mirror of https://github.com/kiwix/libkiwix.git
+ new indexer code
This commit is contained in:
kelson42
parent
735d9afd3a
commit
62daa9ffe5
|
|
@ -35,20 +35,9 @@ namespace kiwix {
|
|||
}
|
||||
|
||||
/* Constructor */
|
||||
Indexer::Indexer(const string &zimFilePath)
|
||||
: zimFileHandler(NULL),
|
||||
articleCount(0),
|
||||
stepSize(0),
|
||||
keywordsBoostFactor(3) {
|
||||
|
||||
this->initialize();
|
||||
this->setZimFilePath(zimFilePath);
|
||||
Indexer::Indexer() :
|
||||
keywordsBoostFactor(3) {
|
||||
|
||||
/* Read the stopwords file */
|
||||
//this->readStopWordsFile("/home/kelson/kiwix/moulinkiwix/stopwords/fr");
|
||||
}
|
||||
|
||||
void Indexer::initialize() {
|
||||
/* Initialize mutex */
|
||||
pthread_mutex_init(&threadIdsMutex, NULL);
|
||||
pthread_mutex_init(&toParseQueueMutex, NULL);
|
||||
|
|
@ -57,44 +46,34 @@ namespace kiwix {
|
|||
pthread_mutex_init(&articleParserRunningMutex, NULL);
|
||||
pthread_mutex_init(&articleIndexerRunningMutex, NULL);
|
||||
pthread_mutex_init(&articleCountMutex, NULL);
|
||||
pthread_mutex_init(&zimPathMutex, NULL);
|
||||
pthread_mutex_init(&indexPathMutex, NULL);
|
||||
pthread_mutex_init(&progressionMutex, NULL);
|
||||
|
||||
/* Article count & Progression */
|
||||
this->setProgression(0);
|
||||
}
|
||||
|
||||
bool Indexer::setZimFilePath(const string &zimFilePath) {
|
||||
/* Open the ZIM file */
|
||||
this->zimFileHandler = new zim::File(zimFilePath);
|
||||
|
||||
/* Define a few values */
|
||||
this->firstArticleOffset = this->zimFileHandler->getNamespaceBeginOffset('A');
|
||||
this->lastArticleOffset = this->zimFileHandler->getNamespaceEndOffset('A');
|
||||
this->currentArticleOffset = this->firstArticleOffset;
|
||||
|
||||
/* Compute few things */
|
||||
kiwix::Reader reader(zimFilePath);
|
||||
this->setArticleCount(reader.getArticleCount());
|
||||
//this->articleCount = this->zimFileHandler->getNamespaceCount('A');
|
||||
this->stepSize = (float)this->articleCount / (float)100;
|
||||
/* Read the stopwords file */
|
||||
//this->readStopWordsFile("/home/kelson/kiwix/moulinkiwix/stopwords/fr");
|
||||
}
|
||||
|
||||
/* Article extractor methods */
|
||||
void *Indexer::extractArticles(void *ptr) {
|
||||
kiwix::Indexer *self = (kiwix::Indexer *)ptr;
|
||||
self->articleExtractorRunning(true);
|
||||
unsigned int startOffset = self->zimFileHandler->getNamespaceBeginOffset('A');
|
||||
unsigned int endOffset = self->zimFileHandler->getNamespaceEndOffset('A');
|
||||
|
||||
/* Get the number of article to index */
|
||||
kiwix::Reader reader(self->getZimPath());
|
||||
self->setArticleCount(reader.getArticleCount());
|
||||
|
||||
/* Goes trough all articles */
|
||||
unsigned int currentOffset = startOffset;
|
||||
zim::File *zimHandler = reader.getZimFileHandler();
|
||||
unsigned int currentOffset = zimHandler->getNamespaceBeginOffset('A');;
|
||||
unsigned int lastOffset = zimHandler->getNamespaceEndOffset('A');;
|
||||
zim::Article currentArticle;
|
||||
|
||||
while (currentOffset <= endOffset) {
|
||||
while (currentOffset <= lastOffset) {
