On 30/06/17 22:06, Christopher Schultz wrote: > Mark, > > (This time, to the list and not just you) > > This seems like the wrong approach. > > I think you want to leave the URLEncoding alone and instead XML-escape > anything you intend to put into an XML document separately. > > You want the XML to end up looking like this: > > <element>url.which.contains.a.&.symbol</element> > > and not this: > > <element>url.which.contains.a.%26.symbol</element> > > Or am I misunderstanding what's going on, here?
Arguably, & is the more correct but either approach is legal as far as the specifications are concerned. Since we need to URL encode anyway, fixing this via tweaking the URL encoding looked to be simpler. It was also consistent with how it was working before the regression was introduced. Mark --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
