1 | package org.nzdl.gsdl.GsdlCollageApplet;
|
---|
2 |
|
---|
3 | import java.awt.*;
|
---|
4 | import java.io.*;
|
---|
5 | import java.net.*;
|
---|
6 | import java.util.*;
|
---|
7 |
|
---|
8 | /**
|
---|
9 | * @author Katrina Edgar
|
---|
10 | * @author David Bainbridge
|
---|
11 | *
|
---|
12 | * Controls retrieval of images from the specified starting url. Follows appropriate
|
---|
13 | * links from this starting point, traversing in a tree-like state through several other
|
---|
14 | * pages. Filters images and links based on specified parameters. Also controls the quantity
|
---|
15 | * of downloading that occurs by restricting the number of downloaded images that are yet to
|
---|
16 | * be displayed to 10, and the total number of downloads allowed is also restricted by
|
---|
17 | * the applet application (to prevent downloading occur infinitely). */
|
---|
18 | public class DownloadUrls extends Thread {
|
---|
19 |
|
---|
20 | /** Refers to applet */
|
---|
21 | GsdlCollageApplet app_ = null;
|
---|
22 | /** Refers to download thread */
|
---|
23 | DownloadImages download_images_ = null;
|
---|
24 |
|
---|
25 | /** The address from which the application should start looking for images */
|
---|
26 | String starting_url_ = null;
|
---|
27 | /** Restricts links followed from the starting url to links that contain this string */
|
---|
28 | String href_musthave_ = null;
|
---|
29 | /** Restricts links followed from the starting url to links that do not contain this string.
|
---|
30 | * Also prevents image names from containing this string */
|
---|
31 | String image_mustnothave_ = null;
|
---|
32 | /** Ignore images whose names begin with this string */
|
---|
33 | String image_ignore_ = null;
|
---|
34 | /** Restricts the types of images included in the collage, for example jpg, gif, etc. */
|
---|
35 | String image_type_ = null;
|
---|
36 |
|
---|
37 | /** A static delay used when attempting to download more images into a full downloading buffer */
|
---|
38 | final int delay_ = 1000;
|
---|
39 | /** The maximum number of images to have downloaded and not yet displayed */
|
---|
40 | final int buffer_size_ = 10;
|
---|
41 |
|
---|
42 | /** Used in cases where the image maps to a url outside of it's original location.
|
---|
43 | * When used with Greenstone the collage images will refer to documents in the collections
|
---|
44 | * from which the images are sourced. When used individually, the images may be saved into
|
---|
45 | * a user directory and the pages they reference may be external hyperlinks. */
|
---|
46 | Hashtable external_links_ = null;
|
---|
47 |
|
---|
48 | /** Records all urls which have already been examined */
|
---|
49 | Hashtable visited_url_ = null;
|
---|
50 | /** Determines whether there are still pages to examine and images to download */
|
---|
51 | boolean thread_running_ = true;
|
---|
52 |
|
---|
53 | /** Constructor to initialise a download thread from which images are found,
|
---|
54 | * saves parameters into local variables for use within the class.
|
---|
55 | *
|
---|
56 | * @param app reference to the applet
|
---|
57 | * @param download_images class which stores the images retrieved in triplets
|
---|
58 | * @param starting_url the url from which the search for images should begin
|
---|
59 | * @param href_musthave restricts links to only those containing this string
|
---|
60 | * @param image_mustnothave restricts links and image names to only those that don't contain this string
|
---|
61 | * @param image_ignore restricts the beginning of image names
|
---|
62 | * @param image_type restricts the type of images included in the collage to those named */
|
---|
63 | public DownloadUrls(GsdlCollageApplet app,
|
---|
64 | DownloadImages download_images, String starting_url,
|
---|
65 | String href_musthave, String image_mustnothave,
|
---|
66 | String image_ignore, String image_type)
|
---|
67 | {
|
---|
68 | super("DownloadUrls");
|
---|
69 | app_ = app;
|
---|
70 | download_images_ = download_images;
|
---|
71 |
|
---|
72 | starting_url_ = starting_url;
|
---|
73 | href_musthave_ = href_musthave;
|
---|
74 | image_mustnothave_ = image_mustnothave;
|
---|
75 | image_ignore_ = image_ignore;
|
---|
76 | image_type_ = image_type;
|
---|
77 | }
|
---|
78 |
|
---|
79 | /** Determines whether or not a url has already been examined
|
---|
80 | *
|
