74 | }
|
---|
75 |
|
---|
76 | private void _extractDocumentMetadata()
|
---|
77 | {
|
---|
78 | NodeList metadata = this.domDocument.getElementsByTagName("META");
|
---|
79 | for (int n = 0; n < metadata.getLength(); n ++) {
|
---|
80 | Node node = metadata.item(n);
|
---|
81 | Element element = (Element) node;
|
---|
82 |
|
---|
83 | String name = element.getAttribute("name");
|
---|
84 | if (name == null || name.length() == 0) {
|
---|
85 | continue;
|
---|
86 | }
|
---|
87 |
|
---|
88 | String value = element.getAttribute("content");
|
---|
89 | if (value == null || value.length() == 0) {
|
---|
90 | value = name;
|
---|
91 | }
|
---|
92 |
|
---|
93 | this.addDocumentMetadata(new MetadataLabel(name), value);
|
---|
94 | }
|
---|
95 |
|
---|
96 | NodeList titles = this.domDocument.getElementsByTagName("TITLE");
|
---|
97 | StringBuffer title = new StringBuffer();
|
---|
98 | for (int n = 0; n < titles.getLength(); n ++) {
|
---|
99 | Node node = titles.item(n);
|
---|
100 | Element element = (Element) node;
|
---|
101 |
|
---|
102 | NodeList childNodes = node.getChildNodes();
|
---|
103 | for (int c = 0; c < childNodes.getLength(); c ++) {
|
---|
104 | Node child = childNodes.item(c);
|
---|
105 | if (child.getNodeType() == org.w3c.dom.Node.TEXT_NODE) {
|
---|
106 | title.append(child.getNodeValue());
|
---|
107 | }
|
---|
108 | }
|
---|
109 | }
|
---|
110 |
|
---|
111 | if (title.length() > 0)
|
---|
112 | { this.addDocumentMetadata(new MetadataLabel("Title"), title.toString());
|
---|
113 | }
|
---|
114 |
|
---|
115 | /* Old HTMLDoc based parsing...
|
---|
116 | HTMLBlock codedContent = htmlDoc.getCodedContent();
|
---|
117 | boolean inTitle = false;
|
---|
118 | title = new StringBuffer();
|
---|
119 |
|
---|
120 | for (int e = 0; e < codedContent.size(); e ++)
|
---|
121 | { if (codedContent.elementAt(e) instanceof HTMLTag)
|
---|
122 | { HTMLTag tag = (HTMLTag) codedContent.elementAt(e);
|
---|
123 |
|
---|
124 | if (tag.tagName().equals("meta"))
|
---|
125 | { // check that the name of the metadata item exists
|
---|
126 | String name = tag.idValue("name");
|
---|
127 | if (name == null || name.length() == 0) {
|
---|
128 | continue;
|
---|
129 | }
|
---|
130 |
|
---|
131 | // get the value, if it exists
|
---|
132 | String value = tag.idValue("content");
|
---|
133 |
|
---|
134 | // if value does not exist, default it to being the same
|
---|
135 | // as the name.
|
---|
136 | if (value == null || value.length() > 0) {
|
---|
137 | value = name;
|
---|
138 | }
|
---|
139 | }
|
---|
140 | else
|
---|
141 | if (tag.tagName().equals("title"))
|
---|
142 | { inTitle = true;
|
---|
143 | }
|
---|
144 | else if (tag.tagName().equals("/title"))
|
---|
145 | { inTitle = false;
|
---|
146 | }
|
---|
147 | // cut off when real body content appears - not a perfect
|
---|
148 | // implementation, just cheap & cheerful
|
---|
149 | else if (tag.tagName().equals("/head"))
|
---|
150 | { break;
|
---|
151 | }
|
---|
152 | else if (tag.tagName().equals("body"))
|
---|
153 | { break;
|
---|
154 | }
|
---|
155 | }
|
---|
156 | else if (inTitle == true)
|
---|
157 | { title.append(codedContent.elementAt(e).toString());
|
---|
158 | }
|
---|
159 | }
|
---|
160 | */
|
---|
161 | }
|
---|
162 |
|
---|
163 | private void _extractDocumentFiles()
|
---|
164 | { URL homeUrl = this.fileSet.getFile(0).getLocation();
|
---|
165 |
|
---|
166 | NodeList metadata = this.domDocument.getElementsByTagName("img");
|
---|
167 | for (int n = 0; n < metadata.getLength(); n ++) {
|
---|
168 | Node node = metadata.item(n);
|
---|
169 | Element element = (Element) node;
|
---|
170 |
|
---|
171 | String location = element.getAttribute("src");
|
---|
172 | if (location == null || location.length() == 0) {
|
---|
173 | System.out.println("No name");
|
---|
174 | continue;
|
---|
175 | }
|
---|
176 | try
|
---|
177 | { // make the url for the image, and then add it to the document list of
|
---|
178 | //
|
---|
179 | URL imgUrl = new URL(homeUrl, location);
|
---|
180 | METSFile file = this.fileSet.addFile(imgUrl);
|
---|
181 | this.structureSet.getStructure("All").getDivision("All").addFileReference(file.getID());
|
---|
182 | }
|
---|
183 | catch (MalformedURLException ex)
|
---|
184 | { // TODO: report exception/failure to resolve...
