/** Create LockssKeystores from config subtree below {@link #PARAM_KEYSTORE} */
void configureKeyStores(Configuration config) {
Configuration allKs = config.getConfigTree(PARAM_KEYSTORE);
for (Iterator iter = allKs.nodeIterator(); iter.hasNext(); ) {
String id = (String) iter.next();
Configuration oneKs = allKs.getConfigTree(id);
try {
LockssKeyStore lk = createLockssKeyStore(oneKs);
String name = lk.getName();
if (name == null) {
log.error("KeyStore definition missing name: " + oneKs);
continue;
}
LockssKeyStore old = keystoreMap.get(name);
if (old != null && !lk.equals(old)) {
log.warning(
"Keystore "
+ name
+ " redefined. "
+ "New definition may not take effect until daemon restart");
}
log.debug("Adding keystore " + name);
keystoreMap.put(name, lk);
} catch (Exception e) {
log.error("Couldn't create keystore: " + oneKs, e);
}
}
}
/**
* Return a TitleConfig for the AU. Returns matching entry from the title db if found, else
* creates one
*/
TitleConfig titleConfigFromAu(InactiveAuProxy au) {
PluginProxy plugin = au.getPlugin();
String auname = au.getName();
Configuration auConfig = au.getConfiguration();
TitleConfig tc = AuUtil.findTitleConfig(auConfig, plugin.getPlugin());
if (tc != null) {
return tc;
}
tc = new TitleConfig(auname, plugin.getPluginId());
ArrayList<ConfigParamAssignment> params = new ArrayList();
for (Iterator iter = auConfig.keyIterator(); iter.hasNext(); ) {
String key = (String) iter.next();
if (!ConfigParamDescr.isReservedParam(key)) {
String val = auConfig.get(key);
ConfigParamDescr descr = plugin.findAuConfigDescr(key);
if (descr != null) {
ConfigParamAssignment cpa = new ConfigParamAssignment(descr, val);
params.add(cpa);
} else {
log.warning("Unknown parameter key: " + key + " in au: " + auname);
}
}
}
params.trimToSize();
tc.setParams(params);
return tc;
}
protected void checkLeaf(SimulatedArchivalUnit sau) {
log.debug("checkLeaf()");
String parent = sau.getUrlRoot() + "/branch1";
CachedUrlSetSpec spec = new RangeCachedUrlSetSpec(parent);
CachedUrlSet set = sau.makeCachedUrlSet(spec);
Iterator setIt = set.contentHashIterator();
ArrayList childL = new ArrayList(16);
while (setIt.hasNext()) {
childL.add(((CachedUrlSetNode) setIt.next()).getUrl());
}
String[] expectedA =
new String[] {
parent,
parent + "/001file.html",
parent + "/001file.txt",
parent + "/002file.html",
parent + "/002file.txt",
parent + "/branch1",
parent + "/branch1/001file.html",
parent + "/branch1/001file.txt",
parent + "/branch1/002file.html",
parent + "/branch1/002file.txt",
parent + "/branch1/index.html",
parent + "/branch2",
parent + "/branch2/001file.html",
parent + "/branch2/001file.txt",
parent + "/branch2/002file.html",
parent + "/branch2/002file.txt",
parent + "/branch2/index.html",
parent + "/index.html",
};
assertIsomorphic(expectedA, childL);
}
/**
* Method that creates a simulated Cached URL from the source code provided by the goodContent
* String. It then asserts that the metadata extracted, by using the
* MetaPressRisMetadataExtractorFactory, match the metadata in the source code.
