myGBPVR: use XMBC as GBPVR FrontEnd

[marcuswar]

Martyn,

Thanks for figuring this out for us.. but what do I do with the AES.py and Setup.py files ?

I tried simply copying them into the libs folder and replacing the _AESEncrypt function GBPVR_Connect.py but it still doesn't work for me
GBPvr script says "unable to connect to GBPVr"
XBMC.log says that it can't import aes.

Help

Marcus

Hey Ton,

I found a pure Python module that does AES encryption and have been attempting to get myGBPVR to work on my original XBoxes with XBMC.

I've got as far as (I think) correctly encrypting the credentials and passing them in the requests, but I'm getting errors from GBPVR with regards to "Padding", so I'm guessing something isn't quite right.

Would you be willing to take a look at that for me? Or even take a look at using the Python module instead of using PyCrypto?

The module is at http://code.google.com/p/slowaes/source/...unk/python

My (poor) code to replace the _AESEncrypt routine is:

Code:

def _AESEncrypt(self, clear_bytes, key_bytes, iv_bytes):

    import aes

    try:        

        mode = aes.AESModeOfOperation.modeOfOperation["CBC"]
        
        key_bytes = map(ord, key_bytes)
        key_size = len(key_bytes)        

        clear_bytes = aes.append_PKCS7_padding(clear_bytes)
        
        iv_bytes = map(ord, iv_bytes)

        aes_cypher = aes.AESModeOfOperation()        
        (mode, orig_len, encrypted) = aes_cypher.encrypt(clear_bytes,mode,key_bytes,key_size,iv_bytes)

        return ''.join(map(chr, encrypted))

    except:
        import traceback
        traceback.print_exc()
        return

Cheers,

Martyn

[Ton]

Hi,

Martyn,

Thanks for figuring this out for us.. but what do I do with the AES.py and Setup.py files ?

I tried simply copying them into the libs folder and replacing the _AESEncrypt function GBPVR_Connect.py but it still doesn't work for me
GBPvr script says "unable to connect to GBPVr"
XBMC.log says that it can't import aes.

Help

Marcus

It looks like you have to copy the "aes.py" file into the "resources/src" directory. That should allow the script to find the file....

I'll try it myself a little bit later, just to see if it works, and post back to you...

[martynw]

Hi,



It looks like you have to copy the "aes.py" file into the "resources/src" directory. That should allow the script to find the file....

I'll try it myself a little bit later, just to see if it works, and post back to you...

Hi Ton / Marcus,

That's correct, just grab the aes.py file and put it into the "resources/src" directory. You can put it in the libs folder if you wish and adjust the "import" line to take that into account.


Cheers,

Martyn

[carpeVideo]

XBMC lets you do 1 of 2 things with an NFO file 1) override all the information. If you have an nfo with nothing but XML then it will not scrape and use whatever is in the NFO - so if the NFO only has a name and description your show or movie will only have a name and description but no art. 2) override a few pieces of info and let it scrape the rest.

To do 2 you need a mix of XML and the URL -

<movie>
<genre clear="true">Action</genre>
<genre>Horror</genre>
<genre>Thriller</genre>
<director clear="true">Frank Miller</director>
<director>Robert Rodriguez</director>
<director>Quentin Tarantino</director>
</movie>
http://www.imdb.com/title/tt0401792

I requested one change that would be helpful over at the XBMC forums: the ability to have an NFO file without a URL and have it still scrape the rest. Plus if the NFO has a name use that name for lookup rather than the filenake. That way if the movie was Superman from 2007 we could simply write and NFO like this:

<movie>
<name>Superman (2007)</name>
<scrape>true<scrape>
</movie>

And it would look up the rest including the art and correctly lookup the right movie or series based on the improved name. They told me to write my own patch :o/

[pBS]

lol...
so very strange the dir structure they decided to implement..doesn't seem to make sense for anything really..
like why separate with a dir when it's already separated by the filename..

so we have to put a scraper url in the file? then xbmc isn't really scraping is it? lol
i'll stick with my already enhanced data and thumbs i got from gbpvr..
thanx for the help...really working well now...
so when the asterisk gets removed i'll be able to play videos! lol
[right now, only about 1% of my shows aren't first runs, so all have a *]

anyone know how to regex out the " *" from filenames just before the play?
[i don't know squat about Python]

[carpeVideo]

You don't need the url to get scrapes, only if you have an NFO, but it is an odd constraint. As to the directories it is not required that they are separated that its only an example of one structure that could be used.