|
||||
/* Redirects are not indexed */
|
||||
do {
|
||||
currentArticle = self->zimFileHandler->getArticle(currentOffset++);
|
||||
} while (currentArticle.isRedirect() && currentOffset != endOffset);
|
||||
currentArticle = zimHandler->getArticle(currentOffset++);
|
||||
} while (currentArticle.isRedirect() && currentOffset != lastOffset);
|
||||
|
||||
/* Add articles to the queue */
|
||||
indexerToken token;
|
||||
|
|
@ -176,7 +155,8 @@ namespace kiwix {
|
|||
self->pushToIndexQueue(token);
|
||||
|
||||
/* Test if the thread should be cancelled */
|
||||
pthread_testcancel(); }
|
||||
pthread_testcancel();
|
||||
}
|
||||
}
|
||||
|
||||
self->articleParserRunning(false);
|
||||
|
|
@ -201,29 +181,36 @@ namespace kiwix {
|
|||
void *Indexer::indexArticles(void *ptr) {
|
||||
kiwix::Indexer *self = (kiwix::Indexer *)ptr;
|
||||
self->articleIndexerRunning(true);
|
||||
|
||||
indexerToken token;
|
||||
unsigned int stepSize = ((self->getArticleCount() / 100) < 1 ? 1 : (self->getArticleCount() / 100));
|
||||
unsigned indexedArticleCount = 0;
|
||||
unsigned int stepSize = ((self->getArticleCount() / 100) < 1 ? 1 : (self->getArticleCount() / 100));
|
||||
self->indexingPrelude(self->getIndexPath());
|
||||
|
||||
while (self->popFromToIndexQueue(token)) {
|
||||
self->indexNextArticle(token.url,
|
||||
token.accentedTitle,
|
||||
token.title,
|
||||
token.keywords,
|
||||
token.content,
|
||||
token.snippet,
|
||||
token.size,
|
||||
token.wordCount
|
||||
);
|
||||
|
||||
self->index(token.url,
|
||||
token.accentedTitle,
|
||||
token.title,
|
||||
token.keywords,
|
||||
token.content,
|
||||
token.snippet,
|
||||
token.size,
|
||||
token.wordCount
|
||||
);
|
||||
|
||||
if (++indexedArticleCount % stepSize == 0) {
|
||||
self->setProgression(self->getProgression() + 1);
|
||||
}
|
||||
|
||||
if (indexedArticleCount % 10000 == 0) {
|
||||
self->flush();
|
||||
}
|
||||
|
||||
/* Test if the thread should be cancelled */
|
||||
pthread_testcancel();
|
||||
}
|
||||
|
||||
self->setProgression(100);
|
||||
self->indexNextPercentPost();
|
||||
|
||||
self->indexingPostlude();
|
||||
self->articleIndexerRunning(false);
|
||||
pthread_exit(NULL);
|
||||
return NULL;
|
||||
|
|
@ -306,8 +293,34 @@ namespace kiwix {
|
|||
return true;
|
||||
}
|
||||
|
||||
/* Article Count & Progression */
|
||||
void Indexer::setArticleCount(unsigned int articleCount) {
|
||||
/* ZIM & Index methods */
|
||||
void Indexer::setZimPath(const string path) {
|
||||
pthread_mutex_lock(&zimPathMutex);
|
||||
this->zimPath = path;
|
||||
pthread_mutex_unlock(&zimPathMutex);
|
||||
}
|
||||
|
||||
string Indexer::getZimPath() {
|
||||
pthread_mutex_lock(&zimPathMutex);
|
||||
string retVal = this->zimPath;
|
||||
pthread_mutex_unlock(&zimPathMutex);
|
||||
return retVal;
|
||||
}
|
||||
|
||||
void Indexer::setIndexPath(const string path) {
|
||||
pthread_mutex_lock(&indexPathMutex);
|
||||
this->indexPath = path;
|
||||
pthread_mutex_unlock(&indexPathMutex);
|
||||
}
|
||||
|
||||
string Indexer::getIndexPath() {
|
||||
pthread_mutex_lock(&indexPathMutex);