---|
81 | * @param url_string the url to check
|
---|
82 | * @return true if the url has been visited, false if not */
|
---|
83 | public boolean already_visited(String url_string)
|
---|
84 | {
|
---|
85 | int hash_pos = url_string.indexOf("#");
|
---|
86 | if (hash_pos>0)
|
---|
87 | {
|
---|
88 | // strip off #anchor reference
|
---|
89 | url_string = url_string.substring(0,hash_pos);
|
---|
90 | }
|
---|
91 |
|
---|
92 | // if the url has been visited before, return true
|
---|
93 | if (visited_url_.containsKey(url_string))
|
---|
94 | {
|
---|
95 | if (app_.verbosity() > 3)
|
---|
96 | {
|
---|
97 | System.err.println("Visited " + url_string + " Before!");
|
---|
98 | }
|
---|
99 | return true;
|
---|
100 | }
|
---|
101 |
|
---|
102 | visited_url_.put(url_string,"visited");
|
---|
103 |
|
---|
104 | return false;
|
---|
105 | }
|
---|
106 |
|
---|
107 | /** Restricts the type of images that can be included in the collage
|
---|
108 | *
|
---|
109 | * @param url_string the url to check
|
---|
110 | * @return true if the image is of a specified type, false if not */
|
---|
111 | public boolean image_file_extension(String url_string)
|
---|
112 | {
|
---|
113 | // lower case comparisons
|
---|
114 | String url_lstring = url_string.toLowerCase();
|
---|
115 |
|
---|
116 | if (image_type_ == null)
|
---|
117 | return true;
|
---|
118 |
|
---|
119 | String tmp = image_type_;
|
---|
120 | String original_image_type_ = image_type_;
|
---|
121 |
|
---|
122 | while (image_type_ != null && image_type_.indexOf("%") >= 0) {
|
---|
123 |
|
---|
124 | tmp = image_type_.substring(0, image_type_.indexOf("%"));
|
---|
125 | if (image_type_.length() > image_type_.indexOf("%") + 1)
|
---|
126 | image_type_ = image_type_.substring(image_type_.indexOf("%") + 1, image_type_.length());
|
---|
127 | else
|
---|
128 | image_type_ = null;
|
---|
129 |
|
---|
130 | if (url_lstring.endsWith(tmp)) {
|
---|
131 | image_type_ = original_image_type_;
|
---|
132 | return true;
|
---|
133 | }
|
---|
134 | }
|
---|
135 |
|
---|
136 | if (image_type_ != null && url_lstring.endsWith(image_type_)) {
|
---|
137 | image_type_ = original_image_type_;
|
---|
138 | return true;
|
---|
139 | }
|
---|
140 |
|
---|
141 | image_type_ = original_image_type_;
|
---|
142 | return false;
|
---|
143 | }
|
---|
144 |
|
---|
145 | /** Restricts images to only those that satisfy several specified conditions
|
---|
146 | * regarding the content of the image name and url.
|
---|
147 | *
|
---|
148 | * @param url_string the url to check
|
---|
149 | * @return true if the image is satisfactory, false if not */
|
---|
150 | public boolean filter_image(String url_string)
|
---|
151 | {
|
---|
152 | if (image_ignore_==null || !url_string.startsWith(image_ignore_))
|
---|
153 | {
|
---|
154 | if (!already_visited(url_string))
|
---|
155 | {
|
---|
156 | if (image_mustnothave_ != null) {
|
---|
157 |
|
---|
158 | String tmp = image_mustnothave_;
|
---|
159 | String original_image_mustnothave_ = image_mustnothave_;
|
---|
160 |
|
---|
161 | while (image_mustnothave_ != null && image_mustnothave_.indexOf("%") >= 0) {
|
---|
162 |
|
---|
163 | tmp = image_mustnothave_.substring(0, image_mustnothave_.indexOf("%"));
|
---|
164 | if (image_mustnothave_.length() > image_mustnothave_.indexOf("%") + 1)
|
---|
165 | image_mustnothave_ = image_mustnothave_.substring(image_mustnothave_.indexOf("%") + 1,
|
---|
166 | image_mustnothave_.length());
|
---|
167 | else
|
---|
168 | image_mustnothave_ = null;
|
---|
169 |
|
---|
170 | if (url_string.indexOf(tmp) >= 0) {
|
---|
171 | image_mustnothave_ = original_image_mustnothave_;
|
---|
172 | return false;
|
---|
173 | }
|
---|
174 | }
|
---|
175 |
|
---|
176 | if (image_mustnothave_ != null && url_string.indexOf(image_mustnothave_) >= 0) {
|
---|
177 | image_mustnothave_ = original_image_mustnothave_;
|
---|
178 | return false;
|
---|
179 | }
|
---|
180 |
|
---|
181 | if (app_.verbosity() > 2) {
|
---|
182 | System.err.println("src url = "+ url_string);
|
---|
183 | }
|
---|
184 |
|
---|
185 | image_mustnothave_ = original_image_mustnothave_;
|
---|
186 | return true;
|
---|
187 | }
|
---|
188 | }
|
---|
189 | }
|
---|
190 |
|
---|
191 | return false;
|
---|
192 | }
|
---|
193 |
|
---|
194 | /** Restricts links to only those that satisfy several specified conditions
|
---|
195 | * regarding the address of the link.
