flickrfs-users

[Flickrfs-users] no sets

[Guldo K <gu...@ti...>]

Hello,
I have just installed flickrfs 1.3.9.1 on debian (unstable) with 
libfuse2 2.7.1 and python 2.4.4.
The uploading of files works, with tags and renaming too, but each time 
cp returns an error:

$ cp fireboard.png /mnt/flickr/sets/fooset/
`fireboard.png' -> `/mnt/flickr/sets/fooset/fireboard.png'
cp: scrittura di `/mnt/flickr/sets/fooset/fireboard.png': Invalid argument

anyway, the file is uploaded.

But, the set looks empty.
A "ls /mnt/flickr/sets/fooset/" returns nothing.
If I unmount and mount back, the sets dir is empty.
If I try to download the file, I get a "No such file or directory" error 
by cp.
If I create another "fooset" dir inside sets dir, a new set with the 
same name is created in flickr.

Why?
It looks like it cannot collect my existing sets info from flickr, but I 
can see no errors in the log file.

Thanks.

-- 
Guldo - Powered by debian sid
www.giapponegiappone.it

Re: [Flickrfs-users] no sets

[Manish R. J. <man...@gm...>]

A lot of people have complained the problem with sets. I'd like to fix that
bug, but I don't have any time to spend upon it. With all the work load, its
difficult for me to maintain the project.

I'd be more than happy if someone would like to take over flickrfs, and
maintain it. Its an interesting project, and allows you to understand
filesystem internals, and flickr apis. I'm still here, and I can provide all
the help that's needed for someone to understand the code base. There's
plenty of functionality that can be added, so its more than just bug fixing.

-Manish


On Jan 22, 2008 1:48 PM, Guldo K <gu...@ti...> wrote:

> Hello,
> I have just installed flickrfs 1.3.9.1 on debian (unstable) with
> libfuse2 2.7.1 and python 2.4.4.
> The uploading of files works, with tags and renaming too, but each time
> cp returns an error:
>
> $ cp fireboard.png /mnt/flickr/sets/fooset/
> `fireboard.png' -> `/mnt/flickr/sets/fooset/fireboard.png'
> cp: scrittura di `/mnt/flickr/sets/fooset/fireboard.png': Invalid argument
>
> anyway, the file is uploaded.
>
> But, the set looks empty.
> A "ls /mnt/flickr/sets/fooset/" returns nothing.
> If I unmount and mount back, the sets dir is empty.
> If I try to download the file, I get a "No such file or directory" error
> by cp.
> If I create another "fooset" dir inside sets dir, a new set with the
> same name is created in flickr.
>
> Why?
> It looks like it cannot collect my existing sets info from flickr, but I
> can see no errors in the log file.
>
> Thanks.
>
> --
> Guldo - Powered by debian sid
> www.giapponegiappone.it
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Flickrfs-users mailing list
> Fli...@li...
> https://lists.sourceforge.net/lists/listinfo/flickrfs-users
>

Re: [Flickrfs-users] no sets

[Liu T. <fli...@ya...>]

Hello Manish Rai Jain,
   
  i have the same problem as many other people...
   
  /set/fooset_xy/ is visible but empty (no pictures).
   
  (everything worked like a charm with an clean account but when it fills up the /set/fooset1 /set/fooset2 etc... are empty)
  I also get "Invalid argument" when i copy files...
   
  So i hope you will find some time to fix your programm - i like it very much!!!
  Also you produce DFO i think this is not what i want..
  i want to copy and update my local /Photos/subdirectory with a simple cp * comand...
   
  I´m not able to program myself so I really hope you can fix this ...
   
  Thanx alot!
  Thomas
   
  

Manish Rai Jain <man...@gm...> wrote:
  A lot of people have complained the problem with sets. I'd like to fix that bug, but I don't have any time to spend upon it. With all the work load, its difficult for me to maintain the project.

I'd be more than happy if someone would like to take over flickrfs, and maintain it. Its an interesting project, and allows you to understand filesystem internals, and flickr apis. I'm still here, and I can provide all the help that's needed for someone to understand the code base. There's plenty of functionality that can be added, so its more than just bug fixing.

-Manish


  On Jan 22, 2008 1:48 PM, Guldo K <gu...@ti...> wrote:
  Hello,
I have just installed flickrfs 1.3.9.1 on debian (unstable) with
libfuse2 2.7.1 and python 2.4.4.
The uploading of files works, with tags and renaming too, but each time
cp returns an error:

$ cp fireboard.png /mnt/flickr/sets/fooset/
`fireboard.png' -> `/mnt/flickr/sets/fooset/fireboard.png'
cp: scrittura di `/mnt/flickr/sets/fooset/fireboard.png': Invalid argument

anyway, the file is uploaded.

But, the set looks empty.
A "ls /mnt/flickr/sets/fooset/" returns nothing.
If I unmount and mount back, the sets dir is empty.
If I try to download the file, I get a "No such file or directory" error
by cp.
If I create another "fooset" dir inside sets dir, a new set with the
same name is created in flickr.

