// . a new interface so Msg3b can call this with "s" set to NULL
// . returns false if blocked, true otherwise
// . sets g_errno on error
bool sendPageParser2 ( TcpSocket *s ,
HttpRequest *r ,
State8 *st ,
long long docId ,
Query *q ,
// in query term space, not imap space
long long *termFreqs ,
// in imap space
float *termFreqWeights ,
// in imap space
float *affWeights ,
void *state ,
void (* callback)(void *state) ) {
//log("parser: read sock=%li",s->m_sd);
// might a simple request to addsomething to validated.*.txt file
// from XmlDoc::print() or XmlDoc::validateOutput()
char *add = r->getString("add",NULL);
//long long uh64 = r->getLongLong("uh64",0LL);
char *uh64str = r->getString("uh64",NULL);
//char *divTag = r->getString("div",NULL);
if ( uh64str ) {
// convert add to number
long addNum = 0;
if ( to_lower_a(add[0])=='t' ) // "true" or "false"?
addNum = 1;
// convert it. skip beginning "str" inserted to prevent
// javascript from messing with the long long since it
// was rounding it!
//long long uh64 = atoll(uh64str);//+3);
// urldecode that
//long divTagLen = gbstrlen(divTag);
//long newLen = urlDecode ( divTag , divTag , divTagLen );
// null term?
//divTag[newLen] = '\0';
// do it. this is defined in XmlDoc.cpp
//addCheckboxSpan ( uh64 , divTag , addNum );
// make basic reply
char *reply;
reply = "HTTP/1.0 200 OK\r\n"
"Connection: Close\r\n";
// that is it! send a basic reply ok
bool status = g_httpServer.sendDynamicPage( s ,
reply,
gbstrlen(reply),
-1, //cachtime
false ,//postreply?
NULL, //ctype
-1 , //httpstatus
NULL,//cookie
"utf-8");
return status;
}
// make a state
if ( st ) st->m_freeIt = false;
if ( ! st ) {
try { st = new (State8); }
catch ( ... ) {
g_errno = ENOMEM;
log("PageParser: new(%i): %s",
(int)sizeof(State8),mstrerror(g_errno));
return g_httpServer.sendErrorReply(s,500,
mstrerror(g_errno));}
mnew ( st , sizeof(State8) , "PageParser" );
st->m_freeIt = true;
}
// msg3b uses this to get a score from the query
st->m_state = state;
st->m_callback = callback;
st->m_q = q;
st->m_termFreqs = termFreqs;
st->m_termFreqWeights = termFreqWeights;
st->m_affWeights = affWeights;
//st->m_total = (score_t)-1;
st->m_indexCode = 0;
st->m_blocked = false;
st->m_didRootDom = false;
st->m_didRootWWW = false;
st->m_wasRootDom = false;
st->m_u = NULL;
st->m_recompute = false;
//st->m_url.reset();
// do not allow more than one to be launched at a time if in
// a quickpoll. will cause quickpoll in quickpoll.
g_inPageParser = true;
// password, too
long pwdLen = 0;
char *pwd = r->getString ( "pwd" , &pwdLen );
if ( pwdLen > 31 ) pwdLen = 31;
if ( pwdLen > 0 ) strncpy ( st->m_pwd , pwd , pwdLen );
st->m_pwd[pwdLen]='\0';
// save socket ptr
st->m_s = s;
st->m_r.copy ( r );
// get the collection
char *coll = r->getString ( "c" , &st->m_collLen ,NULL /*default*/);
if ( st->m_collLen > MAX_COLL_LEN )
return sendErrorReply ( st , ENOBUFS );
if ( ! coll )
return sendErrorReply ( st , ENOCOLLREC );
strcpy ( st->m_coll , coll );
// version to use, if -1 use latest
st->m_titleRecVersion = r->getLong("version",-1);
if ( st->m_titleRecVersion == -1 )
st->m_titleRecVersion = TITLEREC_CURRENT_VERSION;
// default to 0 if not provided
st->m_hopCount = r->getLong("hc",0);
//long ulen = 0;
//char *u = r->getString ( "u" , &ulen , NULL /*default*/);
long old = r->getLong ( "old", 0 );
// set query
long qlen;
char *qs = r->getString("q",&qlen,NULL);
if ( qs ) st->m_tq.set2 ( qs , langUnknown , true );
// url will override docid if given
if ( ! st->m_u || ! st->m_u[0] )
st->m_docId = r->getLongLong ("docid",-1);
else
st->m_docId = -1;
// set url in state class (may have length 0)
//if ( u ) st->m_url.set ( u , ulen );
//st->m_urlLen = ulen;
st->m_u = st->m_r.getString("u",&st->m_ulen,NULL);
// should we recycle link info?
