T10 MTP...for Linux ?

[SonuvBob]

Hm.. that quick fix (setting PTP_USB_BULK_HS_MAX_ PACKET_LEN back to 512) brings back other problems. Hard to keep track of all the issues. Uploads are fixed, but downloads are still buggy.

[ricky715]

hey i'm new here and i got the t30 to work without installing any new software. I was using Fedora Core 5. So far i have to log into root to get this to connect to the t30 so can anyone tell me how i can make it so i dont have to log into rot every time i want to upload?

[SonuvBob]

Originally Posted by ricky715

hey i'm new here and i got the t30 to work without installing any new software. I was using Fedora Core 5. So far i have to log into root to get this to connect to the t30 so can anyone tell me how i can make it so i dont have to log into rot every time i want to upload?

http://www.gphoto.org/doc/manual/permissions-usb.html

[darrengood]

OrthoB posted a python file on the first page of this thread to convert a *.m3u file to a *.pla file for the MTP version of the T10. If you have the UMS version then the script still works great but you have to change the line:

fo.write (struct.pack('>i508s ',n,"iriver MTP PLA" ))
to
fo.write (struct.pack('>i508s ',n,"iriver UMS PLA" ))

... just in case someone was curious but hadn't figured it out yet.

I spent ages last night compiling & installing gphoto2 & trying to get it working with my T10 before it occurred to me that I might not have to... Once I established that I had the UMS version & could simply mount it like a usb disk, then all I had to do is figure out the previous bit about the play lists and everything was very happy. If only I hadn't try to mount it incorrectly in the first place I could have saved many hours of frustration when I assumed it was an MTP version.

How do you know if you have a UMS or MTP version? ... goto:

http://www.iriver.com/html/support/f...pSearchString=

On the page:

http://www.iriver.com/html/support/d...ew.asp?idx=732

there is the latest firmware for the UMS version of the T10 ... not sure if this is required to use playlists but I installed it anyway & everyhing seemed to go ok.

dg

[orthoB]

Originally Posted by darrengood

OrthoB posted a python file on the first page of this thread to convert a *.m3u file to a *.pla file for the MTP version of the T10. If you have the UMS version then the script still works great but you have to change the line:

fo.write (struct.pack('>i508s ',n,"iriver MTP PLA" ))
to
fo.write (struct.pack('>i508s ',n,"iriver UMS PLA" ))

... just in case someone was curious but hadn't figured it out yet.
dg

Thanks! I wrote the script and didn't realize that! I was wondering why my playlist wasn't working after I converted my T10 to UMS last week; It had been a while since I wrote that and I didn't recall the signature line. I was on the verge of posting to ask for a sample UMS playlist.

Thanks again!

[darrengood]

No worries ... only reason I knew was because I made a "quicklist" on the player & then looked at it in a Hex editor & saw the UMS in place of the MTP.

Thanks for making the script in the first place.

[cdsteinkuehler]

I've tried using the script on my T30 (UMS firmware 1.71), and it didn't like the playlists created.

I made a "quick-play" list directly on the T30, which DID work...comparing the lists it looks like the T30 requires a "playlist name" in plain ascii (zero terminated, not unicode) starting at offset 0x30. Once I added this field to the playlist generated by the python script, above, I was successfully able to use the playlist on my T30.

I have not yet updated the python script (I used hexedit to make the change), but when I do, I'll post the mods here (unless someone beats me to it!).

Thanks for the initial script, and all the helpful info in this thread!!!

[blablub]

hy,

running a t10 with firmware 1.70p under linux (ums not mtp). orthoB's python scripts (here: txt2pla.py) work like a charm, except ... when having non-ascii characters in the playlist. then i get an error similar to this:

UnicodeDecodeError: 'ascii' codec can't decode byte 0xff in position 1: ordinal not in range(12

tried to write the playlist.txt either with utf-8 or utf-16, both with no luck.
any hints other than not using non-ascii characters?

[blablub]

has nobody an idea?

[onlyonce]

Try this (line 14):

name = "\\"+t[i].rstrip().replace("/", "\\").decode('iso-8859-15')

Adjust the charset as needed, iso-8859-15 is western european with the € sign.

[blablub]

nope, does not work. no it complains about line 16 (before, it was line 17):

  File "./txt2pla.py", line 16, in ?
    print i,": ",name
UnicodeEncodeError: 'ascii' codec can't encode characters in position 34-35: ordinal not in range(128)

tried it both with 'iso-8859-15' and 'utf-8' - with no success.