*
* @throws Exception
*/
public void testExtractGoodRisContent() throws Exception {
String goodContent = createGoodRisContent();
log.debug3(goodContent);
List<ArticleMetadata> mdlist = setupContentForAU(bau1, RIS_URL, goodContent, false);
assertNotEmpty(mdlist);
ArticleMetadata md = mdlist.get(0);
assertNotNull(md);
assertEquals(goodVolume, md.get(MetadataField.FIELD_VOLUME));
assertEquals(goodIssue, md.get(MetadataField.FIELD_ISSUE));
assertEquals(goodStartPage, md.get(MetadataField.FIELD_START_PAGE));
assertEquals(goodEndPage, md.get(MetadataField.FIELD_END_PAGE));
assertEquals(goodIssn, md.get(MetadataField.FIELD_ISSN));
Iterator<String> actAuthIter = md.getList(MetadataField.FIELD_AUTHOR).iterator();
for (String expAuth : goodAuthors) {
assertEquals(expAuth, actAuthIter.next());
}
assertEquals(goodTitle, md.get(MetadataField.FIELD_ARTICLE_TITLE));
assertEquals(goodJournal, md.get(MetadataField.FIELD_PUBLICATION_TITLE));
assertEquals(goodDate, md.get(MetadataField.FIELD_DATE));
assertEquals(goodPublisher, md.get(MetadataField.FIELD_PUBLISHER));
assertEquals(goodDOI, md.get(MetadataField.FIELD_DOI));
// This shouldn't get set. It will default later to fuill_text_cu
assertNotEquals(doiURL, md.get(MetadataField.FIELD_ACCESS_URL));
}
private V3LcapMessage makeTestVoteMessage(Collection voteBlocks) throws IOException {
mPollMgr.setStateDir("key", tempDir);
V3LcapMessage msg =
new V3LcapMessage(
"ArchivalID_2",
"key",
"Plug42",
m_testBytes,
m_testBytes,
V3LcapMessage.MSG_VOTE,
987654321,
m_testID,
tempDir,
theDaemon);
// Set msg vote blocks.
for (Iterator ix = voteBlocks.iterator(); ix.hasNext(); ) {
msg.addVoteBlock((VoteBlock) ix.next());
}
msg.setHashAlgorithm(LcapMessage.getDefaultHashAlgorithm());
msg.setArchivalId(m_archivalID);
msg.setPluginVersion("PlugVer42");
return msg;
}
protected void checkRoot(SimulatedArchivalUnit sau) {
log.debug("checkRoot()");
CachedUrlSet set = sau.getAuCachedUrlSet();
Iterator setIt = set.flatSetIterator();
ArrayList childL = new ArrayList(1);
CachedUrlSet cus = null;
while (setIt.hasNext()) {
cus = (CachedUrlSet) setIt.next();
childL.add(cus.getUrl());
}
String urlRoot = sau.getUrlRoot();
String[] expectedA = new String[1];
expectedA[0] = urlRoot;
assertIsomorphic(expectedA, childL);
setIt = cus.flatSetIterator();
childL = new ArrayList(7);
while (setIt.hasNext()) {
childL.add(((CachedUrlSetNode) setIt.next()).getUrl());
}
expectedA =
new String[] {
urlRoot + "/001file.html",
urlRoot + "/001file.txt",
urlRoot + "/002file.html",
urlRoot + "/002file.txt",
urlRoot + "/branch1",
urlRoot + "/branch2",
urlRoot + "/index.html"
};
assertIsomorphic(expectedA, childL);
}
/**
* Return the titles in the set.