st->m_recycle = r->getLong("recycle",0);
st->m_recycle2 = r->getLong("recycleimp",0);
st->m_render = r->getLong("render" ,0);
// for quality computation... takes way longer cuz we have to
// lookup the IP address of every outlink, so we can get its root
// quality using Msg25 which needs to filter out voters from that IP
// range.
st->m_oips = r->getLong("oips" ,0);
long linkInfoLen = 0;
// default is NULL
char *linkInfoColl = r->getString ( "oli" , &linkInfoLen, NULL );
if ( linkInfoColl ) strcpy ( st->m_linkInfoColl , linkInfoColl );
else st->m_linkInfoColl[0] = '\0';
// set the flag in our SafeBuf class so that Words.cpp knows to show
// html or html source depending on this value
st->m_xbuf.m_renderHtml = st->m_render;
// should we use the old title rec?
st->m_old = old;
// are we coming from a local machine?
st->m_isLocal = r->isLocal();
//no more setting the default root quality to 30, instead if we do not
// know it setting it to -1
st->m_rootQuality=-1;
// header
SafeBuf *xbuf = &st->m_xbuf;
xbuf->safePrintf("<meta http-equiv=\"Content-Type\" "
"content=\"text/html; charset=utf-8\">\n");
// print standard header
g_pages.printAdminTop ( xbuf , st->m_s , &st->m_r );
// print the standard header for admin pages
char *dd = "";
char *rr = "";
char *rr2 = "";
char *render = "";
char *oips = "";
char *us = "";
if ( st->m_u && st->m_u[0] ) us = st->m_u;
//if ( st->m_sfn != -1 ) sprintf ( rtu , "%li",st->m_sfn );
if ( st->m_old ) dd = " checked";
if ( st->m_recycle ) rr = " checked";
if ( st->m_recycle2 ) rr2 = " checked";
if ( st->m_render ) render = " checked";
if ( st->m_oips ) oips = " checked";
xbuf->safePrintf(
"<style>"
".poo { background-color:#%s;}\n"
"</style>\n" ,
LIGHT_BLUE );
long clen;
char *contentParm = r->getString("content",&clen,"");
// print the input form
xbuf->safePrintf (
"<style>\n"
"h2{font-size: 12px; color: #666666;}\n"
".gbtag { border: 1px solid gray;"
"background: #ffffef;display:inline;}\n"
".gbcomment { border: 1px solid gray;"
"color: #888888; font-style:italic; "
"background: #ffffef;display:inline;}\n"
".token { border: 1px solid gray;"
"background: #f0ffff;display:inline;}\n"
".spam { border: 1px solid gray;"
"background: #af0000;"
"color: #ffffa0;}"
".hs {color: #009900;}"
"</style>\n"
"<center>"
"<table %s>"
"<tr><td colspan=5><center><b>"
"Parser"
"</b></center></td></tr>\n"
"<tr class=poo>"
"<td>"
"<b>url</b>"
"<br><font size=-2>"
"Type in <b>FULL</b> url to parse."
