Rename a file according to a filter

eMule has a nice little feature that lets you automatically rename files you click on. This is useful to replace '.' with a space, or delete information you don't want on there like PROPER or the group name. Basically so it's more readable. Any chance of getting a similar feature in uTorrent?

can be more readable, but then to seed torrents back can be messy

-1

Wouldn't the torrent be according to the file hash though? So the file name doesn't matter?

for single-file torrents MAYBE, but not multi-file.

for multi-file, you could only rename the top-level folder, really.

i however would like to see a feature to be able to reneme files in directory torretns and still be able to seed than. With Root directory (single) files with the "set download location" procedure it is possible.

Having this on mulitfiles possible would be great.

Example: you have a multifiletorrent like this:

"Great CC content show S01E01 unnessesary bla bla"

"Great CC content show S01E02 unnessesary bla bla"

"Great CC content show S01E03 unnessesary bla bla"

...

When i rename the files on the harddrive to get rid of the unnessesary bla bla µtorrent can not seed that torrent anymore.

The renaning thru "set download location" can't handle THIS situation but the single file situation works perfectly.

Would be great if this will be possible in further releases to keep stuff in the seeding queue.

Yes, would also be nice to have predefined variables such as e.g. %SEASON% and %EPISODE% with which you can convert

Title.S01E99.PROPER.HDTV.AVI

to

Title - 01x99.avi

Most sense would it make in the RSS-Filters...

Thanks,

Defender

to do this UT would have to maintain a database of torrent-file-name -> current-windows-file-name mapping.. that's quite a bit of work, you know!

But how come eMule can do it on the fly? It's just about what you save the file as, not the hash or source file name.

eMule only downloads single files, and hashes with eMule are not related to the filename at all. Torrents are multiple files, and the infohash does include the filename when being calculated.

So you cant change file names? Damn never knew that :-/

This is well worth looking into. One reason for this is to sanitize file/folder names to fit various filesystems without a lot of clumsy escapes (this becomes far more of an issue when, as I do, you are dealing with files with multiple operating systems. I deal with them in Windows, Linux, and Cygwin, which is enormously useful, but which brings this problem very much front and center).

Even more, when filenames aren't UTF-8 (or can't be made to be), things get really hairy. I've had some interesting tangles with Japanese files, for instance.

If it's a real problem to approach this with some sort of mapping database between arbitrary filenames and torrents, perhaps something conceptually similar to URL escapes, that is, expressing characters outside the confines of operating system filesystems with %XX, where XX is a character's hexadecimal value. There are other similar approaches, but that's the basic idea, to be able to convert any UTF-8 character to something that itself consists of only characters that do (in this case, UTF-7, even).

Rulesets for various OS filesystems could be created and user-adjustable in the Advanced setup.

It might be necessary to get more creative for charsets other than these (Japanese, Cyrillic, other ISOs), but I'm sure it can be done. I'd be happy to throw in what I could in the way of some code for the purpose, at least for a starting-point.