*
* @return a collection of TitleConfig
*/
public Collection<TitleConfig> getTitles() {
Collection aus = daemon.getRemoteApi().getInactiveAus();
ArrayList<TitleConfig> res = new ArrayList<TitleConfig>(aus.size());
for (Iterator iter = aus.iterator(); iter.hasNext(); ) {
InactiveAuProxy aup = (InactiveAuProxy) iter.next();
res.add(titleConfigFromAu(aup));
}
res.trimToSize();
return res;
}
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;
}
/** Concatenate params for URL string */
String concatParams(Properties props) {
if (props == null) {
return null;
}
java.util.List list = new ArrayList();
for (Iterator iter = props.keySet().iterator(); iter.hasNext(); ) {
String key = (String) iter.next();
String val = props.getProperty(key);
if (!StringUtil.isNullString(val)) {
list.add(key + "=" + urlEncode(val));
}
}
return StringUtil.separatedString(list, "&");
}
public List findExistingRepositoriesFor(String auid) {
List res = null;
for (Iterator iter = getRepositoryList().iterator(); iter.hasNext(); ) {
String repoName = (String) iter.next();
String path = LockssRepositoryImpl.getLocalRepositoryPath(repoName);
if (LockssRepositoryImpl.doesAuDirExist(auid, path)) {
if (res == null) {
res = new ArrayList();
}
res.add(repoName);
}
}
return res == null ? Collections.EMPTY_LIST : res;
}
public void testSimpleDatasetXML() throws Exception {
log.debug3("testSimpleDatasetXML");
String file_input = StringUtil.fromInputStream(getResourceAsStream(testDatasetFile));
String xml_url = TAR_A_BASE + SUBDIR + "dataset.xml";
List<ArticleMetadata> mdList =
extractFromContent(xml_url, "text/xml", file_input, nocheck_mle, null);
assertEquals(6, mdList.size());
Iterator<ArticleMetadata> mdIt = mdList.iterator();
ArticleMetadata mdRecord = null;
while (mdIt.hasNext()) {
mdRecord = (ArticleMetadata) mdIt.next();
validateDatasetMetadataRecord(mdRecord);
}
}
/**
* Checks the consistency of the node, and continues with its children if it's consistent.
*
* @param node RepositoryNodeImpl the node to check
*/
private void recurseConsistencyCheck(RepositoryNodeImpl node) {
logger.debug2("Checking node '" + node.getNodeUrl() + "'...");
// check consistency at each node
// correct/deactivate as necessary
// 'checkNodeConsistency()' will repair if possible
if (node.checkNodeConsistency()) {
logger.debug3("Node consistent; recursing on children...");
List children = node.getNodeList(null, false);
Iterator iter = children.iterator();
while (iter.hasNext()) {
RepositoryNodeImpl child = (RepositoryNodeImpl) iter.next();
recurseConsistencyCheck(child);
}
} else {
logger.debug3("Node inconsistent; deactivating...");
deactivateInconsistentNode(node);
}
}
/**
* Method that creates a simulated Cached URL from the source code provided by the goodContent
* String. It then asserts that the metadata extracted, by using the
* MetaPressRisMetadataExtractorFactory, match the metadata in the source code.
*
* @throws Exception
*/
public void testExtractAlternateRisContent() throws Exception {
String goodContent = createAlternateRisContent();
log.debug3(goodContent);
List<ArticleMetadata> mdlist = setupContentForAU(bau1, RIS_URL, goodContent, false);
assertNotEmpty(mdlist);
ArticleMetadata md = mdlist.get(0);
assertNotNull(md);
Iterator<String> actAuthIter = md.getList(MetadataField.FIELD_AUTHOR).iterator();
for (String expAuth : goodAuthors) {
assertEquals(expAuth, actAuthIter.next());
}
