GH-3281: Update SPARQL/ARQ to check RDF Strings

afs · afs · commit 7bf61f2a5e85 · 2025-06-26T19:07:06.000+01:00
diff --git a/jena-arq/Grammar/arq.jj b/jena-arq/Grammar/arq.jj
@@ -93,8 +93,8 @@ String VersionSpecifier() : { Token t ; String version ; }
     | t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }
     )
     {
-      checkString(version, t.beginLine, t.beginColumn) ;
       version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+      checkRDFString(version, t.beginLine, t.beginColumn) ;
       return version;
     }
 }
@@ -1933,7 +1933,7 @@ String String() : { Token t ; String lex ; }
   | t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }
   )
     { lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;
-      checkString(lex, t.beginLine, t.beginColumn) ;
+      checkRDFString(lex, t.beginLine, t.beginColumn) ;
       return lex ;
     }
 }
diff --git a/jena-arq/Grammar/main.jj b/jena-arq/Grammar/main.jj
@@ -188,8 +188,8 @@ String VersionSpecifier() : { Token t ; String version ; }
     | t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }
     )
     {
-      checkString(version, t.beginLine, t.beginColumn) ;
       version = unescapeStr(version,  t.beginLine, t.beginColumn) ;
+      checkRDFString(version,  t.beginLine, t.beginColumn) ;
       return version;
     }
 }
@@ -2631,7 +2631,7 @@ String String() : { Token t ; String lex ; }
   | t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }
   )
     { lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;
-      checkString(lex, t.beginLine, t.beginColumn) ;
+      checkRDFString(lex,  t.beginLine, t.beginColumn) ;
       return lex ;
     }
 }
diff --git a/jena-arq/Grammar/sparql_12.jj b/jena-arq/Grammar/sparql_12.jj
@@ -83,8 +83,8 @@ String VersionSpecifier() : { Token t ; String version ; }
     | t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }
     )
     {
-      checkString(version, t.beginLine, t.beginColumn) ;
       version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+      checkRDFString(version, t.beginLine, t.beginColumn) ;
       return version;
     }
 }
@@ -1646,7 +1646,7 @@ String String() : { Token t ; String lex ; }
   | t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }
   )
     { lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;
-      checkString(lex, t.beginLine, t.beginColumn) ;
+      checkRDFString(lex, t.beginLine, t.beginColumn) ;
       return lex ;
     }
 }
diff --git a/jena-arq/src/main/java/org/apache/jena/sparql/lang/QueryParserBase.java b/jena-arq/src/main/java/org/apache/jena/sparql/lang/QueryParserBase.java
@@ -212,13 +212,21 @@ protected Node stripSign(Node node) {
         return NodeFactory.createLiteral(lex, lang, dt);
     }
 
-    // Because of Java (Java strings have surrogate pairs) we only detect singleton surrogates.
-    protected void checkString(String string, int line, int column) {
+    /**
+     * Apply any checks for "RDF String" to a string that has already had escape processing applied.
+     * An RDF String is a sequence of codepoints in the range U+0000 to U+10FFFF, excluding surrogates.
+     * Because this is java, we test for no non-paired surrogates.
+     * A surrogate pair is high-low.
+     */
+    protected static void checkRDFString(String string, int line, int column) {
         // Checks for bare surrogate pairs.
         for ( int i = 0; i < string.length(); i++ ) {
             // Not "codePointAt" which does surrogate processing.
             char ch = string.charAt(i);
 
+            if ( ! Character.isValidCodePoint(ch) )
+                throw new QueryParseException(String.format("Illegal code point in \\U sequence value: 0x%08X", ch), line, column);
+
             // Check surrogate pairs are in pairs. Pairs are high-low.
             if ( Character.isLowSurrogate(ch) )
                 throw new QueryParseException("Bad surrogate pair (low surrogate without high surrogate)", line, column);
@@ -369,7 +377,7 @@ protected String resolveQuotedIRI(String iriStr, int line, int column) {
         iriStr = stripQuotes(iriStr);
         iriStr = unescapeUnicode(iriStr, line, column);
         // Check for Unicode surrogates
-        checkString(iriStr, line, column);
+        checkRDFString(iriStr, line, column);
         return resolveIRI(iriStr, line, column);
     }
 