"</font>"
"</td>"
"</td>"
"<td>"
"<input type=text name=u value=\"%s\" size=\"40\">\n"
"</td>"
"</tr>"
/*
"<tr class=poo>"
"<td>"
"Parser version to use: "
"</td>"
"<td>"
"<input type=text name=\"version\" size=\"4\" value=\"-1\"> "
"</td>"
"<td>"
"(-1 means to use latest title rec version)<br>"
"</td>"
"</tr>"
*/
/*
"<tr class=poo>"
"<td>"
"Hop count to use: "
"</td>"
"<td>"
"<input type=text name=\"hc\" size=\"4\" value=\"%li\"> "
"</td>"
"<td>"
"(-1 is unknown. For root urls hopcount is always 0)<br>"
"</td>"
"</tr>"
*/
"<tr class=poo>"
"<td>"
"<b>use cached</b>"
"<br><font size=-2>"
"Load page from cache (titledb)?"
"</font>"
"</td>"
"<td>"
"<input type=checkbox name=old value=1%s> "
"</td>"
"</tr>"
/*
"<tr class=poo>"
"<td>"
"Reparse root:"
"</td>"
"<td>"
"<input type=checkbox name=artr value=1%s> "
"</td>"
"<td>"
"Apply selected ruleset to root to update quality"
"</td>"
"</tr>"
*/
"<tr class=poo>"
"<td>"
"<b>recycle link info</b>"
"<br><font size=-2>"
"Recycle the link info from the title rec"
"Load page from cache (titledb)?"
"</font>"
"</td>"
"<td>"
"<input type=checkbox name=recycle value=1%s> "
"</td>"
"</tr>"
/*
"<tr class=poo>"
"<td>"
"Recycle Link Info Imported:"
"</td>"
"<td>"
"<input type=checkbox name=recycleimp value=1%s> "
"</td>"
"<td>"
"Recycle the link info imported from other coll"
"</td>"
"</tr>"
*/
"<tr class=poo>"
"<td>"
"<b>render html</b>"
"<br><font size=-2>"
"Render document content as HTML"
"</font>"
"</td>"
"<td>"
"<input type=checkbox name=render value=1%s> "
"</td>"
"</tr>"
/*
"<tr class=poo>"
"<td>"
"Lookup outlinks' ruleset, ips, quality:"
"</td>"
"<td>"
"<input type=checkbox name=oips value=1%s> "
"</td>"
"<td>"
"To compute quality lookup IP addresses of roots "
"of outlinks."
"</td>"
"</tr>"
"<tr class=poo>"
"<td>"
"LinkInfo Coll:"
"</td>"
"<td>"
"<input type=text name=\"oli\" size=\"10\" value=\"\"> "
"</td>"
"<td>"
"Leave empty usually. Uses this coll to lookup link info."
"</td>"
"</tr>"
*/
"<tr class=poo>"
"<td>"
"<b>optional query</b>"
"<br><font size=-2>"
"Leave empty usually. For title generation only."
"</font>"
"</td>"
"<td>"
"<input type=text name=\"q\" size=\"20\" value=\"\"> "
"</td>"
"</tr>",
TABLE_STYLE,
us ,
dd,
rr,
render
);
xbuf->safePrintf(
"<tr class=poo>"
"<td>"
"<b>content type below is</b>"
"<br><font size=-2>"
"Is the content below HTML? XML? JSON?"
"</font>"
"</td>"
"<td>"
//"<input type=checkbox name=xml value=1> "
"<select name=ctype>\n"
"<option value=%li selected>HTML</option>\n"
"<option value=%li selected>XML</option>\n"
"<option value=%li selected>JSON</option>\n"
"</select>\n"
"</td>"
"</tr>",
(long)CT_HTML,
(long)CT_XML,
(long)CT_JSON
);
xbuf->safePrintf(
"<tr class=poo>"
"<td><b>content</b>"
"<br><font size=-2>"
"Use this content for the provided <i>url</i> "
"rather than downloading it from the web."