[onlyonce]

just uncomment that line or add ".encode('iso-8859-15')" (use your terminal's charset). The background is, with encode() you are transforming text from a given charset into Python's own internal representation. When you're trying to output it, you have to transform it to a real charset again, using decode().

[blablub]

ahh, fabulous works! thanks

[_kuroyume_]

Originally Posted by cdsteinkuehler

I've tried using the script on my T30 (UMS firmware 1.71), and it didn't like the playlists created.

I made a "quick-play" list directly on the T30, which DID work...comparing the lists it looks like the T30 requires a "playlist name" in plain ascii (zero terminated, not unicode) starting at offset 0x30. Once I added this field to the playlist generated by the python script, above, I was successfully able to use the playlist on my T30.

I have not yet updated the python script (I used hexedit to make the change), but when I do, I'll post the mods here (unless someone beats me to it!).

Thanks for the initial script, and all the helpful info in this thread!!!

any news on this? please?

[AugerMcGunn]

orthoB, thank you for saving me who knows how many hours trying to make playlists in Linux. Your python script worked with just the minor edit of switching MTP to UMS. I and my new T30 thank you. If we're ever in the same drinking establishment, you're drinking for free that night. Darren, I owe you at least one also.

kuroyume, I don't know what the issue is that you may be having, but the txt2pla.py script worked fine for my T30. I took a couple .m3u files, fixed the path to point to the right folders in the T30 (i.e. annie_lennox/medusa/train_in_vain.ogg to Rock/train_in_vain.ogg), ran './txt2pla.py medusa.txt Medusa.pla', and then copied Medusa.pla to T30/Playlists. I then browsed to the file, selected it, and hit the play button. It was a bit of a pleasant surprise to see the little PL next to the file name.

[_kuroyume_]

mmm.;.. i haven't really tried it, since cdsteinkuehler said that it didn't work for him, and that there were some modifications needed... however, if you say it works, i'll give it a spin... thanks for the report...

[polrrr]

Could somewith repost txt2pla.py? There don't appear to be links in OrthoB's original post and I lost my copy with a machine crash.

I found that for my UMS T20 I had to patch txt2pla.py to calculate strings lengths or terminate strings properly. There were some cases where the original script wasn't working for me. Sorry I forgotten the fine detail.

[orthoB]

I'll try to do this from home tonight.

[orthoB]

Here you go. This time it's an actual .zip file. Note that if you have a UMS player, change MTP to UMS in the txt2pla.py script.

Another trick - files are played in inode order. If you have a UMS device you can check the playback order using:
ls -i1 | sort
If the order isn't what you want, mv the files to another folder on the device so the folder is empty, and mv (or drag using some filebrowser bui) them back one by one in the desired playback order.

[polrrr]

[quote=orthoB;560984]Here you go. This time it's an actual .zip file. Note that if you have a UMS player, change MTP to UMS in the txt2pla.py script.
QUOTE]

Thank you. If I need to do the string length trick again I'll post the details.

[polrrr]

Functionally txt2play.py works fine and my T20 reads the playlist
without problem - thanks!

The problem I see with pla2txt.py is incorrectly terminated strings displayed.
With a .pla file created on Windows using the iRiver software
pla2txt.py shows extra garbage on screen (using RH EL 4.5, python
2.3.4):

id = iriver UMS PLA^@^@^@ [repeated 494 times in total - really messy]
1 : Corrina, Corrina.mp3 \DylanMisc\
2 : Hurricane.mp3^@na.mp 3 \DylanMisc\
3 : I Want You.mp3^@a.mp3 \DylanMisc\
4 : Just Like a Woman.mp3 \DylanMisc\
5 : Mr. Tambourine Man.mp3 \DylanMisc\
6 : Desolation Row.mp3^@mp3 \DylanMisc\
7 : Oh, sister.mp3^@mp3^@mp3 \DylanMisc\
8 : Sad Eyed Lady of the Lowlands.mp3 \DylanMisc\
9 : Sara.mp3^@Lady of the Lowlands.mp3 \DylanMisc\
10 : Stuck Inside of Mobile with the Memphis Blues Again.mp3 \DylanMisc\
11 : Visions of Johanna.mp3^@with the Memphis Blues Again.mp3 \DylanMisc\
12 : Talking World War III Blues.mp3^@Memphis Blues Again.mp3 \DylanMisc\
13 : A Hard Rain's A-Gonna Fall.mp3^@^@Memphis Blues Again.mp3 \DylanMisc\
. . .

Where ^@ is a nul byte in the .pla file according to my hex dump.

Even if I create the .pla file with txt2pla.py I still get the
"corrupt" id line even though nulls don't appear in the song titles.

I haven't patched this yet - need to remind myself how to do it in
Python.