diff --git a/jena-arq/src/main/java/org/apache/jena/sparql/lang/arq/javacc/ARQParser.java b/jena-arq/src/main/java/org/apache/jena/sparql/lang/arq/javacc/ARQParser.java
@@ -149,8 +149,8 @@ final public void Prologue() throws ParseException {
       jj_consume_token(-1);
       throw new ParseException();
     }
-checkString(version, t.beginLine, t.beginColumn) ;
-      version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+      checkRDFString(version, t.beginLine, t.beginColumn) ;
       {if ("" != null) return version;}
     throw new Error("Missing return statement in function");
 }
@@ -7520,7 +7520,7 @@ final public Node BooleanLiteral() throws ParseException {
       throw new ParseException();
     }
 lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;
-      checkString(lex, t.beginLine, t.beginColumn) ;
+      checkRDFString(lex, t.beginLine, t.beginColumn) ;
       {if ("" != null) return lex ;}
     throw new Error("Missing return statement in function");
 }
diff --git a/jena-arq/src/main/java/org/apache/jena/sparql/lang/sparql_12/javacc/SPARQLParser12.java b/jena-arq/src/main/java/org/apache/jena/sparql/lang/sparql_12/javacc/SPARQLParser12.java
@@ -128,8 +128,8 @@ final public void Prologue() throws ParseException {
       jj_consume_token(-1);
       throw new ParseException();
     }
-checkString(version, t.beginLine, t.beginColumn) ;
-      version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+version = unescapeStr(version, t.beginLine, t.beginColumn) ;
+      checkRDFString(version, t.beginLine, t.beginColumn) ;
       {if ("" != null) return version;}
     throw new Error("Missing return statement in function");
 }
@@ -5889,7 +5889,7 @@ final public Node BooleanLiteral() throws ParseException {
       throw new ParseException();
     }
 lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;
-      checkString(lex, t.beginLine, t.beginColumn) ;
+      checkRDFString(lex, t.beginLine, t.beginColumn) ;
       {if ("" != null) return lex ;}
     throw new Error("Missing return statement in function");
 }
diff --git a/jena-arq/src/test/java/org/apache/jena/sparql/syntax/TestQueryParser.java b/jena-arq/src/test/java/org/apache/jena/sparql/syntax/TestQueryParser.java
@@ -70,7 +70,7 @@ public void syntax_unicode_escaped_surrogate_strings() {
 
     @Test
     public void syntax_unicode_surrogate_pair_by_unicode_escape() {
-        // Allow - because Java strings may have surrogate pairs so we allow then in unicode escapes if paired.
+        // Allow - because Java strings may have surrogate pairs so we allow them in unicode escapes if paired.
         testParse("SELECT * { ?s ?p '\\uD801\\uDC37'}");
 
 //        QueryParseException ex = assertThrows(QueryParseException.class,  ()->testParse("SELECT * { ?s ?p '\\uD801\\uDC37'}"));