"</td>"
"<td>"
"<textarea rows=10 cols=80 name=content>"
"%s"
"</textarea>"
"</td>"
"</tr>"
"</table>"
"</center>"
"</form>"
"<br>",
//oips ,
contentParm );
xbuf->safePrintf(
"<center>"
"<input type=submit value=Submit>"
"</center>"
);
// just print the page if no url given
if ( ! st->m_u || ! st->m_u[0] ) return processLoop ( st );
XmlDoc *xd = &st->m_xd;
// set this up
SpiderRequest sreq;
sreq.reset();
strcpy(sreq.m_url,st->m_u);
long firstIp = hash32n(st->m_u);
if ( firstIp == -1 || firstIp == 0 ) firstIp = 1;
// parentdocid of 0
sreq.setKey( firstIp, 0LL, false );
sreq.m_isPageParser = 1;
sreq.m_hopCount = st->m_hopCount;
sreq.m_hopCountValid = 1;
sreq.m_fakeFirstIp = 1;
sreq.m_firstIp = firstIp;
Url nu;
nu.set(sreq.m_url);
sreq.m_domHash32 = nu.getDomainHash32();
sreq.m_siteHash32 = nu.getHostHash32();
// . get provided content if any
// . will be NULL if none provided
// . "content" may contain a MIME
long contentLen = 0;
char *content = r->getString ( "content" , &contentLen , NULL );
// is the "content" url-encoded? default is true.
bool contentIsEncoded = true;
// mark doesn't like to url-encode his content
if ( ! content ) {
content = r->getUnencodedContent ();
contentLen = r->getUnencodedContentLen ();
contentIsEncoded = false;
}
// ensure null
if ( contentLen == 0 ) content = NULL;
uint8_t contentType = CT_HTML;
if ( r->getBool("xml",0) ) contentType = CT_XML;
contentType = r->getLong("ctype",contentType);//CT_HTML);
// if facebook, load xml content from title rec...
bool isFacebook = (bool)strstr(st->m_u,"http://www.facebook.com/");
if ( isFacebook && ! content ) {
long long docId = g_titledb.getProbableDocId(st->m_u);
sprintf(sreq.m_url ,"%llu", docId );
sreq.m_isPageReindex = true;
}
// hack
if ( content ) {
st->m_dbuf.purge();
st->m_dbuf.safeStrcpy(content);
//char *data = strstr(content,"\r\n\r\n");
//long dataPos = 0;
//if ( data ) dataPos = (data + 4) - content;
//st->m_dbuf.convertJSONtoXML(0,dataPos);
//st->m_dbuf.decodeJSON(0);
content = st->m_dbuf.getBufStart();
}
// . use the enormous power of our new XmlDoc class
// . this returns false if blocked
if ( ! xd->set4 ( &sreq ,
NULL ,
st->m_coll ,
&st->m_wbuf ,
0 ,//PP_NICENESS ))
content ,
false, // deletefromindex
0, // forced ip
contentType ))
// return error reply if g_errno is set
return sendErrorReply ( st , g_errno );
// make this our callback in case something blocks
xd->setCallback ( st , processLoop );
// . set xd from the old title rec if recycle is true
// . can also use XmlDoc::m_loadFromOldTitleRec flag
if ( st->m_recycle ) xd->m_recycleContent = true;
return processLoop ( st );
}
// for procog
bool sendPageAnalyze ( TcpSocket *s , HttpRequest *r ) {
// make a state
State8 *st;
try { st = new (State8); }
catch ( ... ) {
g_errno = ENOMEM;
log("PageParser: new(%i): %s",
(int)sizeof(State8),mstrerror(g_errno));
return g_httpServer.sendErrorReply(s,500,
mstrerror(g_errno));}
mnew ( st , sizeof(State8) , "PageParser" );
st->m_freeIt = true;
st->m_state = NULL;
//st->m_callback = callback;
//st->m_q = q;
//st->m_termFreqs = termFreqs;
//st->m_termFreqWeights = termFreqWeights;
//st->m_affWeights = affWeights;
//st->m_total = (score_t)-1;
st->m_indexCode = 0;
st->m_blocked = false;
st->m_didRootDom = false;
st->m_didRootWWW = false;
st->m_wasRootDom = false;
st->m_u = NULL;
// password, too
long pwdLen = 0;
char *pwd = r->getString ( "pwd" , &pwdLen );
if ( pwdLen > 31 ) pwdLen = 31;
if ( pwdLen > 0 ) strncpy ( st->m_pwd , pwd , pwdLen );
st->m_pwd[pwdLen]='\0';
// save socket ptr
st->m_s = s;
st->m_r.copy ( r );
// get the collection
char *coll = r->getString ( "c" , &st->m_collLen ,NULL /*default*/);
if ( ! coll ) coll = g_conf.m_defaultColl;
if ( ! coll ) coll = "main";
long collLen = gbstrlen(coll);
