private void deleteBlock(CachedUrl cu) throws IOException {
log.info("deleting " + cu.getUrl());
CachedUrlSetSpec cuss = new SingleNodeCachedUrlSetSpec(cu.getUrl());
ArchivalUnit au = cu.getArchivalUnit();
CachedUrlSet cus = au.makeCachedUrlSet(cuss);
NodeManager nm = au.getPlugin().getDaemon().getNodeManager(au);
nm.deleteNode(cus);
}
public void testArticleCountAndType() throws Exception {
int expCount = 28;
PluginTestUtil.crawlSimAu(sau);
String pat1 = "branch(\\d+)/(\\d+file\\.html)";
String rep1 = "aps/journal/v123/n$1/full/$2";
PluginTestUtil.copyAu(sau, nau, ".*[^.][^p][^d][^f]$", pat1, rep1);
String pat2 = "branch(\\d+)/(\\d+file\\.pdf)";
String rep2 = "aps/journal/v123/n$1/pdf/$2";
PluginTestUtil.copyAu(sau, nau, ".*\\.pdf$", pat2, rep2);
// Remove some URLs
int deleted = 0;
for (Iterator it = nau.getAuCachedUrlSet().contentHashIterator(); it.hasNext(); ) {
CachedUrlSetNode cusn = (CachedUrlSetNode) it.next();
if (cusn instanceof CachedUrl) {
CachedUrl cu = (CachedUrl) cusn;
String url = cu.getUrl();
if (url.contains("/journal/")
&& (url.endsWith("1file.html") || url.endsWith("2file.pdf"))) {
deleteBlock(cu);
++deleted;
}
}
}
assertEquals(8, deleted);
Iterator<ArticleFiles> it = nau.getArticleIterator();
int count = 0;
int countHtmlOnly = 0;
int countPdfOnly = 0;
while (it.hasNext()) {
ArticleFiles af = it.next();
log.info(af.toString());
CachedUrl cu = af.getFullTextCu();
String url = cu.getUrl();
assertNotNull(cu);
String contentType = cu.getContentType();
log.debug("count " + count + " url " + url + " " + contentType);
count++;
if (af.getRoleUrl(ArticleFiles.ROLE_FULL_TEXT_PDF) == null) {
++countHtmlOnly;
}
if (af.getRoleUrl(ArticleFiles.ROLE_FULL_TEXT_PDF) == url) {
++countPdfOnly;
}
}
log.debug("Article count is " + count);
assertEquals(expCount, count);
assertEquals(4, countHtmlOnly);
assertEquals(4, countPdfOnly);
}
@Override
public void extract(MetadataTarget target, CachedUrl cu, Emitter emitter) throws IOException {
ArticleMetadata am = new SimpleHtmlMetaTagMetadataExtractor().extract(target, cu);
am.cook(tagMap);
String url = am.get(MetadataField.FIELD_ACCESS_URL);
ArchivalUnit au = cu.getArchivalUnit();
if (url == null || url.isEmpty() || !au.makeCachedUrl(url).hasContent()) {
url = cu.getUrl();
}
am.replace(
MetadataField.FIELD_ACCESS_URL,
HttpToHttpsUtil.AuUtil.normalizeHttpHttpsFromBaseUrl(au, url));
emitter.emitMetadata(cu, am);
}
/** Setup common to most constructors */
private void commonSetup(CachedUrlSet cus, String lwrBound, String uprBound, int pollType) {
CachedUrlSetSpec cuss = cus.getSpec();
if (cuss instanceof PrunedCachedUrlSetSpec) {
throw new IllegalArgumentException("Polls do not support PrunedCachedUrlSetSpec");
}
this.cus = cus;
ArchivalUnit au = cus.getArchivalUnit();
auId = au.getAuId();
this.pluginVersion = AuUtil.getPollVersion(au);
url = cuss.getUrl();
this.lwrBound = lwrBound;
this.uprBound = uprBound;
this.protocolVersion = protocolVersionFromPollType(pollType);
this.pollType = pollType;
}
List<String> auUrls(ArchivalUnit au) {
List<String> res = new ArrayList<String>();
for (Iterator iter = au.getAuCachedUrlSet().contentHashIterator(); iter.hasNext(); ) {
CachedUrlSetNode cusn = (CachedUrlSetNode) iter.next();
if (cusn.hasContent()) {
