8324665: Loose matching of space separators in the lenient date/time parsing mode

Reviewed-by: joehw, jlu
This commit is contained in:
Naoto Sato 2024-02-06 17:43:12 +00:00
parent 2d252ee06e
commit 96eb0390d6
4 changed files with 177 additions and 6 deletions

View file

@ -1,5 +1,5 @@
/*
* Copyright (c) 1996, 2023, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 1996, 2024, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
@ -746,6 +746,10 @@ public abstract class DateFormat extends Format {
* <p>This leniency value is overwritten by a call to {@link
* #setCalendar(java.util.Calendar) setCalendar()}.
*
* @implSpec A {@link Character#SPACE_SEPARATOR SPACE_SEPARATOR} in the input
* text will match any other {@link Character#SPACE_SEPARATOR SPACE_SEPARATOR}s
* in the pattern with lenient parsing; otherwise, it will not match.
*
* @param lenient when {@code true}, parsing is lenient
* @see java.util.Calendar#setLenient(boolean)
*/

View file

@ -1487,7 +1487,8 @@ public class SimpleDateFormat extends DateFormat {
switch (tag) {
case TAG_QUOTE_ASCII_CHAR:
if (start >= textLength || text.charAt(start) != (char)count) {
if (start >= textLength ||
!charEquals(text.charAt(start), (char)count)) {
pos.index = oldStart;
pos.errorIndex = start;
return null;
@ -1497,7 +1498,8 @@ public class SimpleDateFormat extends DateFormat {
case TAG_QUOTE_CHARS:
while (count-- > 0) {
if (start >= textLength || text.charAt(start) != compiledPattern[i++]) {
if (start >= textLength ||
!charEquals(text.charAt(start), compiledPattern[i++])) {
pos.index = oldStart;
pos.errorIndex = start;
return null;
@ -1580,6 +1582,13 @@ public class SimpleDateFormat extends DateFormat {
return parsedDate;
}
private boolean charEquals(char ch1, char ch2) {
return ch1 == ch2 ||
isLenient() &&
Character.getType(ch1) == Character.SPACE_SEPARATOR &&
Character.getType(ch2) == Character.SPACE_SEPARATOR;
}
/* If the next tag/pattern is a <Numeric_Field> then the parser
* should consider the count of digits while parsing the contiguous digits
* for the current tag/pattern