mirror of
https://github.com/internetarchive/heritrix3.git
synced 2026-10-07 12:51:21 +00:00
Fix HER-1792 backslashes in urls can in some cases really mean backslashes
(specifically, when they're in the query string)
* UURIFactory.java
fixup() - replace \ with / up to the first ? in the url, instead of
anywhere in the url
and update comments
* UURIFactoryTest.java
backslash test
This commit is contained in:
1 parent
2be8af10a9
commit
b064d43db5
2 files changed
+32
-14
No files matched your search
@@ -1166,4 +1166,17 @@ public class UURIFactoryTest extends TestCase {
|
||||
base.resolve(relative);
|
||||
UURIFactory.getInstance(base,relative);
|
||||
}
|
||||
|
||||
/**
|
||||
* To match IE behavior, backslashes in path-info (really, anywhere before
|
||||
* query string) assumed to be slashes, to match IE behavior. In
|
||||
* query-string, they are escaped to %5C.
|
||||
*
|
||||
* @throws URIException
|
||||
*/
|
||||
public void testBackslashes() throws URIException {
|
||||
UURI uuri = UURIFactory.getInstance("http:\\/www.example.com\\a/b\\c/d?q\\r\\|s/t\\v");
|
||||
String expected = "http://www.example.com/a/b/c/d?q%5Cr%5C|s/t%5Cv";
|
||||
assertEquals(expected, uuri.toString());
|
||||
}
|
||||
}
|
||||
Reference in new issue
Block a user