|
---|
185 | }
|
---|
186 | }
|
---|
187 |
|
---|
188 |
|
---|
189 | /**
|
---|
190 | HTMLBlock codedContent = htmlDoc.getCodedContent();
|
---|
191 | for (int e = 0; e < codedContent.size(); e ++)
|
---|
192 | { if (codedContent.elementAt(e) instanceof HTMLTag)
|
---|
193 | { HTMLTag tag = (HTMLTag) codedContent.elementAt(e);
|
---|
194 |
|
---|
195 | if (tag.tagName().equals("img"))
|
---|
196 | { String location = tag.idValue("src");
|
---|
197 |
|
---|
198 | try
|
---|
199 | { // make the url for the image, and then add it to the document list of
|
---|
200 | //
|
---|
201 | URL imgUrl = new URL(homeUrl, location);
|
---|
202 | METSFile file = this.fileSet.addFile(imgUrl);
|
---|
203 | this.structureSet.getStructure("All").getDivision("All").addFileReference(file.getID());
|
---|
204 | }
|
---|
205 | catch (MalformedURLException ex)
|
---|
206 | { // TODO: report exception/failure to resolve...
|
---|
207 | }
|
---|
208 | }
|
---|
209 | }
|
---|
210 | }
|
---|
211 | */
|
---|
212 | }
|
---|
213 |
|
---|
214 | public String getDocumentType()
|
---|
215 | { return HTML_DOCUMENT_TYPE;
|
---|
216 | }
|
---|
217 |
|
---|
218 | // I think this is used for single section documents, while getSectionText
|
---|
219 | // is used for sectioned documents
|
---|
220 | // we will use the domDocument rather than reading it in again to another HTMLDoc.
|
---|
221 | public String getDocumentText()
|
---|
222 | {
|
---|
223 | XMLConverter converter = new XMLConverter();
|
---|
224 | return converter.getPrettyString(this.domDocument.getDocumentElement());
|
---|
225 | /* HTMLDoc htmlDoc;
|
---|
226 | URL url =(URL) this.fileSet.getFile(0).getLocation();
|
---|
227 |
|
---|
228 | this.getSectionText("1");
|
---|
229 |
|
---|
230 | if (url.getProtocol().equals("file"))
|
---|
231 | { htmlDoc = new HTMLDoc(url, url.getPath());
|
---|
232 | }
|
---|
233 | else
|
---|
234 | { htmlDoc = new HTMLDoc(url);
|
---|
235 | }
|
---|
236 | return htmlDoc.getContent();
|
---|
237 | */
|
---|
238 | }
|
---|
239 |
|
---|
240 | private void resolveImages() {
|
---|
241 |
|
---|
242 | // find the path of the url relative to the collection
|
---|
243 | URL full_path = this.fileSet.getFile(0).getLocation();
|
---|
244 |
|
---|
245 | String base_url;
|
---|
246 | if (full_path.getProtocol().equals("file")) {
|
---|
247 | base_url = full_path.getPath();
|
---|
248 | int import_pos = base_url.indexOf("import");
|
---|
249 | base_url = base_url.substring(import_pos);
|
---|
250 | base_url = "_httpcollection_/"+base_url;
|
---|
251 | } else {
|
---|
252 | base_url = full_path.toString();
|
---|
253 | }
|
---|
254 |
|
---|
255 | // need to take off the last part
|
---|
256 | base_url = base_url.substring(0, base_url.lastIndexOf("/")+1);
|
---|
257 |
|
---|
258 | NodeList metadata = this.domDocument.getElementsByTagName("img");
|
---|
259 | for (int n = 0; n < metadata.getLength(); n ++) {
|
---|
260 | Node node = metadata.item(n);
|
---|
261 | Element element = (Element) node;
|
---|
262 |
|
---|
263 | String location = element.getAttribute("src");
|
---|
264 | if (location != null && location.length() > 0 && isRelative(location)) {
|
---|
265 | // modify the source url
