C++ SearchEngineInterface::getResults方法代码示例

void QueryingThread::doWork(void)
{
	// Get the SearchEngine
	SearchEngineInterface *engine = SearchEngineFactory::getSearchEngine(m_engineName, m_engineOption);
	if (engine == NULL)
	{
		m_status = _("Couldn't create search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
		return;
	}
	// Set the maximum number of results
	engine->setMaxResultsCount(m_queryProps.getMaximumResultsCount());

	// Run the query
	if (engine->runQuery(m_queryProps) == false)
	{
		m_status = _("Couldn't run query on search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
	}
	else
	{
		const vector<Result> &resultsList = engine->getResults();

		m_resultsList.clear();
		m_resultsList.reserve(resultsList.size());
		m_resultsCharset = engine->getResultsCharset();

		// Copy the results list
		for (vector<Result>::const_iterator resultIter = resultsList.begin();
			resultIter != resultsList.end(); ++resultIter)
		{
			string title = _("No title");
			string extract = HtmlTokenizer::stripTags(resultIter->getExtract());

			// The title may contain formatting
			if (resultIter->getTitle().empty() == false)
			{
				title = HtmlTokenizer::stripTags(resultIter->getTitle());
			}

			string language = resultIter->getLanguage();
			if (language.empty() == true)
			{
				// Use the query's language
				language = m_queryProps.getLanguage();
			}

			m_resultsList.push_back(Result(resultIter->getLocation(),
				title,
				extract,
				language,
				resultIter->getScore()));
		}
	}
	delete engine;
}

bool DBusServletThread::runQuery(QueryProperties &queryProps, vector<string> &docIds)
{
	PinotSettings &settings = PinotSettings::getInstance();

	docIds.clear();

	SearchEngineInterface *pEngine = ModuleFactory::getSearchEngine(settings.m_defaultBackend,
		settings.m_daemonIndexLocation);
	if (pEngine == NULL)
	{
		return false;
	}

	// Run the query
	pEngine->setDefaultOperator(SearchEngineInterface::DEFAULT_OP_AND);
	if (pEngine->runQuery(queryProps) == false)
	{
		delete pEngine;

		return false;
	}

	const vector<DocumentInfo> &resultsList = pEngine->getResults();
	if (resultsList.empty() == true)
	{
#ifdef DEBUG
		cout << "DBusServletThread::runQuery: trying again" << endl;
#endif
		// Try again, this time with OR as default operator
		pEngine->setDefaultOperator(SearchEngineInterface::DEFAULT_OP_OR);
		if (pEngine->runQuery(queryProps) == false)
		{
			delete pEngine;

			return false;
		}
	}

	for (vector<DocumentInfo>::const_iterator resultIter = resultsList.begin();
		resultIter != resultsList.end(); ++resultIter)
	{
		unsigned int indexId = 0;
		unsigned int docId = resultIter->getIsIndexed(indexId);

		// We only need the document ID
		if (docId > 0)
		{
			char docIdStr[64];

			snprintf(docIdStr, 64, "%u", docId);
			docIds.push_back(docIdStr);
		}
	}

	delete pEngine;

	return true;
}

void QueryingThread::doWork(void)
{
	PinotSettings &settings = PinotSettings::getInstance();

	// Get the SearchEngine
	SearchEngineInterface *pEngine = SearchEngineFactory::getSearchEngine(m_engineName, m_engineOption);
	if (pEngine == NULL)
	{
		m_status = _("Couldn't create search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
		return;
	}

	// Set the maximum number of results
	pEngine->setMaxResultsCount(m_queryProps.getMaximumResultsCount());

	// Set up the proxy
	DownloaderInterface *pDownloader = pEngine->getDownloader();
	if ((pDownloader != NULL) &&
		(settings.m_proxyEnabled == true) &&
		(settings.m_proxyAddress.empty() == false))
	{
		char portStr[64];

		pDownloader->setSetting("proxyaddress", settings.m_proxyAddress);
		snprintf(portStr, 64, "%u", settings.m_proxyPort);
		pDownloader->setSetting("proxyport", portStr);
		pDownloader->setSetting("proxytype", settings.m_proxyType);
	}

	// Run the query
	if (pEngine->runQuery(m_queryProps) == false)
	{
		m_status = _("Couldn't run query on search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
	}
	else
	{
		IndexInterface *pDocsIndex = NULL;
		IndexInterface *pDaemonIndex = NULL;
		const vector<Result> &resultsList = pEngine->getResults();
		unsigned int indexId = 0;
		bool isIndexQuery = false;

		m_resultsList.clear();
		m_resultsList.reserve(resultsList.size());
		m_resultsCharset = pEngine->getResultsCharset();