assertEquals(goodTitle, md.get(MetadataField.FIELD_ARTICLE_TITLE));
assertEquals(goodJournal, md.get(MetadataField.FIELD_PUBLICATION_TITLE));
assertEquals(goodDate, md.get(MetadataField.FIELD_DATE));
assertEquals(goodPublisher, md.get(MetadataField.FIELD_PUBLISHER));
}
public void testRequestMessageCreation() throws Exception {
V3LcapMessage reqMsg =
new V3LcapMessage(
"ArchivalID_2",
"key",
"Plug42",
m_testBytes,
m_testBytes,
V3LcapMessage.MSG_REPAIR_REQ,
987654321,
m_testID,
tempDir,
theDaemon);
reqMsg.setTargetUrl("http://foo.com/");
for (Iterator ix = m_testVoteBlocks.iterator(); ix.hasNext(); ) {
reqMsg.addVoteBlock((VoteBlock) ix.next());
}
assertEquals(3, reqMsg.getProtocolVersion());
assertEquals("Plug42", reqMsg.getPluginVersion());
assertTrue(m_testID == reqMsg.getOriginatorId());
assertEquals(V3LcapMessage.MSG_REPAIR_REQ, reqMsg.getOpcode());
assertEquals("ArchivalID_2", reqMsg.getArchivalId());
assertEquals("http://foo.com/", reqMsg.getTargetUrl());
assertEquals(m_testBytes, reqMsg.getPollerNonce());
assertEquals(m_testBytes, reqMsg.getVoterNonce());
assertEquals(null, reqMsg.getVoterNonce2());
List aBlocks = new ArrayList();
List bBlocks = new ArrayList();
for (VoteBlocksIterator iter = m_testMsg.getVoteBlockIterator(); iter.hasNext(); ) {
aBlocks.add(iter.next());
}
for (VoteBlocksIterator iter = reqMsg.getVoteBlockIterator(); iter.hasNext(); ) {
bBlocks.add(iter.next());
}
assertEquals(aBlocks, bBlocks);
// Actual size of test vote blocks is unpredictable
assertTrue(reqMsg.getEstimatedEncodedLength() > V3LcapMessage.EST_ENCODED_HEADER_LENGTH);
}
private void outputSummary(String baseUrl, Set fetched, Set toCrawl, long elapsedTime) {
int fetchCount = fetched.size();
outputMessage(
"\n\nSummary for starting Url: " + baseUrl + " and depth: " + m_crawlDepth,
TEST_SUMMARY_MESSAGE);
outputMessage(
"\nUrls fetched: " + fetchCount + " Urls extracted: " + m_extracted.size(),
PLAIN_MESSAGE);
outputMessage("\nDepth Fetched Parsed New URLs", PLAIN_MESSAGE);
for (int depth = 1; depth <= m_crawlDepth; depth++) {
PrintfFormat pf = new PrintfFormat("%5d %7d %6d %8d");
Integer[] args =
new Integer[] {
new Integer(depth),
new Integer(depth_fetched[depth - 1]),
new Integer(depth_parsed[depth - 1]),
new Integer(depth_incl[depth - 1]),
};
String s = pf.sprintf(args);
outputMessage(s, PLAIN_MESSAGE);
}
outputMessage("\nRemaining unfetched: " + toCrawl.size(), PLAIN_MESSAGE);
if (false) {
for (Iterator iter = toCrawl.iterator(); iter.hasNext(); ) {
String url = (String) iter.next();
outputMessage(url, PLAIN_MESSAGE);
}
}
long secs = elapsedTime / Constants.SECOND;
long fetchRate = 0;
if (secs > 0) {
fetchRate = fetchCount * 60 * Constants.SECOND / elapsedTime;
}
outputMessage(
"\nElapsed Time: " + secs + " secs." + " Fetch Rate: " + fetchRate + " p/m",
PLAIN_MESSAGE);
}
public void testFunctionalFromTarHierarchy() throws Exception {
log.debug3("in testFromTarHierarchy");
// load the tarballs
InputStream file_input = null;
try {
file_input = getResourceAsStream(realTARFile_A);
// UrlCacher uc = au.makeUrlCacher(TAR_A_BASE);
// uc.storeContent(file_input, tarHeader);
UrlCacher uc = tarAu.makeUrlCacher(new UrlData(file_input, tarHeader, TAR_A_BASE));
uc.storeContent();
IOUtil.safeClose(file_input);
file_input = getResourceAsStream(realTARFile_B);
// uc = au.makeUrlCacher(TAR_B_BASE);
// uc.storeContent(file_input, tarHeader);
