Merge branch 'BABEL_3_X_DEV' into BABEL-4590

contrib/babelfishpg_tds/src/backend/tds/tds_srv.c

@@ -340,6 +340,25 @@ pe_authenticate(Port *port, const char **username)
 
 		appendStringInfo(&logmsg, _(" Tds Version=0x%X."), GetClientTDSVersion());
 
+#ifdef ENABLE_GSS
+		if (port->gss)
+		{
+			const char *princ = be_gssapi_get_princ(port);
+
+			if (princ)
+				appendStringInfo(&logmsg,
+								 _(" GSS (authenticated=%s, encrypted=%s, principal=%s)"),
+								 be_gssapi_get_auth(port) ? _("yes") : _("no"),
+								 be_gssapi_get_enc(port) ? _("yes") : _("no"),
+								 princ);
+			else
+				appendStringInfo(&logmsg,
+								 _(" GSS (authenticated=%s, encrypted=%s)"),
+								 be_gssapi_get_auth(port) ? _("yes") : _("no"),
+								 be_gssapi_get_enc(port) ? _("yes") : _("no"));
+		}
+#endif
+
 		ereport(LOG, errmsg_internal("%s", logmsg.data));
 		pfree(logmsg.data);
 	}

contrib/babelfishpg_tds/src/backend/tds/tdslogin.c

@@ -1462,6 +1462,7 @@ CheckGSSAuth(Port *port)
 	gss_buffer_desc gbuf;
 	MemoryContext oldContext;
 	char	   *at_pos = NULL;
+	char	   *princ;
 
 	if (pg_krb_server_keyfile && strlen(pg_krb_server_keyfile) > 0)
 	{
@@ -1583,6 +1584,24 @@ CheckGSSAuth(Port *port)
 						 _("retrieving GSS user name failed"),
 						 maj_stat, min_stat);
 
+	/*
+	 * gbuf.value might not be null-terminated, so turn it into a regular
+	 * null-terminated string.
+	 */
+	princ = palloc(gbuf.length + 1);
+	memcpy(princ, gbuf.value, gbuf.length);
+	princ[gbuf.length] = '\0';
+
+	/*
+	 * Copy the original name of the authenticated principal into our backend
+	 * memory for display later.
+	 *
+	 * This is also our authenticated identity.  Set it now, rather than
+	 * waiting for the usermap check below, because authentication has already
+	 * succeeded and we want the log file to reflect that.
+	 */
+	port->gss->princ = MemoryContextStrdup(TopMemoryContext, princ);
+
 	/*
 	 * XXX: In PG there are options to match realm names or perform ident
 	 * mappings. We're not going to do those checks now.  If required, we can

contrib/babelfishpg_tds/test/t/002_tdskerberos.pl

@@ -418,6 +418,13 @@ sub test_query
 my @connstr1 = $tsql_node->tsql_connstr('master');
 $tsql_node->connect_ok('Kerberos auth test', (connstr => \@connstr1));
 
+$tsql_node->connect_ok('pg_stat_gssapi and connection log test',
+			(connstr => \@connstr1,
+			 sql => "SELECT gss_authenticated, principal from pg_stat_gssapi where pid = pg_backend_pid();",
+      			 expected_stdout => qr/1.*test1\@$realm/,
+	    		 log_like => [qr/connection authorized: user=test1\@$realm, application=SQLCMD, Tds Version=0x\d\d\d\d\d\d\d\d. GSS \(authenticated=yes, encrypted=no, principal=test1\@$realm\)/]));
+
+
 # Reset pg_hba.conf and mark every connection to use password based auth
 # But we should be able to use kerberos auth through TDS endpoint irrespective
 # of pg_hba.conf file

contrib/babelfishpg_tsql/src/fts_data.h

@@ -6,6 +6,7 @@ extern int	fts_yylex(void);
 extern void fts_yyerror(char **result, const char *message) pg_attribute_noreturn();
 extern void fts_scanner_init(const char *str);
 extern void fts_scanner_finish(void);
+extern bool isNonEnglishString(const char *str);
 
 /* in fts_parser.y */
 extern int fts_yyparse(char **result);

contrib/babelfishpg_tsql/src/fts_parser.y

@@ -23,10 +23,11 @@ static int	scanbuflen;
 
 static char* translate_simple_term(const char* s);
 static char *trim(char *s, bool insideQuotes);
+static bool containsSpecialCharacters(char s);
 
