prop.putHTML("keystring", keystring);
prop.putHTML("keyhash", ASCII.String(keyhash));
// read values from checkboxes
final String[] urls = post.getAll("urlhx.*");
HandleSet urlb = new HandleSet(URIMetadataRow.rowdef.primaryKeyLength, URIMetadataRow.rowdef.objectOrder, urls.length);
if (urls != null) for (final String s: urls) try { urlb.put(s.getBytes()); } catch (final RowSpaceExceededException e) { Log.logException(e); }
final boolean delurl = post.containsKey("delurl");
final boolean delurlref = post.containsKey("delurlref");
if (post.containsKey("keystringsearch")) {
keyhash = Word.word2hash(keystring);
prop.put("keyhash", keyhash);
final RWIProcess ranking = genSearchresult(prop, sb, segment, keyhash, null);
if (ranking.filteredCount() == 0) {
prop.put("searchresult", 1);
prop.putHTML("searchresult_word", keystring);
}
}
if (post.containsKey("keyhashsearch")) {
if (keystring.length() == 0 || !ByteBuffer.equals(Word.word2hash(keystring), keyhash)) {
prop.put("keystring", "<not possible to compute word from hash>");
}
final RWIProcess ranking = genSearchresult(prop, sb, segment, keyhash, null);
if (ranking.filteredCount() == 0) {
prop.put("searchresult", 2);
prop.putHTML("searchresult_wordhash", ASCII.String(keyhash));
}
}
// delete everything
if (post.containsKey("deletecomplete")) {
if (post.get("deleteIndex", "").equals("on")) {
segment.clear();
}
if (post.get("deleteSolr", "").equals("on") && sb.getConfigBool("federated.service.solr.indexing.enabled", false)) try {
sb.indexSegments.segment(Segments.Process.LOCALCRAWLING).getSolr().clear();
} catch (final Exception e) {
Log.logException(e);
}
if (post.get("deleteCrawlQueues", "").equals("on")) {
sb.crawlQueues.clear();
sb.crawlStacker.clear();
ResultURLs.clearStacks();
}
if (post.get("deleteCache", "").equals("on")) {
Cache.clear();
}
if (post.get("deleteRobots", "").equals("on")) {
sb.robots.clear();
}
if (post.get("deleteSearchFl", "").equals("on")) {
sb.tables.clear(WorkTables.TABLE_SEARCH_FAILURE_NAME);
}
post.remove("deletecomplete");
}
// set reference limitation
if (post.containsKey("maxReferencesLimit")) {
if (post.get("maxReferencesRadio", "").equals("on")) {
ReferenceContainer.maxReferences = post.getInt("maxReferences", 0);
} else {
ReferenceContainer.maxReferences = 0;
}
sb.setConfig("index.maxReferences", ReferenceContainer.maxReferences);
}
// delete word
if (post.containsKey("keyhashdeleteall")) try {
if (delurl || delurlref) {
// generate urlx: an array of url hashes to be deleted
ReferenceContainer<WordReference> index = null;
index = segment.termIndex().get(keyhash, null);
final Iterator<WordReference> en = index.entries();
urlb = new HandleSet(URIMetadataRow.rowdef.primaryKeyLength, URIMetadataRow.rowdef.objectOrder, index.size());
while (en.hasNext()) try { urlb.put(en.next().urlhash()); } catch (final RowSpaceExceededException e) { Log.logException(e); }
index = null;
}
if (delurlref) {
segment.removeAllUrlReferences(urlb, sb.loader, CacheStrategy.IFEXIST);
}
// delete the word first because that is much faster than the deletion of the urls from the url database
segment.termIndex().delete(keyhash);
// now delete all urls if demanded
if (delurl || delurlref) {
for (final byte[] b: urlb) sb.urlRemove(segment, b);
}
post.remove("keyhashdeleteall");
post.put("urllist", "generated");
} catch (final IOException e) {
Log.logException(e);
}
// delete selected URLs
if (post.containsKey("keyhashdelete")) try {
if (delurlref) {
segment.removeAllUrlReferences(urlb, sb.loader, CacheStrategy.IFEXIST);
}
if (delurl || delurlref) {
for (final byte[] b: urlb) sb.urlRemove(segment, b);
}
final HandleSet urlHashes = new HandleSet(URIMetadataRow.rowdef.primaryKeyLength, URIMetadataRow.rowdef.objectOrder, 0);
for (final byte[] b: urlb) try { urlHashes.put(b); } catch (final RowSpaceExceededException e) { Log.logException(e); }
segment.termIndex().remove(keyhash, urlHashes);
// this shall lead to a presentation of the list; so handle that the remaining program
// thinks that it was called for a list presentation
post.remove("keyhashdelete");
post.put("urllist", "generated");
} catch (final IOException e) {
Log.logException(e);
}
if (post.containsKey("urllist")) {
if (keystring.length() == 0 || !ByteBuffer.equals(Word.word2hash(keystring), keyhash)) {
prop.put("keystring", "<not possible to compute word from hash>");
}
final Bitfield flags = compileFlags(post);
final int count = (post.get("lines", "all").equals("all")) ? -1 : post.getInt("lines", -1);
final RWIProcess ranking = genSearchresult(prop, sb, segment, keyhash, flags);
genURLList(prop, keyhash, keystring, ranking, flags, count);
}
// transfer to other peer
