Add skiphash as parameter for injecting document so we can force a reindex even when content is the same

PageInject.cpp

@@ -596,7 +596,7 @@ void handleRequest7 ( UdpSlot *slot , int32_t netnice ) {
 			       ir->ptr_contentTypeStr, // text/html text/xml
 			       ir->m_spiderLinks ,
 			       ir->m_newOnly, // index iff new
-
+			       ir->m_skipContentHashCheck,
 			       xd, // state ,
 			       sendUdpReply7 ,
 

PageInject.h

@@ -30,6 +30,7 @@ class InjectionRequest {
 	bool      m_spiderLinks;
 	bool      m_shortReply;
 	bool      m_newOnly;
+	bool      m_skipContentHashCheck;
 	bool      m_deleteUrl;
 	bool      m_hasMime;
 	bool      m_doConsistencyTesting;

Parms.cpp

@@ -6698,6 +6698,16 @@ void Parms::init ( ) {
 	simple_m_set_checkbox(InjectionRequest,m_newOnly);
 	m++;
 
+	m->m_title = "skip content hash check";
+	m->m_desc  = "Skip content hash check to force reindexing of document even "
+		"when content is identical";
+	m->m_cgi   = "skiphash";
+	m->m_def   = "0";
+	m->m_flags = PF_API;
+	m->m_page  = PAGE_INJECT;
+	simple_m_set_checkbox(InjectionRequest,m_skipContentHashCheck);
+	m++;
+
 	m->m_title = "delete from index";
 	m->m_desc  = "Delete the specified url from the index.";
 	m->m_cgi   = "deleteurl";

XmlDoc.cpp

@@ -218,6 +218,7 @@ void XmlDoc::reset ( ) {
 
 	// if this is true, then only index if new
 	m_newOnly = 0;
+	m_skipContentHashCheck = false;
 
 	if ( m_httpReplyValid && m_httpReply ) {
 		mfree(m_httpReply,m_httpReplyAllocSize,"httprep");
@@ -1287,7 +1288,7 @@ bool XmlDoc::injectDoc ( const char *url ,
 			 const char *contentTypeStr, // text/html application/json
 			 bool spiderLinks ,
 			 char newOnly, // index iff new
-
+			 bool skipContentHashCheck,
 			 void *state,
 			 void (*callback)(void *state) ,
 
@@ -1398,6 +1399,8 @@ bool XmlDoc::injectDoc ( const char *url ,
 	// . maybe just set indexCode
 	m_newOnly = newOnly;
 
+	m_skipContentHashCheck = skipContentHashCheck;
+
 	// do not re-lookup the robots.txt
 	m_isAllowed      = true;
 	m_isAllowedValid = true;
@@ -2512,8 +2515,10 @@ int32_t *XmlDoc::getIndexCode ( ) {
 
 		// disable content hash check if language differ (we could have overridden language when injecting doc)
 		bool checkContentHash = true;
-		if (m_wasContentInjected && m_langIdValid && m_langId != od->m_langId) {
-			checkContentHash = false;
+		if (m_wasContentInjected) {
+			if (m_skipContentHashCheck || (m_langIdValid && m_langId != od->m_langId)) {
+				checkContentHash = false;
+			}
 		}
 
 		if (checkContentHash && *ch32 == od->m_contentHash32) {

XmlDoc.h

@@ -907,6 +907,7 @@ public:
 	// for limiting # of iframe tag expansions
 	int32_t m_numExpansions;
 	char m_newOnly;
+	bool m_skipContentHashCheck;
 	char m_isWWWDup;	// May be -1
 
 	SafeBuf m_linkSiteHashBuf;
@@ -1143,7 +1144,7 @@ public:
 			 const char *contentTypeStr, // text/html, text/xml etc.
 			 bool spiderLinks ,
 			 char newOnly, // index iff new
-
+			 bool skipContentHashCheck,
 			 void *state,
 			 void (*callback)(void *state) ,