uc = tarAu.makeUrlCacher(new UrlData(file_input, tarHeader, TAR_B_BASE));
uc.storeContent();
IOUtil.safeClose(file_input);
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
} finally {
IOUtil.safeClose(file_input);
}
CachedUrlSet cus = tarAu.getAuCachedUrlSet();
for (CachedUrl cu : cus.getCuIterable()) {
log.debug3("AU - cu is: " + cu.getUrl());
cu.release();
}
// We need to start from the level of the ArticleMetadataExtractor
MyListEmitter emitter = new MyListEmitter();
ArticleMetadataExtractor amEx =
new ElsevierDeferredArticleMetadataExtractor(ArticleFiles.ROLE_ARTICLE_METADATA);
Iterator<ArticleFiles> it = tarAu.getArticleIterator(MetadataTarget.Any());
while (it.hasNext()) {
ArticleFiles af = it.next();
log.debug3("Metadata test - articlefiles " + af.toString());
// CachedUrl cu = af.getFullTextCu();
CachedUrl cu = af.getRoleCu(ArticleFiles.ROLE_ARTICLE_METADATA);
log.debug3("metadata cu is " + cu.getUrl());
// List<ArticleMetadata> mdlist = mle.extract(MetadataTarget.Any(), cu);
amEx.extract(MetadataTarget.Any(), af, emitter);
List<ArticleMetadata> returnList = emitter.getAmList();
assertNotNull(returnList);
log.debug3("size of returnList is " + returnList.size());
Iterator<ArticleMetadata> mdIt = returnList.iterator();
ArticleMetadata mdRecord = null;
while (mdIt.hasNext()) {
mdRecord = (ArticleMetadata) mdIt.next();
validateCompleteMetadataRecord(mdRecord);
}
}
}
protected CIProperties makeCIProperties(ArchiveRecordHeader elementHeader) throws IOException {
CIProperties ret = new CIProperties();
Set elementHeaderFieldKeys = elementHeader.getHeaderFieldKeys();
for (Iterator i = elementHeaderFieldKeys.iterator(); i.hasNext(); ) {
String key = (String) i.next();
try {
Object valueObject = elementHeader.getHeaderValue(key);
if (valueObject == null) {
logger.warning("Ignoring null value for key '" + key + "'.");
} else {
String value = valueObject.toString();
logger.debug3(key + ": " + value);
ret.put(key, value);
}
} catch (ClassCastException ex) {
logger.error("makeCIProperties: " + key + " threw ", ex);
throw new CacheException.ExploderException(ex);
}
}
return (ret);
}
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);
}
private Set outputUrlResults(String url, Set m_inclset, Set m_exclset) {
Set new_incls = new TreeSet(CollectionUtils.subtract(m_inclset, m_reported));
Set new_excls = new TreeSet(CollectionUtils.subtract(m_exclset, m_reported));
if (!m_inclset.isEmpty()) {
outputMessage(
"\nIncluded Urls: ("
+ new_incls.size()
+ " new, "
+ (m_inclset.size() - new_incls.size())
+ " old)",
URL_SUMMARY_MESSAGE);
depth_incl[m_curDepth - 1] += new_incls.size();
}
for (Iterator it = new_incls.iterator(); it.hasNext(); ) {
outputMessage(it.next().toString(), PLAIN_MESSAGE);
}
if (!m_exclset.isEmpty()) {
outputMessage(
"\nExcluded Urls: ("
+ new_excls.size()
+ " new, "
+ (m_exclset.size() - new_excls.size())
+ " old)",
URL_SUMMARY_MESSAGE);
}
for (Iterator it = new_excls.iterator(); it.hasNext(); ) {
outputMessage(it.next().toString(), PLAIN_MESSAGE);
}
m_reported.addAll(new_incls);
m_reported.addAll(new_excls);
if (m_outWriter != null) {
try {
m_outWriter.flush();
} catch (IOException ex) {
}
}
return new_incls;
}
public void setConfig(
Configuration config, Configuration oldConfig, Configuration.Differences changedKeys) {
// Build list of repositories from list of disk (fs) paths). Needs to
// be generalized if ever another repository implementation.