		// Are we querying an index ?
		if (m_engineName == "xapian")
		{
			// Internal index ?
			if (m_engineOption == settings.m_docsIndexLocation)
			{
				indexId = settings.getIndexId(_("My Web Pages"));
				isIndexQuery = true;
			}
			else if (m_engineOption == settings.m_daemonIndexLocation)
			{
				indexId = settings.getIndexId(_("My Documents"));
				isIndexQuery = true;
			}
		}

		// Will we have to query internal indices ?
		if (isIndexQuery == false)
		{
			pDocsIndex = settings.getIndex(settings.m_docsIndexLocation);
			pDaemonIndex = settings.getIndex(settings.m_daemonIndexLocation);
		}

		// Copy the results list
		for (vector<Result>::const_iterator resultIter = resultsList.begin();
			resultIter != resultsList.end(); ++resultIter)
		{
			Result current(*resultIter);
			string title(_("No title"));
			string location(current.getLocation());
			string language(current.getLanguage());
			unsigned int docId = 0;

			// The title may contain formatting
			if (current.getTitle().empty() == false)
			{
				title = FilterUtils::stripMarkup(current.getTitle());
			}
			current.setTitle(title);
#ifdef DEBUG
			cout << "QueryingThread::doWork: title is " << title << endl;
#endif

			// Use the query's language if the result's is unknown
			if (language.empty() == true)
			{
				language = m_queryProps.getLanguage();
			}
			current.setLanguage(language);

//.........这里部分代码省略.........

//.........这里部分代码省略.........
		pDownloader->setSetting("proxytype", proxyType);
	}

	// Set the query
	QueryProperties queryProps("pinot-search", "", queryType);
	if (queryType == QueryProperties::XAPIAN_QP)
	{
		queryProps.setFreeQuery(pQueryInput);
	}
	else
	{
		string fileContents;

		// Load the query from file
		if (loadFile(pQueryInput, fileContents) == false)
		{
			cerr << "Couldn't load query from file " << pQueryInput << endl;

			DownloaderInterface::shutdown();
			MIMEScanner::shutdown();

			return EXIT_FAILURE;
		}

		queryProps.setFreeQuery(fileContents);
	}

	queryProps.setMaximumResultsCount(maxResultsCount);
	if (pEngine->runQuery(queryProps) == true)
	{
		string resultsPage;

		// Try getting a list of links
		const vector<DocumentInfo> resultsList = pEngine->getResults();
		if (resultsList.empty() == false)
		{
			if (printResults == true)
			{
				unsigned int count = 0;

				cout << "Matching documents are :" << endl;

				vector<DocumentInfo>::const_iterator resultIter = resultsList.begin();
				while (resultIter != resultsList.end())
				{
					string rawUrl(resultIter->getLocation());
					Url thisUrl(rawUrl);

					cout << count << " Raw URL  : '" << rawUrl << "'"<< endl;
					cout << count << " Protocol : " << thisUrl.getProtocol() << endl;
					cout << count << " Host     : " << thisUrl.getHost() << endl;
					cout << count << " Location : " << thisUrl.getLocation() << "/" << thisUrl.getFile() << endl;
					cout << count << " Title    : " << resultIter->getTitle() << endl;
					cout << count << " Type     : " << resultIter->getType() << endl;
					cout << count << " Language : " << resultIter->getLanguage() << endl;
					cout << count << " Extract  : " << resultIter->getExtract() << endl;
					cout << count << " Score    : " << resultIter->getScore() << endl;
					count++;

					// Next
					resultIter++;
				}
			}
			else
			{
				string engineName(SearchEngineFactory::getSearchEngineName(engineType, option));

int main(int argc, char **argv)
{
	string type, option;
	bool bDownloadResults = false;

	if (argc < 5)
	{
		cerr << "Usage: " << argv[0] << " <search engine name> <option> <search string> <max results> [DOWNLOAD]" << endl;
		return EXIT_FAILURE;
	}
	if (argc > 5)
	{
		string flag = argv[5];
		if (flag == "DOWNLOAD")
		{
			bDownloadResults = true;
		}
	}

	// Which SearchEngine ?
	type = argv[1];
	option = argv[2];
	SearchEngineInterface *myEngine = SearchEngineFactory::getSearchEngine(type, option);
	if (myEngine == NULL)
	{
		cerr << "Couldn't obtain search engine instance" << endl;
		return EXIT_FAILURE;
	}