if (post.containsKey("keyhashtransfer")) try {
if (keystring.length() == 0 || !ByteBuffer.equals(Word.word2hash(keystring), keyhash)) {
prop.put("keystring", "<not possible to compute word from hash>");
}
// find host & peer
String host = post.get("host", ""); // get host from input field
Seed seed = null;
if (host.length() != 0) {
if (host.length() == 12) {
// the host string is a peer hash
seed = sb.peers.getConnected(host);
} else {
// the host string can be a host name
seed = sb.peers.lookupByName(host);
}
} else {
host = post.get("hostHash", ""); // if input field is empty, get from select box
seed = sb.peers.getConnected(host);
}
// prepare index
ReferenceContainer<WordReference> index;
final long starttime = System.currentTimeMillis();
index = segment.termIndex().get(keyhash, null);
// built urlCache
final Iterator<WordReference> urlIter = index.entries();
final TreeMap<byte[], URIMetadataRow> knownURLs = new TreeMap<byte[], URIMetadataRow>(Base64Order.enhancedCoder);
final HandleSet unknownURLEntries = new HandleSet(WordReferenceRow.urlEntryRow.primaryKeyLength, WordReferenceRow.urlEntryRow.objectOrder, index.size());
Reference iEntry;
URIMetadataRow lurl;
while (urlIter.hasNext()) {
iEntry = urlIter.next();
lurl = segment.urlMetadata().load(iEntry.urlhash());
if (lurl == null) {
try {
unknownURLEntries.put(iEntry.urlhash());
} catch (final RowSpaceExceededException e) {
Log.logException(e);
}
urlIter.remove();
} else {
knownURLs.put(iEntry.urlhash(), lurl);
}
}
// make an indexContainerCache
final ReferenceContainerCache<WordReference> icc = new ReferenceContainerCache<WordReference>(Segment.wordReferenceFactory, Segment.wordOrder, Word.commonHashLength);
try {
icc.add(index);
} catch (final RowSpaceExceededException e) {
Log.logException(e);
}
// transport to other peer
final boolean gzipBody = sb.getConfigBool("indexControl.gzipBody", false);
final int timeout = (int) sb.getConfigLong("indexControl.timeout", 60000);
final String error = Protocol.transferIndex(
seed,
icc,
knownURLs,
gzipBody,
timeout);
prop.put("result", (error == null) ? ("Successfully transferred " + knownURLs.size() + " words in " + ((System.currentTimeMillis() - starttime) / 1000) + " seconds, " + unknownURLEntries.size() + " URL not found") : "error: " + error);
index = null;
} catch (final IOException e) {
Log.logException(e);
}
// generate list
if (post.containsKey("keyhashsimilar")) try {
final Iterator<ReferenceContainer<WordReference>> containerIt = segment.termIndex().referenceContainer(keyhash, true, 256, false).iterator();
ReferenceContainer<WordReference> container;
i = 0;
int rows = 0, cols = 0;
prop.put("keyhashsimilar", "1");
while (containerIt.hasNext() && i < 256) {
container = containerIt.next();
prop.put("keyhashsimilar_rows_"+rows+"_cols_"+cols+"_wordHash", container.getTermHash());
cols++;
if (cols==8) {
prop.put("keyhashsimilar_rows_"+rows+"_cols", cols);
cols = 0;
rows++;
}
i++;
}
prop.put("keyhashsimilar_rows_"+rows+"_cols", cols);
prop.put("keyhashsimilar_rows", rows + 1);
prop.put("result", "");
} catch (final IOException e) {
Log.logException(e);
}
if (post.containsKey("blacklist")) {
final String blacklist = post.get("blacklist", "");
final HandleSet urlHashes = new HandleSet(URIMetadataRow.rowdef.primaryKeyLength, URIMetadataRow.rowdef.objectOrder, urlb.size());
if (post.containsKey("blacklisturls")) {
PrintWriter pw;
try {
final String[] supportedBlacklistTypes = env.getConfig("BlackLists.types", "").split(",");
pw = new PrintWriter(new FileWriter(new File(ListManager.listsPath, blacklist), true));
DigestURI url;
for (final byte[] b: urlb) {
try {
urlHashes.put(b);
} catch (final RowSpaceExceededException e) {
Log.logException(e);
}
final URIMetadataRow e = segment.urlMetadata().load(b);
segment.urlMetadata().remove(b);
if (e != null) {
url = e.metadata().url();
pw.println(url.getHost() + "/" + url.getFile());
for (final String supportedBlacklistType : supportedBlacklistTypes) {
if (ListManager.listSetContains(supportedBlacklistType + ".BlackLists", blacklist)) {
Switchboard.urlBlacklist.add(
supportedBlacklistType,
url.getHost(),
url.getFile());
}
}
SearchEventCache.cleanupEvents(true);
}
}
pw.close();
} catch (final IOException e) {
}
}
if (post.containsKey("blacklistdomains")) {
PrintWriter pw;
try {
final String[] supportedBlacklistTypes = Blacklist.BLACKLIST_TYPES_STRING.split(",");
pw = new PrintWriter(new FileWriter(new File(ListManager.listsPath, blacklist), true));
DigestURI url;
for (final byte[] b: urlb) {
try {
urlHashes.put(b);
} catch (final RowSpaceExceededException e) {
Log.logException(e);
}
final URIMetadataRow e = segment.urlMetadata().load(b);
segment.urlMetadata().remove(b);