Merge branch 'master' into testing

Collectiondb.cpp

@@ -1570,6 +1570,7 @@ CollectionRec::CollectionRec() {
 	// inits for sortbydatetable
 	m_inProgress = false;
 	m_msg5       = NULL;
+	m_importState = NULL;
 	// JAB - track which regex parsers have been initialized
 	//log(LOG_DEBUG,"regex: %p initalizing empty parsers", m_pRegExParser);
 

Makefile

@@ -567,10 +567,10 @@ master-rpm:
 # deb-master
 master-deb:
 # need to change in changelog too!! dont' forget!!!
-	git archive --format=tar --prefix=gb-1.14/ master > ../gb_1.14.orig.tar
+	git archive --format=tar --prefix=gb-1.16/ master > ../gb_1.16.orig.tar
 	rm -rf debian
 # change "-p gb_1.0" to "-p gb_1.1" to update version for example
-	dh_make -e gigablast@mail.com -p gb_1.14 -f ../gb_1.14.orig.tar
+	dh_make -e gigablast@mail.com -p gb_1.16 -f ../gb_1.16.orig.tar
 # zero this out, it is just filed with the .txt files erroneously and it'll
 # try to automatiicaly install in /usr/docs/
 	rm debian/docs
@@ -595,7 +595,7 @@ master-deb:
 # upload den
 	scp gb*.deb gk268:/w/html/	
 # alien it
-	sudo alien --to-rpm gb_1.14-1_i386.deb
+	sudo alien --to-rpm gb_1.16-1_i386.deb
 # upload rpm
 	scp gb*.rpm gk268:/w/html/	
 

PageInject.cpp

@@ -917,6 +917,8 @@ bool resumeImports ( ) {
 	if ( s_tried ) return true;
 	s_tried = true;
 
+	if ( g_hostdb.m_hostId != 0 ) return true;
+
 	for ( long i = 0 ; i < g_collectiondb.m_numRecs ; i++ ) {
 		CollectionRec *cr = g_collectiondb.m_recs[i];
 		if ( ! cr ) continue;
@@ -1085,7 +1087,7 @@ bool ImportState::importLoop ( ) {
 
 	CollectionRec *cr = g_collectiondb.getRec ( m_collnum );
 
-	if ( ! cr ) { 
+	if ( ! cr || g_hostdb.m_hostId != 0 ) { 
 		// if coll was deleted!
 		log("import: collnum %li deleted while importing into",
 		    (long)m_collnum);
@@ -1108,6 +1110,20 @@ bool ImportState::importLoop ( ) {
 	}
 	
 
+	if ( ! cr->m_importEnabled ) {
+		// wait for all to return
+		if ( out > 0 ) return false;
+		// then delete it
+		log("import: collnum %li import loop disabled",
+		    (long)m_collnum);
+		mdelete ( this, sizeof(ImportState) , "impstate");
+		delete (this);
+		return true;
+	}
+
+
+
+
 	// scan each titledb file scanning titledb0001.dat first,
 	// titledb0003.dat second etc.
 
@@ -1452,6 +1468,7 @@ Multicast *ImportState::getAvailMulticast() { // Msg7 ( ) {
 }
 
 void saveImportStates ( ) {
+	if ( g_hostdb.m_myHost->m_hostId != 0 ) return;
 	for ( long i = 0 ; i < g_collectiondb.m_numRecs ; i++ ) {
 		CollectionRec *cr = g_collectiondb.m_recs[i];
 		if ( ! cr ) continue;

XmlDoc.cpp

@@ -18983,6 +18983,17 @@ char *XmlDoc::getSpiderLinks ( ) {
 	//	m_spiderLinks2     = false; 
 	//	m_spiderLinksValid = true ; }
 
+	// this slows importing down because we end up doing ip lookups
+	// for every outlink if "firstip" not in tagdb.
+	// shoot. set2() already sets m_spiderLinksValid to true so we
+	// have to override if importing.
+	if ( m_isImporting && m_isImportingValid ) {
+		m_spiderLinks  = false;
+		m_spiderLinks2 = false;
+		m_spiderLinksValid = true;
+		return &m_spiderLinks2;
+	}
+
 	// return the valid value
 	if ( m_spiderLinksValid ) return &m_spiderLinks2;
 
@@ -21801,8 +21812,6 @@ char *XmlDoc::getMetaList ( bool forDelete ) {
 	// likewise if there error was ENONCANONICAL treat it like that
 	if ( m_indexCode == EDOCNONCANONICAL )
 		spideringLinks = true;
-	
-
 
 	//
 	// . prepare the outlink info if we are adding links to spiderdb!
@@ -35859,7 +35868,7 @@ char **XmlDoc::getRootTitleBuf ( ) {
 	char *src     = NULL;
 	long  srcSize = 0;
 
-	if ( ptr_rootTitleBuf ) {
+	if ( ptr_rootTitleBuf || m_setFromTitleRec ) {
 		src    =  ptr_rootTitleBuf;
 		srcSize = size_rootTitleBuf;
 	}

changelog

@@ -1,5 +1,5 @@
-gb (1.14-1) unstable; urgency=low
+gb (1.16-1) unstable; urgency=low
 
   * More bug fixes.
 
- -- mwells <gigablast@mail.com>  Tue, 19 Sep 2014 21:38:35 -0700
+ -- mwells <gigablast@mail.com>  Tue, 24 Sep 2014 21:38:35 -0700