	// How many results ?
	unsigned int count = atoi(argv[4]);
	myEngine->setMaxResultsCount(count);

	QueryProperties queryProps("senginetest", argv[3], "", "", "");
	bool bOK = myEngine->runQuery(queryProps);
	if (bOK == true)
	{
		string resultsPage;

		// Try getting a list of links
		const vector<Result> resultsList = myEngine->getResults();
		if (resultsList.empty() == false)
		{
			unsigned int count = 0;

			cout << "Matching documents are :" << endl;

			vector<Result>::const_iterator resultIter = resultsList.begin();
			while (resultIter != resultsList.end())
			{
				string rawUrl = (*resultIter).getLocation();
				Url thisUrl(rawUrl);

				cout << count << " Raw URL  : '" << rawUrl << "'"<< endl;
				cout << count << " Protocol : " << thisUrl.getProtocol() << endl;
				cout << count << " Host     : " << thisUrl.getHost() << endl;
				cout << count << " Location : " << thisUrl.getLocation() << "/" << thisUrl.getFile() << endl;
				cout << count << " Title    : " << HtmlTokenizer::stripTags((*resultIter).getTitle()) << endl;
				cout << count << " Extract  : " << HtmlTokenizer::stripTags((*resultIter).getExtract()) << endl;
				cout << count << " Score    : " << (*resultIter).getScore() << endl;

				if (bDownloadResults == true)
				{
					// Set the name of the file to which this page will be saved
					char num[16];
					sprintf(num, "%d", count);
					string url = (*resultIter).getLocation();
					string file = num;
					file += "_";
					file += thisUrl.getHost();
					file += ".html";

					if (type == "googleapi")
					{
						// Fetch the page from the Google cache
						fetchCachedPage(url, file, option);
					}
					else
					{
						fetchPage(url, file);
					}
				}
				count++;

				// Next
				resultIter++;
			}
		}
		else
		{
			cerr << "Couldn't get a results list !" << endl;
		}
	}
	else
	{
		cerr << "Couldn't run query on search engine " << argv[1] << endl;
	}

	delete myEngine;

//.........这里部分代码省略.........

void EngineQueryThread::doWork(void)
{
	PinotSettings &settings = PinotSettings::getInstance();

	// Get the SearchEngine
	SearchEngineInterface *pEngine = SearchEngineFactory::getSearchEngine(m_engineName, m_engineOption);
	if (pEngine == NULL)
	{
		m_errorNum = UNKNOWN_ENGINE;
		m_errorParam = m_engineDisplayableName;
		return;
	}

	// Set up the proxy
	DownloaderInterface *pDownloader = pEngine->getDownloader();
	if ((pDownloader != NULL) &&
		(settings.m_proxyEnabled == true) &&
		(settings.m_proxyAddress.empty() == false))
	{
		char portStr[64];

		pDownloader->setSetting("proxyaddress", settings.m_proxyAddress);
		snprintf(portStr, 64, "%u", settings.m_proxyPort);
		pDownloader->setSetting("proxyport", portStr);
		pDownloader->setSetting("proxytype", settings.m_proxyType);
	}

	if (m_listingIndex == false)
	{
		pEngine->setLimitSet(m_limitToDocsSet);
	}

	// Run the query
	pEngine->setDefaultOperator(SearchEngineInterface::DEFAULT_OP_AND);
	if (pEngine->runQuery(m_queryProps, m_startDoc) == false)
	{
		m_errorNum = QUERY_FAILED;
		m_errorParam = m_engineDisplayableName;
	}
	else
	{
		const vector<DocumentInfo> &resultsList = pEngine->getResults();

		m_documentsList.clear();
		m_documentsList.reserve(resultsList.size());
		m_documentsCount = pEngine->getResultsCountEstimate();
#ifdef DEBUG
		cout << "EngineQueryThread::doWork: " << resultsList.size() << " off " << m_documentsCount
			<< " results to process, starting at position " << m_startDoc << endl;
#endif

		m_resultsCharset = pEngine->getResultsCharset();
		if (m_listingIndex == false)
		{
			processResults(resultsList);
		}
		else
		{
			processResults(resultsList,
				PinotSettings::getInstance().getIndexIdByName(m_engineDisplayableName));
		}

