http://gwt-code-reviews.appspot.com/1449814/diff/5/user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java
File user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java
(right):
http://gwt-code-reviews.appspot.com/1449814/diff/5/user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java#newcode65
user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java:65:
public static boolean isValidUriCharset(String uri) {
On 2011/06/09 15:16:51, xtof wrote:
Can you make this package private?
I think this might not work with super sourced classes though...
I tried package private but, as you guessed, it triggers some kind of
bug (well, I believe it's a bug) in the DevMode that throws an
IllegalAccessException.
http://gwt-code-reviews.appspot.com/1449814/diff/5/user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java#newcode136
user/src/com/google/gwt/safehtml/shared/SafeUriHostedModeUtils.java:136:
uri = UriUtils.encodeAllowEscapes(uri);
On 2011/06/09 15:16:51, xtof wrote:
On 2011/06/09 14:32:41, tbroyer wrote:
> This is a "breaking change", as it now allows "%foo" ("%" not
followed by
> 2-hex-digits) where it previously didn't.
>
> Given that we allow otherwise-erroneous unescaped unicode chars
(along with a
> few others, such as unescaped "<" and ">"), and that browsers allow
"%foo", I
> think it's a change for the better.
I think that's fine. And it wouldn't actually break anything since it
loosens
validation (I assume that's why you put "breaking change" in quotes?)
Technically, it could break one's test that asserted "%foo" is invalid.
In practice, it probably affects nobody.
http://gwt-code-reviews.appspot.com/1449814/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
