adds unicode braced escaping and tests

src/main/antlr/GraphqlCommon.g4

@@ -125,7 +125,7 @@ fragment StringCharacter:
 '\\' EscapedCharacter;
 
 fragment EscapedCharacter :  ["\\/bfnrt];
-fragment EscapedUnicode : Hex Hex Hex Hex;
+fragment EscapedUnicode : Hex Hex Hex Hex | '{' Hex+ '}';
 fragment Hex : [0-9a-fA-F];
 
 

src/main/java/graphql/parser/StringValueParsing.java

@@ -3,6 +3,7 @@
 import graphql.Assert;
 import graphql.Internal;
 
+import java.io.IOException;
 import java.io.StringWriter;
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -30,7 +31,9 @@ public static String removeIndentation(String rawValue) {
         String[] lines = rawValue.split("\\n");
         Integer commonIndent = null;
         for (int i = 0; i < lines.length; i++) {
-            if (i == 0) continue;
+            if (i == 0) {
+                continue;
+            }
             String line = lines[i];
             int length = line.length();
             int indent = leadingWhitespace(line);
@@ -44,7 +47,9 @@ public static String removeIndentation(String rawValue) {
         if (commonIndent != null) {
             for (int i = 0; i < lineList.size(); i++) {
                 String line = lineList.get(i);
-                if (i == 0) continue;
+                if (i == 0) {
+                    continue;
+                }
                 if (line.length() > commonIndent) {
                     line = line.substring(commonIndent);
                     lineList.set(i, line);
@@ -135,15 +140,37 @@ public static String parseSingleQuotedString(String string) {
                     writer.write('\t');
                     continue;
                 case 'u':
-                    String hexStr = string.substring(i + 1, i + 5);
-                    int codepoint = Integer.parseInt(hexStr, 16);
-                    i += 4;
-                    writer.write(codepoint);
+                    i = parseEscapedUnicode(writer, string, i);
                     continue;
                 default:
                     Assert.assertShouldNeverHappen();
             }
         }
         return writer.toString();
     }
+
+    private static int parseEscapedUnicode(StringWriter writer, String string, int i) {
+        if (string.charAt(i + 1) != '{') {
+            String hexStr = string.substring(i + 1, i + 5);
+            int codepoint = Integer.parseInt(hexStr, 16);
+            writer.write(codepoint);
+            return i + 4;
+        }
+        // this means we have a braced escape which allows code points outside of the BMP: e.g. '\\u{1F37A}'
+        int startIx = i + 2;
+        int endIndexExclusive = startIx;
+        do {
+            if (endIndexExclusive + 1 >= string.length()) {
+                throw new RuntimeException("invalid unicode encoding");
+            }
+        } while (string.charAt(++endIndexExclusive) != '}');
+        String hexStr = string.substring(startIx, endIndexExclusive);
+        char[] chars = Character.toChars(Integer.parseInt(hexStr, 16));
+        try {
+            writer.write(chars);
+        } catch (IOException e) {
+            return Assert.assertShouldNeverHappen();
+        }
+        return endIndexExclusive;
+    }
 }

src/test/groovy/graphql/parser/ParserTest.groovy

@@ -39,6 +39,7 @@ import graphql.language.VariableDefinition
 import graphql.language.VariableReference
 import org.antlr.v4.runtime.CommonTokenStream
 import org.antlr.v4.runtime.ParserRuleContext
+import org.junit.Ignore
 import spock.lang.Specification
 import spock.lang.Unroll
 
@@ -814,6 +815,76 @@ triple3 : """edge cases \\""" "" " \\"" \\" edge cases"""
         operationDefinition.getComments()[0].content == " Represents the 😕 emoji."
     }
 
