Ok, I put in a test case that illustrates the problem and I fixed it. I'm still thinking about that null check thingy and I think it's a good idea.
-----Original Message----- From: James Carman [mailto:[EMAIL PROTECTED] Sent: Monday, May 29, 2006 5:43 PM To: 'Tapestry users' Subject: RE: Tapernate/squeezer chain - squeezing null bug Ted, Great catch! For now, I'll just put in a null check in the Tapernate squeezer. Thanks, James -----Original Message----- From: Ted Steen [mailto:[EMAIL PROTECTED] Sent: Monday, May 29, 2006 1:56 PM To: Tapestry users Subject: Re: Tapernate/squeezer chain - squeezing null bug Here's the stack trace if anyone is interested; http://paste.uni.cc/9180 2006/5/29, Ted Steen <[EMAIL PROTECTED]>: > When squeezing an object that is null, > EntitySqueezerFilter.squeeze(EntitySqueezerFilter.java:57) throws NPE. > It doesn't matter if it is an entity or not, because the object is > passed through the chain, and eventually ends up in the > EntitySqueezerFilter. > > A solution to this problem would be to either check for null in the > squeeze method, or maybe let some kind of "null squeezer" (that > squeezes the object to a single character, like X) be the first > DataSqueezerFilter in the chain. > > If the new squeezing chain is going into the Tapestry 4.1 API, maybe > placing the "null squeezer" first in the chain would be a good idea as > it eliminates the danger of getting NPEs in other, potentially buggy > 3rd party squeezers implementations. > > ted > -- ted --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