|
---|
266 | element.setAttribute("src", base_url+location);
|
---|
267 | }
|
---|
268 | }
|
---|
269 | }
|
---|
270 |
|
---|
271 | private boolean isRelative(String location) {
|
---|
272 |
|
---|
273 | if ( location.startsWith("http:") || location.startsWith("file:")) {
|
---|
274 | return false;
|
---|
275 | }
|
---|
276 | return true;
|
---|
277 | }
|
---|
278 |
|
---|
279 | public Document getDOMDocument()
|
---|
280 | {
|
---|
281 | if (this.domDocument == null) {
|
---|
282 | URL url =(URL) this.fileSet.getFile(0).getLocation();
|
---|
283 | this.loadDocument(url);
|
---|
284 | resolveImages();
|
---|
285 | }
|
---|
286 | return this.domDocument;
|
---|
287 | }
|
---|
288 |
|
---|
289 | private XPointer getXPointer(METSDivision division)
|
---|
290 | { String fileId = division.getDefaultFileReference();
|
---|
291 |
|
---|
292 | if (fileId == null) {
|
---|
293 | System.err.println("Unable to obtain file reference for section " + division.getLabel());
|
---|
294 | return null;
|
---|
295 | }
|
---|
296 |
|
---|
297 | METSFileGroup fileGroup = this.getDocumentFiles().getGroup(fileId);
|
---|
298 | if (fileGroup == null) {
|
---|
299 | System.err.println("Unable to obtain file reference for filegroup " + fileId);
|
---|
300 | return null;
|
---|
301 | }
|
---|
302 |
|
---|
303 | METSFile file = fileGroup.getFile(0);
|
---|
304 | if (file == null) {
|
---|
305 | System.err.println("Unable to obtain any files within filegroup " + fileId);
|
---|
306 | return null;
|
---|
307 | }
|
---|
308 |
|
---|
309 | URL url = file.getLocation();
|
---|
310 | XPointer xpointer = XPointer.processXPointer(this.domDocument, url);
|
---|
311 |
|
---|
312 | return xpointer;
|
---|
313 | }
|
---|
314 |
|
---|
315 | private XPointer getXPointer(String sectionId)
|
---|
316 | { if (this.domDocument == null) {
|
---|
317 | URL url =(URL) this.fileSet.getFile(0).getLocation();
|
---|
318 | this.loadDocument(url);
|
---|
319 | resolveImages();
|
---|
320 | }
|
---|
321 |
|
---|
322 | METSDivision division =
|
---|
323 | this.getDocumentStructure().getDivision(METSStructureSet.GSDL3_SECTION_STRUCTURE,
|
---|
324 | sectionId);
|
---|
325 | if (division == null) {
|
---|
326 | // System.err.println("Unable to locate section " + sectionId);
|
---|
327 | return null;
|
---|
328 | }
|
---|
329 |
|
---|
330 | return this.getXPointer(division);
|
---|
331 | }
|
---|
332 |
|
---|
333 | public Node getSectionStartNode(METSDivision division)
|
---|
334 | { XPointer xpointer = this.getXPointer(division);
|
---|
335 | return xpointer.getStartNode();
|
---|
336 | }
|
---|
337 |
|
---|
338 | public Node getSectionStartNode(String sectionId)
|
---|
339 | { XPointer xpointer = this.getXPointer(sectionId);
|
---|
340 |
|
---|
341 | return xpointer.getStartNode();
|
---|
342 | }
|
---|
343 |
|
---|
344 | public String getSectionText(String sectionId)
|
---|
345 | { XPointer xpointer = this.getXPointer(sectionId);
|
---|
346 |
|
---|
347 | if (xpointer == null) {
|
---|
348 | return "";
|
---|
349 | }
|
---|
350 |
|
---|
351 | // get the XML content of the xpointer...
|
---|
352 | return xpointer.toString();
|
---|
353 | }
|
---|
354 | }
|
---|