DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG� RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://issues.apache.org/bugzilla/show_bug.cgi?id=34592>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND� INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=34592 ------- Additional Comments From [EMAIL PROTECTED] 2005-05-06 11:11 ------- VFS do not force to user to use "System.gc()". This was just to debug this problem. The times it took up VFS it uses a cache which never ever released a file - a memory eater, so I implemented the SoftRefFilesCache. I am still convinced that this sort of cache is good and could work, though I am no longer sure VFS should use a cache in its core system. One problem I constantly have with the cache (any sort of) is that it reports stale data (due to filesytem changes form external processes) and I am forced to constantly call ".close()" to refresh this info. Currently I think about removing the cache, so every VFS.getManager().resolveFile("anyfile.txt") will return a NEW FileObject (like new File("anyfile.txt") will create a new instance), every call to fo.getContent().getSize() will lead to a call to the used library (even if in case of FTP this will be a performance penalty) That way we could have VFS as small as possible to abstract the different libraries (something like commons-logging is for the different logging apis) To reimplement the cache we could try to decorate the file-object with something like a CachedFileObject which will do the caching with all its pros and cons. The biggest con for this could be that it is no longer possible to synchronize on a fileObject if resolved independent. But hey, who will do this? And it will delay the move out of sandbox again as this is a huge change. Any comments? -- Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