Why?
It looks like it cannot collect my existing sets info from flickr, but I
can see no errors in the log file.

Thanks.

--
Guldo - Powered by debian sid
www.giapponegiappone.it

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Flickrfs-users mailing list
Fli...@li...
https://lists.sourceforge.net/lists/listinfo/flickrfs-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/_______________________________________________
Flickrfs-users mailing list
Fli...@li...
https://lists.sourceforge.net/lists/listinfo/flickrfs-users


       
---------------------------------
Looking for last minute shopping deals?  Find them fast with Yahoo! Search.

Re: [Flickrfs-users] no sets

[<rdm...@bi...>]

On Tue, 22 Jan 2008 at 16:34, Manish Rai Jain wrote:
> I'd be more than happy if someone would like to take over flickrfs, and
> maintain it. Its an interesting project, and allows you to understand
> filesystem internals, and flickr apis. I'm still here, and I can provide all
> the help that's needed for someone to understand the code base. There's
> plenty of functionality that can be added, so its more than just bug fixing.

I haven't really used flickrfs since I did the refactoring work almost
two years ago now.  My interests went in other directions.  However,
I find that I'm once again interested in flickrfs and python-fuse,
and I have some time I could spend on this.

So, I'd be interested in taking over maintenance, Manish.  I'm still a
developer on the sourceforge project.  I'm working on getting the CVS
version running on my own machine, at the moment...

--David

Re: [Flickrfs-users] no sets

[Manish R. J. <man...@gm...>]

Sounds great. Let me know what help you need from me.

-Manish


On Jan 26, 2008 11:14 AM, <rdm...@bi...> wrote:

> On Tue, 22 Jan 2008 at 16:34, Manish Rai Jain wrote:
> > I'd be more than happy if someone would like to take over flickrfs, and
> > maintain it. Its an interesting project, and allows you to understand
> > filesystem internals, and flickr apis. I'm still here, and I can provide
> all
> > the help that's needed for someone to understand the code base. There's
> > plenty of functionality that can be added, so its more than just bug
> fixing.
>
> I haven't really used flickrfs since I did the refactoring work almost
> two years ago now.  My interests went in other directions.  However,
> I find that I'm once again interested in flickrfs and python-fuse,
> and I have some time I could spend on this.
>
> So, I'd be interested in taking over maintenance, Manish.  I'm still a
> developer on the sourceforge project.  I'm working on getting the CVS
> version running on my own machine, at the moment...
>
> --David
>

[Flickrfs-users] fuse-python version

[<rdm...@bi...>]

On Sat, 26 Jan 2008 at 11:32, Manish Rai Jain wrote:
> Sounds great. Let me know what help you need from me.

OK, so flickrfs was written against fuse-python 0.1, right?  Setting
fuse.fuse_python_api to (0, 1) gets flickrfs to run against fuse-python
2.0 (you can do this without touching the code by setting the environment
variable FUSE_PYTHON_API to '0.1'), but it doesn't work right, at least on
my system.  It appears that flickrfs is doing things to the filesystem
before it is mounted.  That is, the mount happens when sever.main()
is called, but directories are created and threads are launched in the
__init__, before main() is called.  This causes flickrfs to hang on my
system some time after the mount is done.

Are others successfully running flickrfs using fuse-python 0.2?
If you are running 0.1, are you having the problem with the empty
'sets' directory?

The 'new api' page for fuse-python says that this is where the mount
took place in 0.1 as well (in main()), but I'm wondering if maybe that
isn't in fact true, and under 0.1 the mount happened in the __init__.
The problem is that I can't get fuse-python 0.1 to compile on my system,
so I can't test it :(

I'm going to fiddle with the flickrfs initialization and see if I can get
it running on my system.  The question is, is anybody using fuse-python
0.1 who can't or doesn't want to upgrade?  My thought is that we should
eventually migrate flickrfs to the 0.2 API, so if I need instead to
maintain backward compatibility I'd better know now :)

--David

Re: [Flickrfs-users] fuse-python version

[<rdm...@bi...>]

On Sat, 26 Jan 2008 at 17:06, rdm...@bi... wrote:
> my system.  It appears that flickrfs is doing things to the filesystem
> before it is mounted.  That is, the mount happens when sever.main()
> is called, but directories are created and threads are launched in the
> __init__, before main() is called.  This causes flickrfs to hang on my
> system some time after the mount is done.

Hah, like I said, it is taking me a bit to get back up to speed on this
code.  This initialization sequence is not a problem since it is our
model of the file system we are updating, and that is available before
the mount.  So why doing it before the mount causes later commands to
hang for me is mysterious.  If I do the fs/thread stuff after the mount,
then flickrfs doesn't hang for me, but an 'ls' on my mount point says
there's no such file...and df shows all question marks.  So I'm poking
around trying to figure out what's going on here....

--David