Original file line number	Diff line number	Diff line change
`@@ -93,8 +93,8 @@ String VersionSpecifier() : { Token t ; String version ; }`
`93`	`93`	`\| t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }`
`94`	`94`	`)`
`95`	`95`	`{`
`96`		`- checkString(version, t.beginLine, t.beginColumn) ;`
`97`	`96`	`version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`97`	`+ checkRDFString(version, t.beginLine, t.beginColumn) ;`
`98`	`98`	`return version;`
`99`	`99`	`}`
`100`	`100`	`}`
`@@ -1933,7 +1933,7 @@ String String() : { Token t ; String lex ; }`
`1933`	`1933`	`\| t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }`
`1934`	`1934`	`)`
`1935`	`1935`	`{ lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;`
`1936`		`- checkString(lex, t.beginLine, t.beginColumn) ;`
	`1936`	`+ checkRDFString(lex, t.beginLine, t.beginColumn) ;`
`1937`	`1937`	`return lex ;`
`1938`	`1938`	`}`
`1939`	`1939`	`}`
Original file line number	Diff line number	Diff line change
`@@ -188,8 +188,8 @@ String VersionSpecifier() : { Token t ; String version ; }`
`188`	`188`	`\| t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }`
`189`	`189`	`)`
`190`	`190`	`{`
`191`		`- checkString(version, t.beginLine, t.beginColumn) ;`
`192`	`191`	`version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`192`	`+ checkRDFString(version, t.beginLine, t.beginColumn) ;`
`193`	`193`	`return version;`
`194`	`194`	`}`
`195`	`195`	`}`
`@@ -2631,7 +2631,7 @@ String String() : { Token t ; String lex ; }`
`2631`	`2631`	`\| t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }`
`2632`	`2632`	`)`
`2633`	`2633`	`{ lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;`
`2634`		`- checkString(lex, t.beginLine, t.beginColumn) ;`
	`2634`	`+ checkRDFString(lex, t.beginLine, t.beginColumn) ;`
`2635`	`2635`	`return lex ;`
`2636`	`2636`	`}`
`2637`	`2637`	`}`
Original file line number	Diff line number	Diff line change
`@@ -83,8 +83,8 @@ String VersionSpecifier() : { Token t ; String version ; }`
`83`	`83`	`\| t = <STRING_LITERAL2> { version = stripQuotes(t.image) ; }`
`84`	`84`	`)`
`85`	`85`	`{`
`86`		`- checkString(version, t.beginLine, t.beginColumn) ;`
`87`	`86`	`version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`87`	`+ checkRDFString(version, t.beginLine, t.beginColumn) ;`
`88`	`88`	`return version;`
`89`	`89`	`}`
`90`	`90`	`}`
`@@ -1646,7 +1646,7 @@ String String() : { Token t ; String lex ; }`
`1646`	`1646`	`\| t = <STRING_LITERAL_LONG2> { lex = stripQuotes3(t.image) ; }`
`1647`	`1647`	`)`
`1648`	`1648`	`{ lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;`
`1649`		`- checkString(lex, t.beginLine, t.beginColumn) ;`
	`1649`	`+ checkRDFString(lex, t.beginLine, t.beginColumn) ;`
`1650`	`1650`	`return lex ;`
`1651`	`1651`	`}`
`1652`	`1652`	`}`
Original file line number	Diff line number	Diff line change
`@@ -149,8 +149,8 @@ final public void Prologue() throws ParseException {`
`149`	`149`	`jj_consume_token(-1);`
`150`	`150`	`throw new ParseException();`
`151`	`151`	`}`
`152`		`-checkString(version, t.beginLine, t.beginColumn) ;`
`153`		`- version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`152`	`+version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`153`	`+ checkRDFString(version, t.beginLine, t.beginColumn) ;`
`154`	`154`	`{if ("" != null) return version;}`
`155`	`155`	`throw new Error("Missing return statement in function");`
`156`	`156`	`}`
`@@ -7520,7 +7520,7 @@ final public Node BooleanLiteral() throws ParseException {`
`7520`	`7520`	`throw new ParseException();`
`7521`	`7521`	`}`
`7522`	`7522`	`lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;`
`7523`		`- checkString(lex, t.beginLine, t.beginColumn) ;`
	`7523`	`+ checkRDFString(lex, t.beginLine, t.beginColumn) ;`
`7524`	`7524`	`{if ("" != null) return lex ;}`
`7525`	`7525`	`throw new Error("Missing return statement in function");`
`7526`	`7526`	`}`
Original file line number	Diff line number	Diff line change
`@@ -128,8 +128,8 @@ final public void Prologue() throws ParseException {`
`128`	`128`	`jj_consume_token(-1);`
`129`	`129`	`throw new ParseException();`
`130`	`130`	`}`
`131`		`-checkString(version, t.beginLine, t.beginColumn) ;`
`132`		`- version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`131`	`+version = unescapeStr(version, t.beginLine, t.beginColumn) ;`
	`132`	`+ checkRDFString(version, t.beginLine, t.beginColumn) ;`
`133`	`133`	`{if ("" != null) return version;}`
`134`	`134`	`throw new Error("Missing return statement in function");`
`135`	`135`	`}`
`@@ -5889,7 +5889,7 @@ final public Node BooleanLiteral() throws ParseException {`
`5889`	`5889`	`throw new ParseException();`
`5890`	`5890`	`}`
`5891`	`5891`	`lex = unescapeStr(lex, t.beginLine, t.beginColumn) ;`
`5892`		`- checkString(lex, t.beginLine, t.beginColumn) ;`
	`5892`	`+ checkRDFString(lex, t.beginLine, t.beginColumn) ;`
`5893`	`5893`	`{if ("" != null) return lex ;}`
`5894`	`5894`	`throw new Error("Missing return statement in function");`
`5895`	`5895`	`}`