Hi Damjan, Damjan Jovanovic wrote:
> As one of the perpetrators of the problem, I have now fixed it. The > reasons I swallowed exceptions were simple: [snip] > * when an exception is thrown and close() then throws another > exception, the close() exception is propagated and the original > exception - which could reveal the cause of the problem - swallowed [snip] And this *is* a real problem. And it is exactly why the IOException of close() are normally ignored. While I don't like IOUtils.closeQietly(closeable), I use normally a method "IO.closeLogged(closeable)" resp. "IO.closeLogged(closeable, logger)". If e.g. an image is corrupted and later on an additional exception occurs closing any resource, you will simply *never* learn about the corrupted image that caused the problem in first case. The original exception must never be swallowed! - Jörg --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