+    def "allow braced escaped unicode"() {
+        def input = '''
+              {
+              foo(arg: "\\u{1F37A}")
+               }
+    '''
+        when:
+        Document document = Parser.parse(input)
+        OperationDefinition operationDefinition = (document.definitions[0] as OperationDefinition)
+        def field = operationDefinition.getSelectionSet().getSelections()[0] as Field
+        def argValue = field.arguments[0].value as StringValue
+
+        then:
+        argValue.getValue() == "🍺"
+
+    }
+
+    def "allow surrogate pairs escaped unicode"() {
+        def input = '''
+              {
+              foo(arg: "\\ud83c\\udf7a")
+               }
+    '''
+        when:
+        Document document = Parser.parse(input)
+        OperationDefinition operationDefinition = (document.definitions[0] as OperationDefinition)
+        def field = operationDefinition.getSelectionSet().getSelections()[0] as Field
+        def argValue = field.arguments[0].value as StringValue
+
+        then:
+        argValue.getValue() == "🍺"
+
+    }
+
+    @Ignore
+    def "invalid surrogate pair"() {
+        def input = '''
+              {
+              foo(arg: "\\uD83D\\uDBFF")
+               }
+    '''
+        when:
+        Document document = Parser.parse(input)
+        OperationDefinition operationDefinition = (document.definitions[0] as OperationDefinition)
+        def field = operationDefinition.getSelectionSet().getSelections()[0] as Field
+        def argValue = field.arguments[0].value as StringValue
+
+        then:
+        argValue.getValue() == "🍺"
+
+    }
+
+    def "invalid unicode code point"() {
+        def input = '''
+              {
+              foo(arg: "\\u{fffffff}")
+               }
+    '''
+        when:
+        Document document = Parser.parse(input)
+        OperationDefinition operationDefinition = (document.definitions[0] as OperationDefinition)
+        def field = operationDefinition.getSelectionSet().getSelections()[0] as Field
+        def argValue = field.arguments[0].value as StringValue
+
+        then:
+        argValue.getValue() == "🍺"
+
+    }
+
+
     def "can override antlr to ast"() {
 
         def query = '''
@@ -897,4 +968,5 @@ triple3 : """edge cases \\""" "" " \\"" \\" edge cases"""
         '1.23.4'  | _
         '1.2e3e'  | _
     }
+
 }

src/test/groovy/graphql/parser/StringValueParsingTest.groovy

@@ -40,7 +40,7 @@ class StringValueParsingTest extends Specification {
         parsed == '''"'''
     }
 
-    def "parsing emoji should work"() {
+    def "parsing beer mug as surrogate pair"() {
         // needs surrogate pairs for this emoji
         given:
         def input = '''"\\ud83c\\udf7a"'''
@@ -52,15 +52,39 @@ class StringValueParsingTest extends Specification {
         parsed == '''🍺''' // contains the beer icon 	U+1F37A  : http://www.charbase.com/1f37a-unicode-beer-mug
     }
 
+    def "parsing beer mug as escaped unicode"() {
+        // needs surrogate pairs for this emoji
+        given:
+        def input = '''"\\u{1F37A} hello"'''
+
+        when:
+        String parsed = StringValueParsing.parseSingleQuotedString(input)
+
+        then:
+        parsed == '''🍺 hello''' // contains the beer icon 	U+1F37A  : http://www.charbase.com/1f37a-unicode-beer-mug
+    }
+
+    def "parsing beer mug non escaped"() {
+        // needs surrogate pairs for this emoji
+        given:
+        def input = '''"🍺 hello"'''
+
+        when:
+        String parsed = StringValueParsing.parseSingleQuotedString(input)
+
+        then:
+        parsed == '''🍺 hello''' // contains the beer icon 	U+1F37A  : http://www.charbase.com/1f37a-unicode-beer-mug
+    }
+
     def "parsing simple unicode should work"() {
         given:
-        def input = '''"\\u56fe"'''
+        def input = '''"\\u56fe hello"'''
 
         when:
         String parsed = StringValueParsing.parseSingleQuotedString(input)
 
         then:
-        parsed == '''图'''
+        parsed == '''图 hello'''
     }