		// Any spelling correction ?
		string correctedFreeQuery(pEngine->getSpellingCorrection());
		if (correctedFreeQuery.empty() == false)
		{
			m_correctedSpelling = true;
			m_queryProps.setFreeQuery(correctedFreeQuery);
		}
	}

	delete pEngine;
}

//.........这里部分代码省略.........
	{
		clog << "Couldn't obtain search engine instance" << endl;

		DownloaderInterface::shutdown();
		MIMEScanner::shutdown();

		return EXIT_FAILURE;
	}

	// Set up the proxy
	WebEngine *pWebEngine = dynamic_cast<WebEngine *>(pEngine);
	if (pWebEngine != NULL)
	{
		DownloaderInterface *pDownloader = pWebEngine->getDownloader();
		if ((pDownloader != NULL) &&
			(settings.m_proxyEnabled == true) &&
			(settings.m_proxyAddress.empty() == false))
		{
			char portStr[64];

			pDownloader->setSetting("proxyaddress", settings.m_proxyAddress);
			snprintf(portStr, 64, "%u", settings.m_proxyPort);
			pDownloader->setSetting("proxyport", portStr);
			pDownloader->setSetting("proxytype", settings.m_proxyType);
		}

		pWebEngine->setEditableValues(settings.m_editablePluginValues);
	}

	pEngine->setDefaultOperator(SearchEngineInterface::DEFAULT_OP_AND);
	if (pEngine->runQuery(queryProps) == true)
	{
		string resultsPage;
		unsigned int estimatedResultsCount = pEngine->getResultsCountEstimate();

		const vector<DocumentInfo> &resultsList = pEngine->getResults();
		if (resultsList.empty() == false)
		{
			if (printResults == true)
			{
				unsigned int count = 0;

				if (locationOnly == false)
				{
					clog << "Showing " << resultsList.size() << " results of about " << estimatedResultsCount << endl;
				}

				vector<DocumentInfo>::const_iterator resultIter = resultsList.begin();
				while (resultIter != resultsList.end())
				{
					string rawUrl(resultIter->getLocation(true));

					if (locationOnly == false)
					{
						clog << count << " Location : '" << rawUrl << "'"<< endl;
						clog << count << " Title    : " << resultIter->getTitle() << endl;
						clog << count << " Type     : " << resultIter->getType() << endl;
						clog << count << " Language : " << resultIter->getLanguage() << endl;
						clog << count << " Date     : " << resultIter->getTimestamp() << endl;
						clog << count << " Size     : " << resultIter->getSize() << endl;
						clog << count << " Extract  : " << resultIter->getExtract() << endl;
						clog << count << " Score    : " << resultIter->getScore() << endl;
					}
					else
					{
						clog << rawUrl << endl;

void QueryingThread::doWork(void)
{
	PinotSettings &settings = PinotSettings::getInstance();

	// Get the SearchEngine
	SearchEngineInterface *pEngine = SearchEngineFactory::getSearchEngine(m_engineName, m_engineOption);
	if (pEngine == NULL)
	{
		m_status = _("Couldn't create search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
		return;
	}

	// Set up the proxy
	DownloaderInterface *pDownloader = pEngine->getDownloader();
	if ((pDownloader != NULL) &&
		(settings.m_proxyEnabled == true) &&
		(settings.m_proxyAddress.empty() == false))
	{
		char portStr[64];

		pDownloader->setSetting("proxyaddress", settings.m_proxyAddress);
		snprintf(portStr, 64, "%u", settings.m_proxyPort);
		pDownloader->setSetting("proxyport", portStr);
		pDownloader->setSetting("proxytype", settings.m_proxyType);
	}

	// Run the query
	if (pEngine->runQuery(m_queryProps, m_startDoc) == false)
	{
		m_status = _("Couldn't run query on search engine");
		m_status += " ";
		m_status += m_engineDisplayableName;
	}
	else
	{
		const vector<DocumentInfo> &resultsList = pEngine->getResults();

		m_documentsList.clear();
		m_documentsList.reserve(resultsList.size());
		m_documentsCount = pEngine->getResultsCountEstimate();
#ifdef DEBUG
		cout << "QueryingThread::doWork: " << resultsList.size() << " off " << m_documentsCount
			<< " results to process, starting at position " << m_startDoc << endl;
#endif

		m_resultsCharset = pEngine->getResultsCharset();
		if (m_listingIndex == false)
		{
			processResults(resultsList);
		}
		else
		{
			processResults(resultsList,
				PinotSettings::getInstance().getIndexId(m_engineDisplayableName));
		}
	}

	delete pEngine;
}