|
---|
196 | *
|
---|
197 | * @param url_string the url to check
|
---|
198 | * @param new_url_string the url from which this link was found
|
---|
199 | * @param depth the number of links followed on this path
|
---|
200 | * @return true if the image is satisfactory, false if not */
|
---|
201 | public boolean filter_href(String url_string, String new_url_string, int depth)
|
---|
202 | {
|
---|
203 | boolean has_href = false;
|
---|
204 | String tmp = href_musthave_;
|
---|
205 | String original_href_musthave_ = href_musthave_;
|
---|
206 |
|
---|
207 | // checks the it does contain this content
|
---|
208 | if (href_musthave_ != null) {
|
---|
209 |
|
---|
210 | while (href_musthave_ != null && href_musthave_.indexOf("%") >= 0) {
|
---|
211 |
|
---|
212 | tmp = href_musthave_.substring(0, href_musthave_.indexOf("%"));
|
---|
213 | if (href_musthave_.length() > href_musthave_.indexOf("%") + 1)
|
---|
214 | href_musthave_ = href_musthave_.substring(href_musthave_.indexOf("%") + 1, href_musthave_.length());
|
---|
215 | else
|
---|
216 | href_musthave_ = null;
|
---|
217 |
|
---|
218 | if (url_string.indexOf(tmp) >= 0)
|
---|
219 | has_href = true;
|
---|
220 | }
|
---|
221 |
|
---|
222 | if (href_musthave_ != null && url_string.indexOf(href_musthave_) >= 0)
|
---|
223 | has_href = true;
|
---|
224 |
|
---|
225 | href_musthave_ = original_href_musthave_;
|
---|
226 | }
|
---|
227 |
|
---|
228 | tmp = image_mustnothave_;
|
---|
229 | String original_image_mustnothave_ = image_mustnothave_;
|
---|
230 |
|
---|
231 | // checks that it doesn't contain this content
|
---|
232 | if (image_mustnothave_ != null) {
|
---|
233 |
|
---|
234 | while (image_mustnothave_ != null && image_mustnothave_.indexOf("%") >= 0) {
|
---|
235 |
|
---|
236 | tmp = image_mustnothave_.substring(0, image_mustnothave_.indexOf("%"));
|
---|
237 | if (image_mustnothave_.length() > image_mustnothave_.indexOf("%") + 1)
|
---|
238 | image_mustnothave_ = image_mustnothave_.substring(image_mustnothave_.indexOf("%") + 1, image_mustnothave_.length());
|
---|
239 | else
|
---|
240 | image_mustnothave_ = null;
|
---|
241 |
|
---|
242 | if (url_string.indexOf(tmp) >= 0)
|
---|
243 | has_href = false;
|
---|
244 | }
|
---|
245 | if (image_mustnothave_ != null && url_string.indexOf(image_mustnothave_) >= 0)
|
---|
246 | has_href = false;
|
---|
247 |
|
---|
248 | image_mustnothave_ = original_image_mustnothave_;
|
---|
249 | }
|
---|
250 |
|
---|
251 | // return true if the link is valid and false if not
|
---|
252 | if (href_musthave_==null || has_href)
|
---|
253 | {
|
---|
254 | // might be another URL
|
---|
255 | if (depth < app_.maxDepth())
|
---|
256 | {
|
---|
257 | if (!new_url_string.startsWith(url_string))
|
---|
258 | {
|
---|
259 | return true;
|
---|
260 | }
|
---|
261 | }
|
---|
262 | }
|
---|
263 | return false;
|
---|
264 | }
|
---|
265 |
|
---|
266 | /** Adds an image to the stored downloaded images as a triplet.