res.add(cusn.getUrl());
}
}
return res;
}
public BePressArticleIterator(
ArchivalUnit au, SubTreeArticleIterator.Spec spec, boolean isSection) {
super(au, spec);
String volumeAsString = au.getConfiguration().get(ConfigParamDescr.VOLUME_NUMBER.getKey());
String journalAbbr = au.getConfiguration().get(ConfigParamDescr.JOURNAL_ABBR.getKey());
if (isSection) {
journalAbbr = journalAbbr + "/" + au.getConfiguration().get("journal_section");
}
// pick up issue level and lower (make (art)?[0-9]+ optional because a few au's have article
// at issue level
this.pattern =
Pattern.compile(
String.format(
"/%s/((([^0-9]+/)?(vol)?%s/(iss)?[0-9]+(/(art)?[0-9]+)?)|(vol%s/(?-i:[A-Z])[0-9]+))$",
journalAbbr, volumeAsString, volumeAsString),
Pattern.CASE_INSENSITIVE);
this.TOC_pattern =
Pattern.compile(
String.format("/%s/([^0-9]+/)?(vol)?%s/(iss)?[0-9]+$", journalAbbr, volumeAsString),
Pattern.CASE_INSENSITIVE);
}
public static V1Poll createCompletedPoll(
LockssDaemon daemon,
ArchivalUnit au,
V1LcapMessage testmsg,
int numAgree,
int numDisagree,
PollManager pollmanager)
throws Exception {
log.debug(
"createCompletedPoll: au: "
+ au.toString()
+ " peer "
+ testmsg.getOriginatorId()
+ " votes "
+ numAgree
+ "/"
+ numDisagree);
CachedUrlSetSpec cusSpec = null;
if ((testmsg.getLwrBound() != null)
&& (testmsg.getLwrBound().equals(PollSpec.SINGLE_NODE_LWRBOUND))) {
cusSpec = new SingleNodeCachedUrlSetSpec(testmsg.getTargetUrl());
} else {
cusSpec =
new RangeCachedUrlSetSpec(
testmsg.getTargetUrl(), testmsg.getLwrBound(), testmsg.getUprBound());
}
CachedUrlSet cus = au.makeCachedUrlSet(cusSpec);
PollSpec spec = new PollSpec(cus, Poll.V1_CONTENT_POLL);
((MockCachedUrlSet) spec.getCachedUrlSet()).setHasContent(false);
V1Poll p = null;
if (testmsg.isContentPoll()) {
p =
new V1ContentPoll(
spec,
pollmanager,
testmsg.getOriginatorId(),
testmsg.getChallenge(),
testmsg.getDuration(),
testmsg.getHashAlgorithm());
} else if (testmsg.isNamePoll()) {
p =
new V1NamePoll(
spec,
pollmanager,
testmsg.getOriginatorId(),
testmsg.getChallenge(),
testmsg.getDuration(),
testmsg.getHashAlgorithm());
} else if (testmsg.isVerifyPoll()) {
p =
new V1VerifyPoll(
spec,
pollmanager,
testmsg.getOriginatorId(),
testmsg.getChallenge(),
testmsg.getDuration(),
testmsg.getHashAlgorithm(),
testmsg.getVerifier());
}
assertNotNull(p);
p.setMessage(testmsg);
p.m_tally.quorum = numAgree + numDisagree;
p.m_tally.numAgree = numAgree;
p.m_tally.numDisagree = numDisagree;
p.m_tally.wtAgree = 2000;
p.m_tally.wtDisagree = 200;
p.m_tally.localEntries = makeEntries(1, 3);
p.m_tally.votedEntries = makeEntries(1, 5);
p.m_tally.votedEntries.remove(1);
p.m_pollstate = V1Poll.PS_COMPLETE;
p.m_callerID = testmsg.getOriginatorId();
log.debug3("poll " + p.toString());
p.m_tally.tallyVotes();
return p;
}
/** Explode the archive into its constituent elements */
public void explode() throws CacheException {
CachedUrl cachedUrl = null;
int goodEntries = 0;
int badEntries = 0;
int ignoredEntries = 0;
int entriesBetweenSleep = 0;
ArchiveReader arcReader = null;
logger.info(
(storeArchive ? "Storing" : "Fetching") + " WARC file: " + origUrl + " will explode");
try {
if (storeArchive) {
UrlCacher uc = au.makeUrlCacher(new UrlData(arcStream, arcProps, fetchUrl));
