> On 2011-12-05 19:15:25, Brian Lillie wrote: > > In doFetchConcatResources, if one of the 2nd through nth requests > > encounters an error such that false would be returned, we still have the > > content from the 1st up to the failing request in the > > VerbatimConcatOutputStream, and when we do the close on that stream, it > > will be written out as well as the the 400 error code. Seems like the > > cases where that might occur are probably remote, and having extra data in > > the response shouldn't matter. Not sure that it needs to be fixed .. just > > pointing it out. > > Ryan Baxter wrote: > Yeah I thought of that too Brian. I think setting the error code when we > return should let be an indication to ignore the content in the response. > Jesse would you agree?
Yup - I saw that too and thought it was odd but that's the way it worked before your changes so I'm fine with leaving it as is. At least with your changes in place we should get back proper 400 codes consistently even if the previously buffered content is large enough to fill the response buffer. - Jesse ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/3006/#review3628 ----------------------------------------------------------- On 2011-12-04 21:04:16, Ryan Baxter wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/3006/ > ----------------------------------------------------------- > > (Updated 2011-12-04 21:04:16) > > > Review request for shindig, Jesse Ciancetta and Brian Lillie. > > > Summary > ------- > > After we return from doFetchConcatResources(..) we set the status in doGet. > So lets say we return false from doFetchConcatResources and the content > written to the response was bigger than the buffer. This means we will flush > the buffer and set the status to the OK status then write the rest of the > content and set the status to bad request, but it will be to late we already > set the status to OK > > > This addresses bug SHINDIG-1667. > https://issues.apache.org/jira/browse/SHINDIG-1667 > > > Diffs > ----- > > > http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/main/java/org/apache/shindig/gadgets/servlet/ConcatProxyServlet.java > 1210178 > > Diff: https://reviews.apache.org/r/3006/diff > > > Testing > ------- > > Ran unit tests and rendered gadgets in common container > > > Thanks, > > Ryan > >