|
---|
267 | * Ensures that the number of images downloaded but not displayed at
|
---|
268 | * anyone time is controlled by using a buffer. If the buffer is
|
---|
269 | * full this function will wait until space becomes available before
|
---|
270 | * continuing. It also restricts the
|
---|
271 | * total number of images to download as specified by the applet.
|
---|
272 | *
|
---|
273 | * @param url the image to download
|
---|
274 | * @param from_url the url that this image was sourced from
|
---|
275 | * @param img_name the name of the image */
|
---|
276 | public void add_image(URL url, String from_url, String img_name)
|
---|
277 | {
|
---|
278 | try {
|
---|
279 |
|
---|
280 | boolean had_to_wait = false;
|
---|
281 |
|
---|
282 | // ensure that we don't download too many images
|
---|
283 | while (download_images_.size() >= buffer_size_)
|
---|
284 | {
|
---|
285 | had_to_wait = true;
|
---|
286 | Thread.sleep(delay_);
|
---|
287 | }
|
---|
288 |
|
---|
289 | // get the image from the url
|
---|
290 | Image image = Toolkit.getDefaultToolkit().getImage(url);
|
---|
291 |
|
---|
292 | // push image onto the downloaded images
|
---|
293 | download_images_.push(image,from_url, img_name);
|
---|
294 |
|
---|
295 | // if have completed the maximum number of downloads for the
|
---|
296 | // application then stop
|
---|
297 | if (download_images_.size() == app_.maxDownloads())
|
---|
298 | {
|
---|
299 | stop();
|
---|
300 | }
|
---|
301 |
|
---|
302 | }
|
---|
303 | catch (Exception e) {
|
---|
304 | thread_running_ = false;
|
---|
305 | stop();
|
---|
306 | e.printStackTrace();
|
---|
307 | }
|
---|
308 | }
|
---|
309 |
|
---|
310 | /** Connects to the starting url and looks for all images and links from this
|
---|
311 | * original page. Image links are processed first, so that any images found can be
|
---|
312 | * downloaded immediately and placed on the applet. Secondly, the links to other
|
---|
313 | * pages are recursively processed by this function and treated as a starting url
|
---|
314 | *
|
---|
315 | * @param new_url the url from which to start searching for images and links
|
---|
316 | * @param depth the number of links that have been followed on this path */
|
---|
317 | public void rec_add_images(String new_url, int depth)
|
---|
318 | {
|
---|
319 | System.err.println("Parsing url = " + new_url);
|
---|
320 |
|
---|
321 | if (already_visited(new_url)) return;
|
---|
322 |
|
---|
323 | // check if there is a scenario where external hyperlinks are being used
|
---|
324 | externalLinks();
|
---|
325 | String img_name = new String();
|
---|
326 |
|
---|
327 | // connect to the url
|
---|
328 | CURL curl = new CURL(new_url);
|
---|
329 | if (curl.connected_ok())
|
---|
330 | {
|
---|
331 | // read the page
|
---|
332 | curl.readAll();
|
---|
333 |
|
---|
334 | // get all the <code><img src=</code> links into a vector
|
---|
335 | Vector src_links = curl.getSrcLinks();
|
---|
336 |
|
---|
337 | // process each of the image links according to the parameters given.
|
---|
338 | for (int i = 0; i < src_links.size(); i++)
|
---|
339 | {
|
---|
340 | URL url = (URL)src_links.elementAt(i);
|
---|
341 | String url_string = url.toString();
|
---|
342 |
|
---|
343 | if (image_file_extension(url_string))
|
---|
344 | {
|
---|
345 | if (filter_image(url_string))
|
---|
346 | {
|
---|
347 | img_name = url_string.substring(url_string.lastIndexOf("/") + 1, url_string.length());
|
---|
348 |
|
---|
349 | if (external_links_ != null && ! external_links_.isEmpty()) {
|
---|
350 | String ext = (String) external_links_.get(img_name);
|
---|
351 |
|
---|
352 | if (ext != null)
|
---|
353 | add_image(url, ext, img_name);
|
---|
354 | else
|
---|
355 | add_image(url, new_url, img_name);
|
---|
356 | }
|
---|
357 | else {
|
---|
358 | add_image(url, new_url, img_name);
|
---|
359 | }
|
---|
360 | }
|
---|
361 | }
|
---|
362 |
|
---|
363 | }
|
---|
364 |
|
---|
365 | // get all the <code><a href=</code> links into a vector
|
---|
366 | Vector href_links = curl.getHrefLinks();
|
---|
367 |
|
---|
368 | // process each of the href links according to the parameters given.