|
||||
string retVal = this->indexPath;
|
||||
pthread_mutex_unlock(&indexPathMutex);
|
||||
return retVal;
|
||||
}
|
||||
|
||||
void Indexer::setArticleCount(const unsigned int articleCount) {
|
||||
pthread_mutex_lock(&articleCountMutex);
|
||||
this->articleCount = articleCount;
|
||||
pthread_mutex_unlock(&articleCountMutex);
|
||||
|
|
@ -320,7 +333,7 @@ namespace kiwix {
|
|||
return retVal;
|
||||
}
|
||||
|
||||
void Indexer::setProgression(unsigned int progression) {
|
||||
void Indexer::setProgression(const unsigned int progression) {
|
||||
pthread_mutex_lock(&progressionMutex);
|
||||
this->progression = progression;
|
||||
pthread_mutex_unlock(&progressionMutex);
|
||||
|
|
@ -333,8 +346,12 @@ namespace kiwix {
|
|||
return retVal;
|
||||
}
|
||||
|
||||
bool Indexer::start() {
|
||||
this->indexNextPercentPre();
|
||||
/* Manage */
|
||||
bool Indexer::start(const string &zimPath, const string &indexPath) {
|
||||
this->setProgression(0);
|
||||
this->setZimPath(zimPath);
|
||||
this->setIndexPath(indexPath);
|
||||
|
||||
pthread_mutex_lock(&threadIdsMutex);
|
||||
pthread_create(&(this->articleExtractor), NULL, Indexer::extractArticles, (void*)this);
|
||||
pthread_detach(this->articleExtractor);
|
||||
|
|
@ -343,15 +360,7 @@ namespace kiwix {
|
|||
pthread_create(&(this->articleIndexer), NULL, Indexer::indexArticles, (void*)this);
|
||||
pthread_detach(this->articleIndexer);
|
||||
pthread_mutex_unlock(&threadIdsMutex);
|
||||
return true;
|
||||
}
|
||||
|
||||
bool Indexer::stop() {
|
||||
pthread_mutex_lock(&threadIdsMutex);
|
||||
pthread_cancel(this->articleExtractor);
|
||||
pthread_cancel(this->articleParser);
|
||||
pthread_cancel(this->articleIndexer);
|
||||
pthread_mutex_unlock(&threadIdsMutex);
|
||||
return true;
|
||||
}
|
||||
|
||||
|
|
@ -359,12 +368,27 @@ namespace kiwix {
|
|||
return this->isArticleExtractorRunning() || this->isArticleIndexerRunning() || this->isArticleParserRunning();
|
||||
}
|
||||
|
||||
void Indexer::setCurrentArticleOffset(unsigned int offset) {
|
||||
this->currentArticleOffset = offset;
|
||||
}
|
||||
bool Indexer::stop() {
|
||||
if (this->isRunning()) {
|
||||
bool isArticleExtractorRunning = this->isArticleExtractorRunning();
|
||||
bool isArticleIndexerRunning = this->isArticleIndexerRunning();
|
||||
bool isArticleParserRunning = this->isArticleParserRunning();
|
||||
|
||||
pthread_mutex_lock(&threadIdsMutex);
|
||||
|
||||
if (isArticleExtractorRunning)
|
||||
pthread_cancel(this->articleExtractor);
|
||||
if (isArticleIndexerRunning)
|
||||
pthread_cancel(this->articleParser);
|
||||
if (isArticleParserRunning)
|
||||
pthread_cancel(this->articleIndexer);
|
||||
|
||||
pthread_mutex_unlock(&threadIdsMutex);
|
||||
|
||||
this->articleIndexerRunning(false);
|
||||
}
|
||||
|
||||
unsigned int Indexer::getCurrentArticleOffset() {
|
||||
return this->currentArticleOffset;
|
||||
return true;
|
||||
}
|
||||
|
||||
/* Read the file containing the stopwords */
|
||||
|
|
@ -382,102 +406,4 @@ namespace kiwix {
|
|||
return true;
|
||||
}
|
||||