 %}
 
-%token WORD_TOKEN WS_TOKEN TEXT_TOKEN PREFIX_TERM_TOKEN GENERATION_TERM_TOKEN AND_TOKEN AND_NOT_TOKEN OR_TOKEN INFLECTIONAL_TOKEN THESAURUS_TOKEN FORMSOF_TOKEN O_PAREN_TOKEN C_PAREN_TOKEN COMMA_TOKEN
+%token WORD_TOKEN WS_TOKEN TEXT_TOKEN PREFIX_TERM_TOKEN GENERATION_TERM_TOKEN AND_TOKEN NOT_TOKEN AND_NOT_TOKEN OR_TOKEN INFLECTIONAL_TOKEN THESAURUS_TOKEN FORMSOF_TOKEN O_PAREN_TOKEN C_PAREN_TOKEN COMMA_TOKEN SPECIAL_CHAR_TOKEN NON_ENGLISH_TOKEN
 %left OR_TOKEN
 %left AND_TOKEN
 %left AND_NOT_TOKEN
@@ -86,10 +87,10 @@ generation_term:
 
 generation_type:
     INFLECTIONAL_TOKEN {
-        $$ = $1;
+        fts_yyerror(NULL, "Generation term is not currently supported in Babelfish");
     }
     | THESAURUS_TOKEN {
-        $$ = $1;
+        fts_yyerror(NULL, "Generation term is not currently supported in Babelfish");
     }
     ;
 
@@ -138,27 +139,35 @@ char* translate_simple_term(const char* inputStr) {
         inputLength = strlen(trimmedInputStr);
 
         initStringInfo(&output);
+        appendStringInfoString(&output, "");
 
         // Initialize pointers for input and output
-        inputPtr = trimmedInputStr;
-
-        while (*inputPtr != '\0') {
-            if (*inputPtr == ' ') {
+        for (inputPtr = trimmedInputStr; *inputPtr != '\0'; inputPtr++) {
+            if (isspace((unsigned char)*inputPtr)) {
                 // Replace space with "<->"
-                while (*(inputPtr + 1) == ' ') {
+                while (isspace((unsigned char)*(inputPtr + 1))) {
                     // Handle multiples spaces between words and skip over additional spaces
                     inputPtr++;
                 }
                 appendStringInfoString(&output, "<->");
+            } else if (*inputPtr != '-' && containsSpecialCharacters((unsigned char)*inputPtr)) {   // Handle special characters and other languages
+                pfree(trimmedInputStr);
+                yyerror(NULL, "Full-text search conditions with special characters or languages other than English are not currently supported in Babelfish");
             } else {
                 // Copy the character
                 appendStringInfoChar(&output, *inputPtr);
             }
-            inputPtr++;
         }
         pfree(trimmedInputStr);
         return output.data;
     } else {
+         // Initialize pointers for input and output
+        for (inputPtr = trimmedInputStr; *inputPtr != '\0'; inputPtr++) {
+            if (*inputPtr != '-' && containsSpecialCharacters((unsigned char)*inputPtr)) {   // Handle special characters and other languages
+                pfree(trimmedInputStr);
+                yyerror(NULL, "Full-text search conditions with special characters or languages other than English are not currently supported in Babelfish");
+            }
+        }
         // It's a single word, so no transformation needed
         return trimmedInputStr;
     }
@@ -173,7 +182,6 @@ static char *trim(char *s, bool insideQuotes) {
     size_t start;
     size_t end;
     size_t newLength;
-    bool inQuotes;
 
     /*
      * Empty string, nothing to trim
@@ -184,26 +192,22 @@ static char *trim(char *s, bool insideQuotes) {
         return s;
     }
 
-    inQuotes = false;
     start = 0;
     end = length - 1;
 
-    for (size_t i = 0; i < length; i++) {
-        if (s[i] == '"') {
-            inQuotes = !inQuotes;
-        }
+    if(insideQuotes) {
+        start++;
+        end--;
+    }
 
-        if (!inQuotes || insideQuotes) {
-            // Trim leading spaces
-            while (start < length && isspace(s[start])) {
-                start++;
-            }
+    // Trim leading spaces
+    while (start < length && isspace(s[start])) {
+        start++;
+    }
 
-            // Trim trailing spaces
-            while (end > start && isspace(s[end])) {
-                end--;
-            }
-        }
+    // Trim trailing spaces
+    while (end > start && isspace(s[end])) {
+        end--;
     }
 
     // Calculate the new length
@@ -218,4 +222,9 @@ static char *trim(char *s, bool insideQuotes) {
     return s;
 }
 
+static bool containsSpecialCharacters(char str) {
+    // Ignore Non-breaking space character
+    return str != '\302' && str != -62 && str != '\240' && str != -96 && !isalnum(str) && !isspace(str);
+}
+
 # include "fts_scan.c"

contrib/babelfishpg_tsql/src/fts_scan.l

@@ -21,9 +21,10 @@ static YY_BUFFER_STATE scanbufhandle;
 
