| Summary: | Response.setCharacterEncoding() fails after Response.getWriter() and Response.reset() | ||
|---|---|---|---|
| Product: | Tomcat 5 | Reporter: | Olaf Meske <bugs> |
| Component: | Catalina | Assignee: | Tomcat Developers Mailing List <dev> |
| Status: | RESOLVED FIXED | ||
| Severity: | major | CC: | Udo.Walker |
| Priority: | P2 | ||
| Version: | 5.0.23 | ||
| Target Milestone: | --- | ||
| Hardware: | All | ||
| OS: | All | ||
This has been fixed in trunk and proposed for back port to 5.5.x and 6.0.x *** Bug 37072 has been marked as a duplicate of this bug. *** Fixed in 6.0.x Fixed in 5.5.x and will be included in 5.5.26 onwards. |
The class org.apache.catalina.connector.Response resets the internal hold coyoteResponse and outputBuffer during reset(): public void reset() { if (included) return; // Ignore any call from an included servlet coyoteResponse.reset(); outputBuffer.reset(); } but if anyone has already set the characterEncoding (e.g. to UTF-8) and already got the writer, than all following invocations of request.setCharacterEncoding() will do nothing, because the internal save flag for usingWriter is not reseted. This is very annoying, because the coyoteResponse.reset(), resets the characterEncoding to iso8859-1. So I have no change to change/reset the charaterEncoding to UTF-8. I'll think if you reset the response, you should also reset the usingWriter flag or add another special flag, that works with this circumstances.