if (changedKeys.contains(ConfigManager.PARAM_PLATFORM_DISK_SPACE_LIST)) {
List lst = new ArrayList();
String dspace = config.get(ConfigManager.PARAM_PLATFORM_DISK_SPACE_LIST, "");
List paths = StringUtil.breakAt(dspace, ';');
if (paths != null) {
for (Iterator iter = paths.iterator(); iter.hasNext(); ) {
lst.add("local:" + (String) iter.next());
}
}
repoList = lst;
}
if (changedKeys.contains(PARAM_MAX_PER_AU_CACHE_SIZE)) {
paramNodeCacheSize =
config.getInt(PARAM_MAX_PER_AU_CACHE_SIZE, DEFAULT_MAX_PER_AU_CACHE_SIZE);
for (Iterator iter = getDaemon().getAllLockssRepositories().iterator(); iter.hasNext(); ) {
LockssRepository repo = (LockssRepository) iter.next();
if (repo instanceof LockssRepositoryImpl) {
LockssRepositoryImpl repoImpl = (LockssRepositoryImpl) repo;
repoImpl.setNodeCacheSize(paramNodeCacheSize);
}
}
}
if (changedKeys.contains(PARAM_MAX_SUSPECT_VERSIONS_CACHE_SIZE)) {
paramSuspectVersionsCacheSize =
config.getInt(
PARAM_MAX_SUSPECT_VERSIONS_CACHE_SIZE, DEFAULT_MAX_SUSPECT_VERSIONS_CACHE_SIZE);
suspectVersionsCache.setMaxSize(paramSuspectVersionsCacheSize);
}
if (changedKeys.contains(GLOBAL_CACHE_PREFIX)) {
paramIsGlobalNodeCache =
config.getBoolean(PARAM_GLOBAL_CACHE_ENABLED, DEFAULT_GLOBAL_CACHE_ENABLED);
if (paramIsGlobalNodeCache) {
paramGlobalNodeCacheSize =
config.getInt(PARAM_MAX_GLOBAL_CACHE_SIZE, DEFAULT_MAX_GLOBAL_CACHE_SIZE);
log.debug("global node cache size: " + paramGlobalNodeCacheSize);
globalNodeCache.setMaxSize(paramGlobalNodeCacheSize);
}
}
if (changedKeys.contains(DISK_PREFIX)) {
int minMB = config.getInt(PARAM_DISK_WARN_FRRE_MB, DEFAULT_DISK_WARN_FRRE_MB);
double minPer =
config.getPercentage(PARAM_DISK_WARN_FRRE_PERCENT, DEFAULT_DISK_WARN_FRRE_PERCENT);
paramDFWarn = PlatformUtil.DF.makeThreshold(minMB, minPer);
minMB = config.getInt(PARAM_DISK_FULL_FRRE_MB, DEFAULT_DISK_FULL_FRRE_MB);
minPer = config.getPercentage(PARAM_DISK_FULL_FRRE_PERCENT, DEFAULT_DISK_FULL_FRRE_PERCENT);
paramDFFull = PlatformUtil.DF.makeThreshold(minMB, minPer);
}
if (changedKeys.contains(PARAM_SIZE_CALC_MAX_LOAD)) {
sizeCalcMaxLoad = config.getPercentage(PARAM_SIZE_CALC_MAX_LOAD, DEFAULT_SIZE_CALC_MAX_LOAD);
}
if (changedKeys.contains(PREFIX)) {
maxUnusedDirSearch =
config.getInt(PARAM_MAX_UNUSED_DIR_SEARCH, DEFAULT_MAX_UNUSED_DIR_SEARCH);
isStatefulUnusedDirSearch =
config.getBoolean(
PARAM_IS_STATEFUL_UNUSED_DIR_SEARCH, DEFAULT_IS_STATEFUL_UNUSED_DIR_SEARCH);
enableLongComponents =
config.getBoolean(PARAM_ENABLE_LONG_COMPONENTS, DEFAULT_ENABLE_LONG_COMPONENTS);
enableLongComponentsCompatibility =
config.getBoolean(
PARAM_ENABLE_LONG_COMPONENTS_COMPATIBILITY,
DEFAULT_ENABLE_LONG_COMPONENTS_COMPATIBILITY);
maxComponentLength = config.getInt(PARAM_MAX_COMPONENT_LENGTH, DEFAULT_MAX_COMPONENT_LENGTH);
checkUnnormalized =
(CheckUnnormalizedMode)
config.getEnum(
CheckUnnormalizedMode.class,
PARAM_CHECK_UNNORMALIZED,
DEFAULT_CHECK_UNNORMALIZED);
}
}
/** 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);
}
}