 %%
 
-"AND"           { yylval = "AND"; yyerror(NULL, "Boolean operator is not currently supported in Babelfish"); return AND_TOKEN; }
-"AND NOT"       { yylval = "AND NOT"; yyerror(NULL, "Boolean operator is not currently supported in Babelfish"); return AND_NOT_TOKEN; }
-"OR"            { yylval = "OR"; yyerror(NULL, "Boolean operator is not currently supported in Babelfish"); return OR_TOKEN; }
+"AND"           { yylval = "AND"; yyerror(NULL, "Full-text search conditions with boolean operators are not currently supported in Babelfish"); return AND_TOKEN; }
+"AND NOT"       { yylval = "AND NOT"; yyerror(NULL, "Full-text search conditions with boolean operators are not currently supported in Babelfish"); return AND_NOT_TOKEN; }
+"OR"            { yylval = "OR"; yyerror(NULL, "Full-text search conditions with boolean operators are not currently supported in Babelfish"); return OR_TOKEN; }
+"NOT"			{ yylval = "NOT"; yyerror(NULL, "Full-text search conditions with boolean operators are not currently supported in Babelfish"); return NOT_TOKEN; }
 "INFLECTIONAL"  { yylval = "INFLECTIONAL"; return INFLECTIONAL_TOKEN; }
 "THESAURUS"     { yylval = "THESAURUS"; return THESAURUS_TOKEN; }
 "FORMSOF"       { yylval = "FORMSOF"; return FORMSOF_TOKEN; }
@@ -32,13 +33,13 @@ static YY_BUFFER_STATE scanbufhandle;
 ","             { yylval = ","; return COMMA_TOKEN; }
 [ \t\r\n]+      { yylval = " "; return WS_TOKEN; }
 
-\"[^"]+\*\"              { if (yytext[strlen(yytext)-2] == ' ') { yyerror(NULL, "Syntax error, space is not allowed before *"); } yylval = yytext; return PREFIX_TERM_TOKEN; }  // Handle prefix terms
-[a-zA-Z0-9]+       { yylval = yytext; return WORD_TOKEN; }  // Handle individual words
-\"[^"*]*\"          { yylval = yytext; return TEXT_TOKEN; }  // Handle double-quoted phrases
 
-FORMSOF\s*\(\s*(INFLECTIONAL|THESAURUS)\s*,\s*((\w+)|(\"[\w\s]+\"))(\s*,\s*((\w+)|(\"[\w\s]+\")))*\s*\) { yylval = yytext; return GENERATION_TERM_TOKEN; }  // Handle FORMSOF generation term
+FORMSOF\s*\(\s*(INFLECTIONAL|THESAURUS)\s*,\s*((\w+)|(\"[\w\s]+\"))(\s*,\s*((\w+)|(\"[\w\s]+\")))*\s*\)				{ yylval = yytext; yyerror(NULL, "Generation term is not currently supported in Babelfish"); yylval = yytext; return GENERATION_TERM_TOKEN; }  // Handle FORMSOF generation term
+\"[^"]+\*\"               																							{ if (yytext[strlen(yytext)-2] == ' ') { yyerror(NULL, "Syntax error, space is not allowed before *"); } yylval = yytext; return PREFIX_TERM_TOKEN; }  // Handle prefix terms
+\"[^"]*\"                																							{ yylval = yytext; return TEXT_TOKEN; }  // Handle double-quoted phrases
+[a-zA-Z0-9]+ 																										{ yylval = yytext; return WORD_TOKEN; }  // Handle individual words
 
-.               { return yytext[0]; }  // Other characters are returned as-is
+.               																									{ return yytext[0]; }  // Other characters are returned as-is
 
 %%
 
@@ -72,6 +73,11 @@ fts_scanner_init(const char *str)
 		yyerror(NULL, "Null or empty full-text predicate.");
 	}
 
+	// Check for Non-English Language, if true, throw unsupported error
+	if(isNonEnglishString(str)) {
+		yyerror(NULL, "Full-text search conditions with special characters or languages other than English are not currently supported in Babelfish");
+	}
+
 	/*
 	 * Make a scan buffer with special termination needed by flex.
 	 */
@@ -93,3 +99,20 @@ fts_scanner_finish(void)
 	yy_delete_buffer(scanbufhandle);
 	pfree(scanbuf);
 }
+
+/*
+ * Util function to check if the string is of Non-English language
+ */
+bool
+isNonEnglishString(const char *str) {
+	for (const char *p = str; *p; p++) {
+		// Ignore non-breaking space character
+		if ((*p == '\302' && *p == -62 ) || (*p == '\240' && *p == -96)) {
+			continue;
+		}
+		if (*p < 0 || *p > 127) {
+			return true;
+		}
+	}
+	return false;
+}