Re: [Flickrfs-users] Version 1.1. released
Brought to you by:
manishrjain
From: Paul P. <pe...@ee...> - 2005-11-10 06:53:00
|
On Thu, Nov 10, 2005 at 01:45:28PM +0800, Manish Rai Jain wrote: > > which corresponds to the last photo in the last set. Not having much > > luck with debugging so far, but then again I don't know much about FUSE. > > Any debugging suggestions you have would be very helpful. > > > Some checks: > 1. ls -al /dev/fuse -> does it show a character device? (Though, this shd be > fine, otherwise flickrfs won't mount) crw-rw-rw- 1 root root 10, 229 Nov 8 00:43 /dev/fuse > 2. Try out flickr account of your friend, on the same installation, and see > if the prob remains. You'll need to remove ~/.flickr directory containing > the authentication information, so that flickrfs will show the login prompt > again. I registered a new account and uploaded a couple of photos. Same problem. However, I did uncover another bug: if the user has no sets, then the main for loop in sets_thread fails because rsp.photosets[0] has no 'photoset' attr. I guess you can fix it with something like if hasattr(rsp.photosets[0], "photoset"): for a in rsp.photosets[0].photoset: ... > 3. Add some debugging statements to sets_thread, and see if the thread > returns before the error shows up. sets_thread does return. However, I'm not sure whether the "short read" error occurs before or after sets_thread returns. The error message doesn't show up until I hit enter. If I don't touch the keyboard then the flickrfs process just sits there doing nothing after iterating through all the tags and sets (and the mount point remains empty), so I assume the error has already occurred even though the message is not printed until I hit enter. Must be some weird multithreading thing. Paul |