if ( collLen > MAX_COLL_LEN ) return sendErrorReply ( st , ENOBUFS );
strcpy ( st->m_coll , coll );
// version to use, if -1 use latest
st->m_titleRecVersion = r->getLong("version",-1);
if ( st->m_titleRecVersion == -1 )
st->m_titleRecVersion = TITLEREC_CURRENT_VERSION;
// default to 0 if not provided
st->m_hopCount = r->getLong("hc",0);
long old = r->getLong ( "old", 0 );
// set query
long qlen;
char *qs = r->getString("q",&qlen,NULL);
if ( qs ) st->m_tq.set2 ( qs , langUnknown , true );
// url will override docid if given
st->m_docId = r->getLongLong ("d",-1);
st->m_docId = r->getLongLong ("docid",st->m_docId);
long ulen;
char *u = st->m_r.getString("u",&ulen,NULL);
if ( ! u ) u = st->m_r.getString("url",&ulen,NULL);
if ( ! u && st->m_docId == -1LL )
return sendErrorReply ( st , EBADREQUEST );
// set url in state class (may have length 0)
//if ( u ) st->m_url.set ( u , ulen );
//st->m_urlLen = ulen;
st->m_u = u;
st->m_ulen = 0;
if ( u ) st->m_ulen = gbstrlen(u);
// should we recycle link info?
st->m_recycle = r->getLong("recycle",1);
st->m_recycle2 = r->getLong("recycleimp",0);
st->m_render = r->getLong("render" ,0);
st->m_recompute = r->getLong("recompute" ,0);
// for quality computation... takes way longer cuz we have to
// lookup the IP address of every outlink, so we can get its root
// quality using Msg25 which needs to filter out voters from that IP
// range.
st->m_oips = r->getLong("oips" ,0);
//st->m_page = r->getLong("page",1);
long linkInfoLen = 0;
// default is NULL
char *linkInfoColl = r->getString ( "oli" , &linkInfoLen, NULL );
if ( linkInfoColl ) strcpy ( st->m_linkInfoColl , linkInfoColl );
else st->m_linkInfoColl[0] = '\0';
// set the flag in our SafeBuf class so that Words.cpp knows to show
// html or html source depending on this value
//st->m_xbuf.m_renderHtml = st->m_render;
// should we use the old title rec?
st->m_old = old;
// are we coming from a local machine?
st->m_isLocal = r->isLocal();
//no more setting the default root quality to 30, instead if we do not
// know it setting it to -1
st->m_rootQuality=-1;
// header
//xbuf->safePrintf("<meta http-equiv=\"Content-Type\" "
// "content=\"text/html; charset=utf-8\">\n");
XmlDoc *xd = &st->m_xd;
long isXml = r->getLong("xml",0);
// if got docid, use that
if ( st->m_docId != -1 ) {
if ( ! xd->set3 ( st->m_docId,
st->m_coll,
0 ) ) // niceness
// return error reply if g_errno is set
return sendErrorReply ( st , g_errno );
// make this our callback in case something blocks
xd->setCallback ( st , gotXmlDoc );
xd->m_pbuf = &st->m_wbuf;
// reset this flag
st->m_donePrinting = false;
// . set xd from the old title rec if recycle is true
// . can also use XmlDoc::m_loadFromOldTitleRec flag
//if ( st->m_recycle ) xd->m_recycleContent = true;
xd->m_recycleContent = true;
// force this on
//xd->m_useSiteLinkBuf = true;
//xd->m_usePageLinkBuf = true;
if ( isXml ) xd->m_printInXml = true;
// now tell it to fetch the old title rec
if ( ! xd->loadFromOldTitleRec () )
// return false if this blocks
return false;
return gotXmlDoc ( st );
}
// set this up
SpiderRequest sreq;
sreq.reset();
if ( st->m_u ) strcpy(sreq.m_url,st->m_u);
long firstIp = hash32n(st->m_u);
if ( firstIp == -1 || firstIp == 0 ) firstIp = 1;
// parentdocid of 0
sreq.setKey( firstIp, 0LL, false );
sreq.m_isPageParser = 1;
sreq.m_hopCount = st->m_hopCount;
sreq.m_hopCountValid = 1;
sreq.m_fakeFirstIp = 1;
sreq.m_firstIp = firstIp;
Url nu;
nu.set(sreq.m_url);
sreq.m_domHash32 = nu.getDomainHash32();
sreq.m_siteHash32 = nu.getHostHash32();
// . get provided content if any
// . will be NULL if none provided
// . "content" may contain a MIME
long contentLen = 0;
char *content = r->getString ( "content" , &contentLen , NULL );
// is the "content" url-encoded? default is true.
