Some characters cause FileBot to stop job?

[ByteJunk]

Heya!

I caught an exception on my logs:
"InvalidPathException: Malformed input or input contains unmappable chacraters: <snip>"

The character in question is "ä", snipped the full error output above for privacy.

Filebot continues gracefully (kudos on catching it hehe), and sends me a mail saying it renamed all the files in the folder, although it hasn't, it stopped iterating the files after the exception. Not sure if it's a java thing or a filesytem thing, it was trying to copy/rename the file to a NTFS partition.

Might be relevant:

Code: Select all

$uname -a
Linux xbmc-pc 3.5.0-23-generic #35-Ubuntu SMP Thu Jan 24 13:05:29 UTC 2013 i686 i686 i686 GNU/Linux
$filebot -version
FileBot 3.3 (r1412) / OpenJDK Runtime Environment 1.7.0_09 (headless)

[rednoah]

No idea. "ä" works fine on NTFS. I'd guess filesystem thing so nothing I can do about it. Email? The utorrent-postprocess script just processes everything it can. Set --conflict fail if you want it to abort on error. Also I'm pretty sure that email only contains the files that have been renamed, not the ones that failed renaming.

[ByteJunk]

The email isn't a problem, just goes to show that filebot doesn't crash. Can you reproduce it trying to rename something that has "ä" not in the original filename, but in the renamed name? For instance, NCIS.s02e12.avi (should be renamed "Doppelgänger")? I can create the filename manually, doesn't seem like a filesystem issue.

I assumed it was a filebot's path sanity check that was too restrictive.

[rednoah]

Nope, that InvalidPathException is coming from somewhere deep down in the Java runtime when trying to interface with the filesystem.

Works on Windows:

Code: Select all

E:\testdata>filebot -rename NCIS.s02e12.avi
Filename pattern: [1,00] SxE, [0,00] CWS
Rename episodes using [TheTVDB]
Auto-detected query: [NCIS]
Fetching episode data for [NCIS]
[MOVE] Rename [E:\testdata\NCIS.s02e12.avi] to [NCIS - 2x12 - Doppelgänger.avi]
Processed 1 files
Done ?(?????)?

[rednoah]

Found this:
http://code.google.com/p/opendedup/issu ... l?id=34#c1

Yep, I guess it's the issue as with some of the embedded linux systems that don't set up locale and filesystem character encoding. Same as this in the FAQ:

Q: I'm running FileBot on a Linux machine and non-ASCII characters get all messed up. Why do unicode characters not work?
A: On some machines the locale is not set up. You'll need to tell Java what charset filenames are encoded with by setting the environment variable LANG.

Code: Select all

export LANG=en_US.utf8

What machine is that? With locale not being set up can't be a standard Ubuntu install can it?