|
||||
/* Index next percent */
|
||||
bool Indexer::indexNextPercent(const bool &verbose) {
|
||||
float thresholdOffset = this->currentArticleOffset + this->stepSize;
|
||||
size_t found;
|
||||
|
||||
/* Check if we can start */
|
||||
if (this->zimFileHandler == NULL) {
|
||||
return false;
|
||||
}
|
||||
|
||||
this->indexNextPercentPre();
|
||||
|
||||
while(this->currentArticleOffset < thresholdOffset &&
|
||||
this->currentArticleOffset <= this->lastArticleOffset) {
|
||||
|
||||
zim::Article currentArticle;
|
||||
|
||||
/* Get next non redirect article */
|
||||
do {
|
||||
currentArticle = this->zimFileHandler->getArticle(this->currentArticleOffset);
|
||||
} while (this->currentArticleOffset++ &&
|
||||
currentArticle.isRedirect() &&
|
||||
this->currentArticleOffset != this->lastArticleOffset);
|
||||
|
||||
if (!currentArticle.isRedirect()) {
|
||||
|
||||
/* Index the content */
|
||||
this->htmlParser.reset();
|
||||
string content (currentArticle.getData().data(), currentArticle.getData().size());
|
||||
|
||||
/* The parser generate a lot of exceptions which should be avoided */
|
||||
try {
|
||||
this->htmlParser.parse_html(content, "UTF-8", true);
|
||||
} catch (...) {
|
||||
}
|
||||
|
||||
/* If content does not have the noindex meta tag */
|
||||
/* Seems that the parser generates an exception in such case */
|
||||
found = this->htmlParser.dump.find("NOINDEX");
|
||||
|
||||
if (found == string::npos) {
|
||||
string url = currentArticle.getLongUrl();
|
||||
|
||||
/* Debug output */
|
||||
if (verbose) {
|
||||
std::cout << "Indexing " << url << "..." << std::endl;
|
||||
}
|
||||
|
||||
/* Get the title */
|
||||
string accentedTitle = this->htmlParser.title;
|
||||
if (accentedTitle.empty()) {
|
||||
accentedTitle = currentArticle.getTitle();
|
||||
}
|
||||
|
||||
/* count words */
|
||||
stringstream countWordStringStream;
|
||||
countWordStringStream << countWords(this->htmlParser.dump);
|
||||
const std::string wordCountString = countWordStringStream.str();
|
||||
|
||||
/* snippet */
|
||||
std::string snippet = std::string(this->htmlParser.dump, 0, 300);
|
||||
std::string::size_type last = snippet.find_last_of('.');
|
||||
if (last == snippet.npos)
|
||||
last = snippet.find_last_of(' ');
|
||||
if (last != snippet.npos)
|
||||
snippet = snippet.substr(0, last);
|
||||
|
||||
/* size */
|
||||
stringstream sizeStringStream;
|
||||
sizeStringStream << content.size() / 1024;
|
||||
const std::string size = sizeStringStream.str();
|
||||
|
||||
this->indexNextArticle(url,
|
||||
accentedTitle,
|
||||
removeAccents(this->htmlParser.title),
|
||||
removeAccents(this->htmlParser.keywords),
|
||||
removeAccents(this->htmlParser.dump),
|
||||
snippet,
|
||||
size,
|
||||
wordCountString
|
||||
);
|
||||
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
this->indexNextPercentPost();
|
||||
|
||||
/* increment the offset and set returned value */
|
||||
if (this->currentArticleOffset <= this->lastArticleOffset) {
|
||||
return true;
|
||||
} else {
|
||||
// commented as it never returns on OSX.