|
---|
369 | for (int i = 0; i < href_links.size(); i++)
|
---|
370 | {
|
---|
371 | URL url = (URL)href_links.elementAt(i);
|
---|
372 | String url_string = url.toString();
|
---|
373 |
|
---|
374 | if (image_file_extension(url_string))
|
---|
375 | {
|
---|
376 | if (filter_image(url_string))
|
---|
377 | {
|
---|
378 | img_name = url_string.substring(url_string.lastIndexOf("/") + 1, url_string.length());
|
---|
379 |
|
---|
380 | if (external_links_ != null && ! external_links_.isEmpty()) {
|
---|
381 | String ext = (String) external_links_.get(img_name);
|
---|
382 |
|
---|
383 | if (ext != null)
|
---|
384 | add_image(url, ext, img_name);
|
---|
385 | else
|
---|
386 | add_image(url, new_url, img_name);
|
---|
387 | }
|
---|
388 | else {
|
---|
389 | add_image(url, url_string, img_name);
|
---|
390 | }
|
---|
391 | }
|
---|
392 | }
|
---|
393 | else
|
---|
394 | {
|
---|
395 | if (filter_href(url_string,new_url,depth))
|
---|
396 | {
|
---|
397 | rec_add_images(url_string,depth+1);
|
---|
398 |
|
---|
399 | }
|
---|
400 | }
|
---|
401 | }
|
---|
402 | }
|
---|
403 |
|
---|
404 | else {
|
---|
405 | System.err.println("Unable able to download "+new_url);
|
---|
406 | }
|
---|
407 | }
|
---|
408 |
|
---|
409 |
|
---|
410 | /** Used in cases where the image maps to a url outside of it's original location.
|
---|
411 | * When used with Greenstone the collage images will refer to documents in the collections
|
---|
412 | * from which the images are sourced. When used individually, the images may be saved into
|
---|
413 | * a user directory and the pages they reference may be external hyperlinks.
|
---|
414 | * This function reads that external links file and creates a hash map of the image to
|
---|
415 | * its external hyperlink. If the file does not exist the download thread will continue
|
---|
416 | * and assume the first case, that links are internal. */
|
---|
417 | public void externalLinks() {
|
---|
418 |
|
---|
419 | try {
|
---|
420 |
|
---|
421 | if (starting_url_.indexOf("gsdl") >= 0) {
|
---|
422 | external_links_ = null;
|
---|
423 | return;
|
---|
424 | }
|
---|
425 |
|
---|
426 | // open a url to the file written
|
---|
427 | URL u = new URL(starting_url_ + "externallinks");
|
---|
428 |
|
---|
429 | BufferedReader r = new BufferedReader(new InputStreamReader(u.openStream()));
|
---|
430 |
|
---|
431 | external_links_ = new Hashtable();
|
---|
432 |
|
---|
433 | String l = r.readLine();
|
---|
434 | // split the line of the space, first part is the image, second part the link
|
---|
435 | while (l != null) {
|
---|
436 |
|
---|
437 | String tmp1 = new String();
|
---|
438 | String tmp2 = new String();
|
---|
439 |
|
---|
440 | if (l.indexOf(" ") >= 0) {
|
---|
441 |
|
---|
442 | tmp1 = l.substring(0, l.indexOf(" "));
|
---|
443 | if (l.length() > l.indexOf(" ") + 1)
|
---|
444 | tmp2 = l.substring(l.indexOf(" ") + 1, l.length());
|
---|
445 | else
|
---|
446 | tmp2 = null;
|
---|
447 |
|
---|
448 | if (tmp2 != null) {
|
---|
449 | external_links_.put(tmp1, tmp2);
|
---|
450 | //System.err.println(tmp1 + " " + tmp2);
|
---|
451 | }
|
---|
452 | }
|
---|
453 | l = r.readLine();
|
---|
454 | }
|
---|
455 |
|
---|
456 | r.close();
|
---|
457 |
|
---|
458 | } catch (Exception e) {
|
---|
459 | e.printStackTrace();
|
---|
460 | return;
|
---|
461 | }
|
---|
462 | }
|
---|
463 |
|
---|
464 | /** Controls the download thread */
|
---|
465 | public void run ()
|
---|
466 | {
|
---|
467 | System.err.println("Starting download thread.");
|
---|
468 | visited_url_ = new Hashtable();
|
---|
469 | rec_add_images(starting_url_,1);
|
---|
470 |
|
---|
471 | System.err.println("Download thread finished.");
|
---|
472 | }
|
---|
473 | }
|
---|