bool contentIsEncoded = true;
// mark doesn't like to url-encode his content
if ( ! content ) {
content = r->getUnencodedContent ();
contentLen = r->getUnencodedContentLen ();
contentIsEncoded = false;
}
// ensure null
if ( contentLen == 0 ) content = NULL;
//uint8_t contentType = CT_HTML;
//if ( isXml ) contentType = CT_XML;
long ctype = r->getLong("ctype",CT_HTML);
// . use the enormous power of our new XmlDoc class
// . this returns false if blocked
if ( ! xd->set4 ( &sreq ,
NULL ,
st->m_coll ,
// we need this so the term table is set!
&st->m_wbuf , // XmlDoc::m_pbuf
0, // try 0 now! 1 ,//PP_NICENESS ))
content ,
false, // deletefromindex
0, // forced ip
ctype ))
// return error reply if g_errno is set
return sendErrorReply ( st , g_errno );
// make this our callback in case something blocks
xd->setCallback ( st , gotXmlDoc );
// reset this flag
st->m_donePrinting = false;
// prevent a core here in the event we download the page content
xd->m_crawlDelayValid = true;
xd->m_crawlDelay = 0;
// . set xd from the old title rec if recycle is true
// . can also use XmlDoc::m_loadFromOldTitleRec flag
//if ( st->m_recycle ) xd->m_recycleContent = true;
// only recycle if docid is given!!
if ( st->m_recycle ) xd->m_recycleContent = true;
// force this on
//xd->m_useSiteLinkBuf = true;
//xd->m_usePageLinkBuf = true;
if ( isXml ) xd->m_printInXml = true;
return gotXmlDoc ( st );
}
int main ( int argc , char *argv[] ) {
bool addWWW = true;
bool stripSession = true;
// check for arguments
for (int32_t i = 1; i < argc; i++) {
if (strcmp(argv[i], "-w") == 0)
addWWW = false;
else if (strcmp(argv[i], "-s") == 0)
stripSession = false;
}
// initialize
//g_mem.init(100*1024);
hashinit();
//g_conf.m_tfndbExtBits = 23;
loop:
// read a url from stddin
char sbuf[1024];
if ( ! fgets ( sbuf , 1024 , stdin ) ) exit(1);
char *s = sbuf;
char fbuf[1024];
// decode if we should
if ( strncmp(s,"http%3A%2F%2F",13) == 0 ||
strncmp(s,"https%3A%2F%2F",13) == 0 ) {
urlDecode(fbuf,s,gbstrlen(s));
s = fbuf;
}
// old url
printf("###############\n");
printf("old: %s",s);
int32_t slen = gbstrlen(s);
// remove any www. if !addWWW
if (!addWWW) {
if (slen >= 4 &&
strncasecmp(s, "www.", 4) == 0) {
slen -= 4;
memmove(s, &s[4], slen);
}
else {
// get past a ://
int32_t si = 0;
while (si < slen &&
( s[si] != ':' ||
s[si+1] != '/' ||
s[si+2] != '/' ) )
si++;
// remove the www.