|
||||
//this->stopIndexing();
|
||||
return false;
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
|
|
|||
|
|
@ -54,17 +54,14 @@ namespace kiwix {
|
|||
class Indexer {
|
||||
|
||||
public:
|
||||
Indexer(const string &zimFilePath);
|
||||
bool indexNextPercent(const bool &verbose = false);
|
||||
bool setZimFilePath(const string &zimFilePath);
|
||||
bool start();
|
||||
Indexer();
|
||||
bool start(const string &zimPath, const string &indexPath);
|
||||
bool stop();
|
||||
bool isRunning();
|
||||
unsigned int getProgression();
|
||||
|
||||
private:
|
||||
pthread_mutex_t threadIdsMutex;
|
||||
void initialize();
|
||||
|
||||
/* Article extraction */
|
||||
pthread_t articleExtractor;
|
||||
|
|
@ -107,46 +104,47 @@ namespace kiwix {
|
|||
/* Article Count & Progression */
|
||||
unsigned int articleCount;
|
||||
pthread_mutex_t articleCountMutex;
|
||||
void setArticleCount(unsigned int articleCount);
|
||||
void setArticleCount(const unsigned int articleCount);
|
||||
unsigned int getArticleCount();
|
||||
|
||||
/* Progression */
|
||||
unsigned int progression;
|
||||
pthread_mutex_t progressionMutex;
|
||||
void setProgression(unsigned int progression);
|
||||
void setProgression(const unsigned int progression);
|
||||
/* getProgression() is public */
|
||||
|
||||
/* ZIM path */
|
||||
pthread_mutex_t zimPathMutex;
|
||||
string zimPath;
|
||||
void setZimPath(const string path);
|
||||
string getZimPath();
|
||||
|
||||
/* Index path */
|
||||
pthread_mutex_t indexPathMutex;
|
||||
string indexPath;
|
||||
void setIndexPath(const string path);
|
||||
string getIndexPath();
|
||||
|
||||
protected:
|
||||
virtual void indexNextPercentPre() = 0;
|
||||
virtual void indexNextArticle(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount) = 0;
|
||||
virtual void indexNextPercentPost() = 0;
|
||||
virtual void stopIndexing() = 0;
|
||||
|
||||
/* Article offset */
|
||||
void setCurrentArticleOffset(unsigned int offset);
|
||||
unsigned int getCurrentArticleOffset();
|
||||
|
||||
/* ZIM file handling */
|
||||
zim::File* zimFileHandler;
|
||||
zim::size_type firstArticleOffset;
|
||||
zim::size_type lastArticleOffset;
|
||||
zim::size_type currentArticleOffset;
|
||||
virtual void indexingPrelude(const string &indexPath) = 0;
|
||||
virtual void index(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount) = 0;
|
||||
virtual void flush() = 0;
|
||||
virtual void indexingPostlude() = 0;
|
||||
|
||||
/* HTML parsing */
|
||||
MyHtmlParser htmlParser;
|
||||
/* Others */
|
||||
unsigned int countWords(const string &text);
|
||||
|
||||
/* Stopwords */
|
||||
bool readStopWordsFile(const string path);
|
||||
std::vector<std::string> stopWords;
|
||||
|
||||
/* Others */
|
||||
float stepSize;
|
||||
|
||||
/* Boost factor */
|
||||
unsigned int keywordsBoostFactor;
|
||||
inline unsigned int getTitleBoostFactor(const unsigned int contentLength) {
|
||||
|
|
|
|||
|
|
@ -62,6 +62,10 @@ namespace kiwix {
|
|||
delete this->zimFileHandler;
|
||||
}
|
||||
}
|
||||
|
||||
zim::File* Reader::getZimFileHandler() {
|
||||
return this->zimFileHandler;
|
||||
}
|
||||
|
||||
/* Reset the cursor for GetNextArticle() */
|
||||
void Reader::reset() {
|
||||
|
|
|
|||
|
|
@ -63,6 +63,7 @@ namespace kiwix {
|
|||
bool canCheckIntegrity();
|
||||
bool isCorrupted();
|
||||
bool parseUrl(const string &urlStr, char *ns, string &titleStr);
|
||||
zim::File* getZimFileHandler();
|
||||
|
||||
protected:
|
||||
zim::File* zimFileHandler;
|
||||
|
|
|
|||
|
|
@ -22,20 +22,15 @@
|
|||
namespace kiwix {
|
||||
|
||||
/* Constructor */
|
||||
XapianIndexer::XapianIndexer(const string &zimFilePath, const string &xapianDirectoryPath) :
|
||||
