Merge branch 'lexer-recover-eof' of github.com:sharwell/antlr4

This commit is contained in:
Terence Parr 2012-11-04 09:35:28 -08:00
commit c073835186
2 changed files with 49 additions and 6 deletions

View File

@ -134,17 +134,17 @@ public abstract class Lexer extends Recognizer<Integer, LexerATNSimulator>
throw new IllegalStateException("nextToken requires a non-null input stream.");
}
if (_hitEOF) {
emitEOF();
return _token;
}
// Mark start location in char stream so unbuffered streams are
// guaranteed at least have text of current token
int tokenStartMarker = _input.mark();
try{
outer:
while (true) {
if (_hitEOF) {
emitEOF();
return _token;
}
_token = null;
_channel = Token.DEFAULT_CHANNEL;
_tokenStartCharIndex = _input.index();
@ -369,7 +369,10 @@ public abstract class Lexer extends Recognizer<Integer, LexerATNSimulator>
}
public void recover(LexerNoViableAltException e) {
getInterpreter().consume(_input); // skip a char and try again
if (_input.LA(1) != IntStream.EOF) {
// skip a char and try again
getInterpreter().consume(_input);
}
}
public void notifyListeners(LexerNoViableAltException e) {

View File

@ -46,6 +46,46 @@ public class TestLexerErrors extends BaseTest {
assertEquals(expectingError, error);
}
@Test
public void tesStringsEmbeddedInActions() {
String grammar =
"lexer grammar Actions;\n"
+ "ACTION2 : '[' (STRING | ~'\"')*? ']';\n"
+ "STRING : '\"' ('\\\"' | .)*? '\"';\n"
+ "WS : [ \\t\\r\\n]+ -> skip;\n";
String tokens = execLexer("Actions.g4", grammar, "Actions", "[\"foo\"]");
String expectingTokens =
"[@0,0:6='[\"foo\"]',<1>,1:0]\n" +
"[@1,7:6='<EOF>',<-1>,1:7]\n";
assertEquals(expectingTokens, tokens);
assertNull(stderrDuringParse);
tokens = execLexer("Actions.g4", grammar, "Actions", "[\"foo]");
expectingTokens =
"[@0,6:5='<EOF>',<-1>,1:6]\n";
assertEquals(expectingTokens, tokens);
assertEquals("line 1:0 token recognition error at: '[\"foo]'\n", stderrDuringParse);
}
@Test public void testEnforcedGreedyNestedBrances() {
String grammar =
"lexer grammar R;\n"
+ "ACTION : '{' (ACTION | ~[{}])* '}';\n"
+ "WS : [ \\r\\n\\t]+ -> skip;\n";
String tokens = execLexer("R.g4", grammar, "R", "{ { } }");
String expectingTokens =
"[@0,0:6='{ { } }',<1>,1:0]\n" +
"[@1,7:6='<EOF>',<-1>,1:7]\n";
assertEquals(expectingTokens, tokens);
assertEquals(null, stderrDuringParse);
tokens = execLexer("R.g4", grammar, "R", "{ { }");
expectingTokens =
"[@0,5:4='<EOF>',<-1>,1:5]\n";
assertEquals(expectingTokens, tokens);
assertEquals("line 1:0 token recognition error at: '{ { }'\n", stderrDuringParse);
}
@Test public void testInvalidCharAtStartAfterDFACache() throws Exception {
String grammar =
"lexer grammar L;\n" +