static void generateSummary( Summary &summary, char *htmlInput, const char *queryStr, const char *urlStr ) {
Xml xml;
ASSERT_TRUE(xml.set(htmlInput, strlen(htmlInput), 0, CT_HTML));
Words words;
ASSERT_TRUE(words.set(&xml, true));
Bits bits;
ASSERT_TRUE(bits.set(&words));
Url url;
url.set(urlStr);
Sections sections;
ASSERT_TRUE(sections.set(&words, &bits, &url, "", CT_HTML));
Query query;
ASSERT_TRUE(query.set2(queryStr, langEnglish, true));
LinkInfo linkInfo;
memset ( &linkInfo , 0 , sizeof(LinkInfo) );
linkInfo.m_lisize = sizeof(LinkInfo);
Title title;
ASSERT_TRUE(title.setTitle(&xml, &words, 80, &query, &linkInfo, &url, NULL, 0, CT_HTML, langEnglish));
Pos pos;
ASSERT_TRUE(pos.set(&words));
Bits bitsForSummary;
ASSERT_TRUE(bitsForSummary.setForSummary(&words));
Phrases phrases;
ASSERT_TRUE(phrases.set(&words, &bits));
Matches matches;
matches.setQuery(&query);
ASSERT_TRUE(matches.set(&words, &phrases, §ions, &bitsForSummary, &pos, &xml, &title, &url, &linkInfo));
summary.setSummary(&xml, &words, §ions, &pos, &query, 180, 3, 3, 180, &url, &matches, title.getTitle(), title.getTitleLen());
}
/// @brief ClientXMLDocument constructor creates an XML document header and root element
/// and puts them in the output string stream.
ClientXMLDocument::ClientXMLDocument(BOSS::CommandLine & config,
const Phrases & phrases): cl(config)
{
outs << "<?xml version='1.0' encoding='ISO-8859-1' ?>" << endl;
outs << "<TEXT>" << endl;
for (int i = 0; i < phrases.size(); i++)
{
phraseToXML(phrases[i]);
}
outs << "</TEXT>" << endl;
}
// . return length stored into "buf"
// . content must be NULL terminated
// . if "useAnchors" is true we do click and scroll
// . if "isQueryTerms" is true, we do typical anchors in a special way
int32_t Highlight::set ( SafeBuf *sb,
//char *buf ,
//int32_t bufLen ,
char *content ,
int32_t contentLen ,
// primary language of the document (for synonyms)
char docLangId ,
Query *q ,
bool doStemming ,
bool useAnchors ,
const char *baseUrl ,
const char *frontTag ,
const char *backTag ,
int32_t fieldCode ,
int32_t niceness ) {
Words words;
if ( ! words.set ( content ,
contentLen ,
TITLEREC_CURRENT_VERSION,
true , // computeId
true ) ) // has html entites?
return -1;
int32_t version = TITLEREC_CURRENT_VERSION;
Bits bits;
if ( ! bits.set (&words,version,niceness) ) return -1;
Phrases phrases;
if ( !phrases.set(&words,&bits,true,false,version,niceness))return -1;
//SafeBuf langBuf;
//if ( !setLangVec ( &words , &langBuf , niceness )) return 0;
//uint8_t *langVec = (uint8_t *)langBuf.getBufStart();
// make synonyms
//Synonyms syns;
//if(!syns.set(&words,NULL,docLangId,&phrases,niceness,NULL)) return 0;
Matches matches;
matches.setQuery ( q );
if ( ! matches.addMatches ( &words , &phrases ) ) return -1;
// store
m_numMatches = matches.getNumMatches();
return set ( sb ,
//buf ,
//bufLen ,
&words ,
&matches ,
doStemming ,
useAnchors ,
baseUrl ,
frontTag ,
backTag ,
fieldCode ,
q );
}