Indexer(zimFilePath) {
|
||||
|
||||
/* Open the Xapian directory */
|
||||
this->writableDatabase = new Xapian::WritableDatabase(xapianDirectoryPath,
|
||||
Xapian::DB_CREATE_OR_OVERWRITE);
|
||||
|
||||
XapianIndexer::XapianIndexer() {
|
||||
/* Stemming */
|
||||
/*
|
||||
stemmer = Xapian::Stem("french");
|
||||
indexer.set_stemmer(stemmer);
|
||||
*/
|
||||
|
||||
/* Stop words
|
||||
/* Stop words */
|
||||
/*
|
||||
std::vector<std::string>::const_iterator stopWordsIterator = this->stopWords.begin();
|
||||
this->stopper.add("ceci");
|
||||
while (stopWordsIterator != this->stopWords.end()) {
|
||||
|
|
@ -46,19 +41,20 @@ namespace kiwix {
|
|||
*/
|
||||
}
|
||||
|
||||
void XapianIndexer::indexNextPercentPre() {
|
||||
this->writableDatabase->begin_transaction(true);
|
||||
void XapianIndexer::indexingPrelude(const string &indexPath) {
|
||||
this->writableDatabase = Xapian::WritableDatabase(indexPath, Xapian::DB_CREATE_OR_OVERWRITE);
|
||||
this->writableDatabase.begin_transaction(true);
|
||||
}
|
||||
|
||||
void XapianIndexer::indexNextArticle(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount) {
|
||||
|
||||
void XapianIndexer::index(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount) {
|
||||
|
||||
/* Put the data in the document */
|
||||
Xapian::Document currentDocument;
|
||||
currentDocument.clear_values();
|
||||
|
|
@ -85,26 +81,17 @@ namespace kiwix {
|
|||
}
|
||||
|
||||
/* add to the database */
|
||||
this->writableDatabase->add_document(currentDocument);
|
||||
}
|
||||
|
||||
void XapianIndexer::indexNextPercentPost() {
|
||||
/* Flush and close Xapian transaction*/
|
||||
this->writableDatabase->commit_transaction();
|
||||
this->writableDatabase.add_document(currentDocument);
|
||||
}
|
||||
|
||||
/* Stop indexing. TODO: using it crashs the soft under windows. Have to do it in indexNextPercent() */
|
||||
void XapianIndexer::stopIndexing() {
|
||||
/* Delete the zimFileHandler */
|
||||
if (this->zimFileHandler != NULL) {
|
||||
delete this->zimFileHandler;
|
||||
this->zimFileHandler = NULL;
|
||||
}
|
||||
|
||||
/* Delete the Xapian writableDatabase */
|
||||
if (this->writableDatabase != NULL) {
|
||||
delete this->writableDatabase;
|
||||
this->writableDatabase = NULL;
|
||||
}
|
||||
void XapianIndexer::flush() {
|
||||
this->writableDatabase.commit_transaction();
|
||||
this->writableDatabase.begin_transaction(true);
|
||||
}
|
||||
|
||||
void XapianIndexer::indexingPostlude() {
|
||||
this->flush();
|
||||
this->writableDatabase.commit_transaction();
|
||||
this->writableDatabase.commit();
|
||||
}
|
||||
}
|
||||
|
|
|
|||
|
|
@ -30,22 +30,23 @@ namespace kiwix {
|
|||
class XapianIndexer : public Indexer {
|
||||
|
||||
public:
|
||||
XapianIndexer(const string &zimFilePath, const string &xapianDirectoryPath);
|
||||
XapianIndexer();
|
||||
~XapianIndexer();
|
||||
|
||||
protected:
|
||||
void indexNextPercentPre();
|
||||
void indexNextArticle(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount);
|
||||
void indexNextPercentPost();
|
||||
void stopIndexing();
|
||||
void indexingPrelude(const string &indexPath);
|
||||
void index(const string &url,
|
||||
const string &title,
|
||||
const string &unaccentedTitle,
|
||||
const string &keywords,
|
||||
const string &content,
|
||||
const string &snippet,
|
||||
const string &size,
|
||||
const string &wordCount);
|
||||
void flush();
|
||||
void indexingPostlude();
|
||||
|
||||
Xapian::WritableDatabase *writableDatabase;
|
||||
Xapian::WritableDatabase writableDatabase;
|
||||
Xapian::Stem stemmer;
|
||||
Xapian::SimpleStopper stopper;
|
||||
Xapian::TermGenerator indexer;
|
||||
|
|
|
|||
Loading…
Reference in New Issue