indexSegment.urlMetadata().remove(urlhash);
sb.crawlQueues.noticeURL.removeByURLHash(urlhash);
sb.crawlQueues.errorURL.remove(urlhash);
// get a scraper to get the title
final ContentScraper scraper = sb.loader.parseResource(url, CacheStrategy.IFFRESH);
final String title = scraper == null ? url.toNormalform(true, true) : scraper.getTitle();
final String description = scraper.getDescription();
// stack url
sb.crawler.removePassive(crawlingStartURL.hash()); // if there is an old entry, delete it
final CrawlProfile pe = new CrawlProfile(
(crawlingStartURL.getHost() == null) ? crawlingStartURL.toNormalform(true, false) : crawlingStartURL.getHost(),
crawlingStartURL,
newcrawlingMustMatch,
newcrawlingMustNotMatch,
ipMustMatch,
ipMustNotMatch,
countryMustMatch,
newcrawlingdepth,
directDocByURL,
crawlingIfOlder,
crawlingDomMaxPages,
crawlingQ,
indexText, indexMedia,
storeHTCache,
crawlOrder,
xsstopw,
xdstopw,
xpstopw,
cachePolicy);
sb.crawler.putActive(pe.handle().getBytes(), pe);
final String reasonString = sb.crawlStacker.stackCrawl(new Request(
sb.peers.mySeed().hash.getBytes(),
url,
null,
"CRAWLING-ROOT",
new Date(),
pe.handle(),
0,
0,
0,
0
));
if (reasonString == null) {
// create a bookmark from crawl start url
//final Set<String> tags=ListManager.string2set(BookmarkHelper.cleanTagsString(post.get("bookmarkFolder","/crawlStart")));
final Set<String> tags=ListManager.string2set(BookmarkHelper.cleanTagsString("/crawlStart"));
tags.add("crawlStart");
final String[] keywords = scraper.getKeywords();
if (keywords != null) {
for (final String k: keywords) {
final String kk = BookmarkHelper.cleanTagsString(k);
if (kk.length() > 0) tags.add(kk);
}
}
String tagStr = tags.toString();
if (tagStr.length() > 2 && tagStr.startsWith("[") && tagStr.endsWith("]")) tagStr = tagStr.substring(1, tagStr.length() - 2);
// we will create always a bookmark to use this to track crawled hosts
final BookmarksDB.Bookmark bookmark = sb.bookmarksDB.createBookmark(crawlingStart, "admin");
if (bookmark != null) {
bookmark.setProperty(BookmarksDB.Bookmark.BOOKMARK_TITLE, title);
bookmark.setProperty(BookmarksDB.Bookmark.BOOKMARK_DESCRIPTION, description);
bookmark.setOwner("admin");
bookmark.setPublic(false);
bookmark.setTags(tags, true);
sb.bookmarksDB.saveBookmark(bookmark);
}
// do the same for ymarks
// TODO: could a non admin user add crawls?
sb.tables.bookmarks.createBookmark(sb.loader, url, YMarkTables.USER_ADMIN, true, "crawlStart", "/Crawl Start");
// liftoff!
prop.put("info", "8");//start msg
prop.putHTML("info_crawlingURL", post.get("crawlingURL"));
// generate a YaCyNews if the global flag was set
if (!sb.isRobinsonMode() && crawlOrder) {
final Map<String, String> m = new HashMap<String, String>(pe); // must be cloned
m.remove("specificDepth");
m.remove("indexText");
m.remove("indexMedia");
m.remove("remoteIndexing");
m.remove("xsstopw");
m.remove("xpstopw");
m.remove("xdstopw");
m.remove("storeTXCache");
m.remove("storeHTCache");
m.remove("generalFilter");
m.remove("specificFilter");
m.put("intention", post.get("intention", "").replace(',', '/'));
sb.peers.newsPool.publishMyNews(sb.peers.mySeed(), NewsPool.CATEGORY_CRAWL_START, m);
}
} else {
prop.put("info", "5"); //Crawling failed
prop.putHTML("info_crawlingURL", (post.get("crawlingURL")));
prop.putHTML("info_reasonString", reasonString);
sb.crawlQueues.errorURL.push(
new Request(
sb.peers.mySeed().hash.getBytes(),
crawlingStartURL,
null,
"",
new Date(),
pe.handle(),
0,
0,
0,
0),
sb.peers.mySeed().hash.getBytes(),
new Date(),
1,
FailCategory.FINAL_LOAD_CONTEXT,
reasonString, -1);
}
} catch (final PatternSyntaxException e) {
prop.put("info", "4"); // crawlfilter does not match url