if (si + 7 < slen) {
si += 3;
if (strncasecmp(&s[si], "www.", 4) == 0) {
slen -= 4;
memmove(&s[si], &s[si+4], slen-si);
}
}
}
}
// set it
Url u;
u.set ( s , slen ,
addWWW , /*add www?*/
stripSession ); /*strip session ids?*/
// print it
char out[1024*4];
char *p = out;
p += sprintf(p,"tld: ");
gbmemcpy ( p, u.getTLD(),u.getTLDLen());
p += u.getTLDLen();
char c = *p;
*p = '\0';
printf("%s\n",out);
*p = c;
// dom
p = out;
sprintf ( p , "dom: ");
p += gbstrlen ( p );
gbmemcpy ( p , u.getDomain() , u.getDomainLen() );
p += u.getDomainLen();
c = *p;
*p = '\0';
printf("%s\n",out);
*p = c;
// host
p = out;
sprintf ( p , "host: ");
p += gbstrlen ( p );
gbmemcpy ( p , u.getHost() , u.getHostLen() );
p += u.getHostLen();
c = *p;
*p = '\0';
printf("%s\n",out);
*p = c;
// then the whole url
printf("url: %s\n", u.getUrl() );
/*
int32_t siteLen;
char *site = u.getSite ( &siteLen , NULL , false );
if ( site ) {
c = site[siteLen];
site[siteLen] = '\0';
}
printf("site: %s\n", site );
if ( site ) site[siteLen] = c;
*/
SiteGetter sg;
sg.getSite ( u.getUrl() ,
NULL , // tagrec
0 , // timestamp
NULL, // coll
0 , // niceness
//false , // addtags
NULL , // state
NULL ); // callback
if ( sg.m_siteLen )
printf("site: %s\n",sg.m_site);
printf("isRoot: %"INT32"\n",(int32_t)u.isRoot());
/*
bool perm = ::isPermalink ( NULL , // coll
NULL , // Links ptr
&u , // the url
CT_HTML , // contentType
NULL , // LinkInfo ptr
false );// isRSS?
printf ("isPermalink: %"INT32"\n",(int32_t)perm);
*/
// print the path too
p = out;
p += sprintf ( p , "path: " );
gbmemcpy ( p , u.getPath(), u.getPathLen() );
p += u.getPathLen();
if ( u.getFilename() ) {
p += sprintf ( p , "\nfilename: " );
gbmemcpy ( p , u.getFilename(), u.getFilenameLen() );
p += u.getFilenameLen();
*p = '\0';
printf("%s\n", out );
}
// encoded
char dst[MAX_URL_LEN+200];
urlEncode ( dst,MAX_URL_LEN+100,
u.getUrl(), u.getUrlLen(),
false ); // are we encoding a request path?
printf("encoded: %s\n",dst);
// the probable docid
int64_t pd = g_titledb.getProbableDocId(&u);
printf("pdocid: %"UINT64"\n", pd );
printf("dom8: 0x%"XINT32"\n", (int32_t)g_titledb.getDomHash8FromDocId(pd) );
//printf("ext23: 0x%"XINT32"\n",g_tfndb.makeExt(&u));
if ( u.isLinkLoop() ) printf("islinkloop: yes\n");
else printf("islinkloop: no\n");
int64_t hh64 = u.getHostHash64();
printf("hosthash64: 0x%016"XINT64"\n",hh64);
uint32_t hh32 = u.getHostHash32();
printf("hosthash32: 0x%08"XINT32" (%"UINT32")\n",hh32,hh32);
int64_t dh64 = u.getDomainHash64();
printf("domhash64: 0x%016"XINT64"\n",dh64);
int64_t uh64 = u.getUrlHash64();
printf("urlhash64: 0x%016"XINT64"\n",uh64);
//if(isUrlUnregulated(NULL ,0,&u)) printf("unregulated: yes\n");
//else printf("unregulated: no\n");
goto loop;
}
