RawTextComparator.WS_IGNORE_CHANGE must not compare whitespace
Only the presence or absence of whitespace is significant; but not the
actual whitespace characters. Don't compare whitespace bytes.
Compare the C git implementation at [1].
[1] https://github.com/git/git/blob/0d0e1e8/xdiff/xutils.c#L173
Bug: 563570
Change-Id: I2d0522b637ba6b5c8b911b3376a9df5daa9d4c27
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
diff --git a/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/RawTextIgnoreWhitespaceChangeTest.java b/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/RawTextIgnoreWhitespaceChangeTest.java
index b271a04..73b2a72 100644
--- a/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/RawTextIgnoreWhitespaceChangeTest.java
+++ b/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/RawTextIgnoreWhitespaceChangeTest.java
@@ -75,4 +75,32 @@ public void testEqualsWithWhitespace() {
assertTrue(cmp.equals(a, 5, b, 5));
assertTrue(cmp.equals(b, 5, a, 5));
}
+
+ @Test
+ public void testEqualsWithTabs() {
+ RawText a = new RawText(
+ Constants.encodeASCII("a\tb\t \na\tb\t c \n foo\na b\na b"));
+ RawText b = new RawText(
+ Constants.encodeASCII("a b \na b c\n\tfoo\nab\na \tb"));
+
+ // "a\tb\t \n" == "a b \n"
+ assertTrue(cmp.equals(a, 0, b, 0));
+ assertTrue(cmp.equals(b, 0, a, 0));
+
+ // "a\tb\t c \n" == "a b c\n"
+ assertTrue(cmp.equals(a, 1, b, 1));
+ assertTrue(cmp.equals(b, 1, a, 1));
+
+ // " foo" == "\tfoo"
+ assertTrue(cmp.equals(a, 2, b, 2));
+ assertTrue(cmp.equals(b, 2, a, 2));
+
+ // "a b" != "ab"
+ assertFalse(cmp.equals(a, 3, b, 3));
+ assertFalse(cmp.equals(b, 3, a, 3));
+
+ // "a b" == "a \tb "
+ assertTrue(cmp.equals(a, 4, b, 4));
+ assertTrue(cmp.equals(b, 4, a, 4));
+ }
}
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/diff/RawTextComparator.java b/org.eclipse.jgit/src/org/eclipse/jgit/diff/RawTextComparator.java
index 508d07c..0c41b85 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/diff/RawTextComparator.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/diff/RawTextComparator.java
@@ -191,21 +191,15 @@ public boolean equals(RawText a, int ai, RawText b, int bi) {
be = trimTrailingWhitespace(b.content, bs, be);
while (as < ae && bs < be) {
- byte ac = a.content[as];
- byte bc = b.content[bs];
+ byte ac = a.content[as++];
+ byte bc = b.content[bs++];
- if (ac != bc)
- return false;
-
- if (isWhitespace(ac))
+ if (isWhitespace(ac) && isWhitespace(bc)) {
as = trimLeadingWhitespace(a.content, as, ae);
- else
- as++;
-
- if (isWhitespace(bc))
bs = trimLeadingWhitespace(b.content, bs, be);
- else
- bs++;
+ } else if (ac != bc) {
+ return false;
+ }
}
return as == ae && bs == be;
}
@@ -215,12 +209,12 @@ protected int hashRegion(byte[] raw, int ptr, int end) {
int hash = 5381;
end = trimTrailingWhitespace(raw, ptr, end);
while (ptr < end) {
- byte c = raw[ptr];
- hash = ((hash << 5) + hash) + (c & 0xff);
- if (isWhitespace(c))
+ byte c = raw[ptr++];
+ if (isWhitespace(c)) {
ptr = trimLeadingWhitespace(raw, ptr, end);
- else
- ptr++;
+ c = ' ';
+ }
+ hash = ((hash << 5) + hash) + (c & 0xff);
}
return hash;
}