prop.putHTML("info_newcrawlingfilter", newcrawlingMustMatch);
prop.putHTML("info_error", e.getMessage());
} catch (final Exception e) {
// mist
prop.put("info", "6"); // Error with url
prop.putHTML("info_crawlingStart", crawlingStart);
prop.putHTML("info_error", e.getMessage());
Log.logException(e);
}
} else if ("file".equals(crawlingMode)) {
if (post.containsKey("crawlingFile")) {
final String crawlingFileContent = post.get("crawlingFile$file", "");
try {
// check if the crawl filter works correctly
Pattern.compile(newcrawlingMustMatch);
final ContentScraper scraper = new ContentScraper(new DigestURI(crawlingFile));
final Writer writer = new TransformerWriter(null, null, scraper, null, false);
if (crawlingFile != null && crawlingFile.exists()) {
FileUtils.copy(new FileInputStream(crawlingFile), writer);
} else {
FileUtils.copy(crawlingFileContent, writer);
}
writer.close();
// get links and generate filter
final Map<MultiProtocolURI, Properties> hyperlinks = scraper.getAnchors();
if (fullDomain && newcrawlingdepth > 0) newcrawlingMustMatch = siteFilter(hyperlinks.keySet());
final DigestURI crawlURL = new DigestURI("file://" + crawlingFile.toString());
final CrawlProfile profile = new CrawlProfile(
crawlingFileName,
crawlURL,
newcrawlingMustMatch,
CrawlProfile.MATCH_NEVER_STRING,
ipMustMatch,
ipMustNotMatch,
countryMustMatch,
newcrawlingdepth,
false,
crawlingIfOlder,
crawlingDomMaxPages,
crawlingQ,
indexText,
indexMedia,
storeHTCache,
crawlOrder,
xsstopw,
xdstopw,
xpstopw,
cachePolicy);
sb.crawler.putActive(profile.handle().getBytes(), profile);
sb.pauseCrawlJob(SwitchboardConstants.CRAWLJOB_LOCAL_CRAWL);
sb.crawlStacker.enqueueEntriesAsynchronous(sb.peers.mySeed().hash.getBytes(), profile.handle(), hyperlinks, true);
} catch (final PatternSyntaxException e) {
prop.put("info", "4"); // crawlfilter does not match url
prop.putHTML("info_newcrawlingfilter", newcrawlingMustMatch);
prop.putHTML("info_error", e.getMessage());
} catch (final Exception e) {
// mist
prop.put("info", "7"); // Error with file
prop.putHTML("info_crawlingStart", crawlingFileName);
prop.putHTML("info_error", e.getMessage());
Log.logException(e);
}
sb.continueCrawlJob(SwitchboardConstants.CRAWLJOB_LOCAL_CRAWL);
}
} else if ("sitemap".equals(crawlingMode)) {
final String sitemapURLStr = post.get("sitemapURL","");
try {
final DigestURI sitemapURL = new DigestURI(sitemapURLStr);
final CrawlProfile pe = new CrawlProfile(
sitemapURLStr,
sitemapURL,
CrawlProfile.MATCH_ALL_STRING,
CrawlProfile.MATCH_NEVER_STRING,
ipMustMatch,
ipMustNotMatch,
countryMustMatch,
0,
false,
crawlingIfOlder,
crawlingDomMaxPages,
true,
indexText,
indexMedia,
storeHTCache,
crawlOrder,
xsstopw,
xdstopw,
xpstopw,
cachePolicy);
sb.crawler.putActive(pe.handle().getBytes(), pe);
final SitemapImporter importer = new SitemapImporter(sb, sitemapURL, pe);
importer.start();
} catch (final Exception e) {
// mist
prop.put("info", "6");//Error with url
prop.putHTML("info_crawlingStart", sitemapURLStr);
prop.putHTML("info_error", e.getMessage());
Log.logException(e);
}
} else if ("sitelist".equals(crawlingMode)) {
try {
final DigestURI sitelistURL = new DigestURI(crawlingStart);
// download document
ContentScraper scraper = null;
scraper = sb.loader.parseResource(sitelistURL, CacheStrategy.IFFRESH);
// String title = scraper.getTitle();
// String description = scraper.getDescription();
// get links and generate filter
final Map<MultiProtocolURI, Properties> hyperlinks = scraper.getAnchors();
if (fullDomain && newcrawlingdepth > 0) newcrawlingMustMatch = siteFilter(hyperlinks.keySet());
// put links onto crawl queue
final CrawlProfile profile = new CrawlProfile(
sitelistURL.getHost(),