BitSet bs = new BitSet();
bs.set(UrlCacher.DONT_CLOSE_INPUT_STREAM_FLAG);
uc.setFetchFlags(bs);
uc.storeContent();
archiveData.resetInputStream();
arcStream = archiveData.input;
}
// Wrap it in an ArchiveReader
logger.debug3("About to wrap stream");
arcReader = wrapStream(fetchUrl, arcStream);
logger.debug3("wrapStream() returns " + (arcReader == null ? "null" : "non-null"));
// Explode it
if (arcReader == null) {
throw new CacheException.ExploderException("no WarcReader for " + origUrl);
}
ArchivalUnit au = crawlFacade.getAu();
Set stemSet = new HashSet();
logger.debug("Exploding " + fetchUrl);
// Iterate through the elements in the WARC file, except the first
Iterator i = arcReader.iterator();
// Skip first record
for (i.next(); i.hasNext(); ) {
// XXX probably not necessary
helper.pokeWDog();
if ((++entriesBetweenSleep % sleepAfter) == 0) {
long pauseTime =
CurrentConfig.getTimeIntervalParam(PARAM_RETRY_PAUSE, DEFAULT_RETRY_PAUSE);
Deadline pause = Deadline.in(pauseTime);
logger.debug3("Sleeping for " + StringUtil.timeIntervalToString(pauseTime));
while (!pause.expired()) {
try {
pause.sleep();
} catch (InterruptedException ie) {
// no action
}
}
}
ArchiveRecord element = (ArchiveRecord) i.next();
// Each element is a URL to be cached in a suitable AU
ArchiveRecordHeader elementHeader = element.getHeader();
String elementUrl = elementHeader.getUrl();
String elementMimeType = elementHeader.getMimetype();
long elementLength = elementHeader.getLength();
logger.debug2("WARC url " + elementUrl + " mime " + elementMimeType);
if (elementUrl.startsWith("http:")) {
ArchiveEntry ae =
new ArchiveEntry(
elementUrl,
elementLength,
0, // XXX need to convert getDate string to long
element, // ArchiveRecord extends InputStream
this,
fetchUrl);
ae.setHeaderFields(makeCIProperties(elementHeader));
long bytesStored = elementLength;
logger.debug3("ArchiveEntry: " + ae.getName() + " bytes " + bytesStored);
try {
helper.process(ae);
} catch (PluginException ex) {
throw new CacheException.ExploderException("helper.process() threw", ex);
}
if (ae.getBaseUrl() != null) {
if (ae.getRestOfUrl() != null && ae.getHeaderFields() != null) {
storeEntry(ae);
handleAddText(ae);
goodEntries++;
crawlFacade.getCrawlerStatus().addContentBytesFetched(bytesStored);
} else {
ignoredEntries++;
}
} else {
badEntries++;
logger.debug2("Can't map " + elementUrl + " from " + archiveUrl);
}
}
}
} catch (IOException ex) {
throw new CacheException.ExploderException(ex);
} finally {
if (arcReader != null)
try {
arcReader.close();
arcReader = null;
} catch (IOException ex) {
throw new CacheException.ExploderException(ex);
}
if (cachedUrl != null) {
cachedUrl.release();
}
IOUtil.safeClose(arcStream);
}
if (badEntries == 0 && goodEntries > 0) {
// Make it look like a new crawl finished on each AU to which
// URLs were added.
for (Iterator it = touchedAus.iterator(); it.hasNext(); ) {
ArchivalUnit au = (ArchivalUnit) it.next();
logger.debug3(archiveUrl + " touching " + au.toString());
AuUtil.getDaemon(au).getNodeManager(au).newContentCrawlFinished();
}
} else {
ArchivalUnit au = crawlFacade.getAu();
String msg = archiveUrl + ": " + badEntries + "/" + goodEntries + " bad entries";
throw new CacheException.UnretryableException(msg);
}
}