To get rid of the star I used :
smbPath = smbPath.replace("*","")

[Ton]

Hi,

I've got it setup as follows (for tvshows):
1) I have a "tvshow.nfo" file that has an URL to the TVshow database (to get the art and logo's). I put that down manually (it's only once for each show, wasn't that big a deal for me)..
2) I then have "NFO" files for each recording, that contain just XML. I (personally) don't really care about the exact episode numbers / season numbers. All I care about is getting the right meta-data (title, description) into XBMC.

That said, here in Europe we're stuck with crappy EPG data, so all the cool things you get in the US (first showing, repeat, season-number, episode numbers etc.) are not part of the EPG data (so it's almost impossible to match). A lot of shows don't even come with correct title information, so there's no way to automatically match them to real season/episode numbers.

What are you trying to achieve? Is it just the meta-data in the files (I think the method I described above will do that), or do you want exact matching...

I don't use GBPVR for recording any movies, when I watch those, I prefer better quality then analog TV...

One thing I wanted to experiment with (but haven't gotten round to) is to use "shortcuts" to the actual recording files, in the name format that XBMC expects. That way you could leave the original recordings in place, and there would be no need to copy any files. You could do this just for tvshows (recurring recordings)...

You don't need the url to get scrapes, only if you have an NFO, but it is an odd constraint. As to the directories it is not required that they are separated that its only an example of one structure that could be used.

To get rid of the star I used :
smbPath = smbPath.replace("*","")

[Ton]

Hi Martyn / Marcus,

Hi Ton / Marcus,

That's correct, just grab the aes.py file and put it into the "resources/src" directory. You can put it in the libs folder if you wish and adjust the "import" line to take that into account.


Cheers,

Martyn

Just tried it (copied the aes.py in the resrources/src directory and modifying the function in GBPVR_Connect.py), and it works like a charm. It's a little slower (on my laptop) then using PyCrypto, but not all that much....

[pBS]

cool...starting to get the gist of it..
btw, i had to use
smbPath = smbPath.replace(" " + "*","")
to get the space in there also..
now it plays all my recordings... sweet...that makes it usable finally..

i think i'll stick with the regular file mode and the built-in info..
i already generate my own thumbs so that's ok..
is there a way to specify a different thumb than one named same as file, in .nfo?

basically i just want the gb info to show up in library mode...i see thumbs but no info unless click on 'episode info'
or does it work that way? [i thought that's what the .nfo did]
or just thumbs in mygbpvr interface..what's supposed to be in bottom of details box? i wonder if that needs the star removed to find the thumg?

sorry to be such a pain with this part..it is kinda funky..

[marcuswar]

Martyn/Ton

Thanks for pointing out the obvious for me ... I've now copied the aes folder into the src folder and am getting further. Now I'm getting an error about the padding being invalid :confused:

Code:

18:57:47 M: 22249472  NOTICE:
18:57:50 M: 20283392  NOTICE: TimeString used: 11-05-2010 18:57:00
18:57:50 M: 20283392  NOTICE:
18:57:52 M: 20172800  NOTICE: (webServiceAuthentication){
                                R = "3rl/z9LNb2SddDhrUaeTDoSmV7ckQnZZbmXJ/958fUZIXHYCkn3BCVzhc/vz0bx1ad8fx8qVKruX8WtVPxOhaQ==ad8fx8qVKruX8WtVPxOhaQ==0000bd8bb1398e60915a371f4e3f420baa48"
                                RL = "qrLv6MCz+ljj+8C9FLp8VQ=="
                                RTL = "KvM3O70GDKD9Ci6JMl07hg=="
                                UserName = None
                                Password = None
                              }
18:57:52 M: 20230144  NOTICE:
18:57:52 M: 20230144  NOTICE: No handlers could be found for logger "suds.client"
18:57:52 M: 20209664  NOTICE: EXCEPTION: : Traceback (most recent call last):
                               File "Q:\scripts\myGBPVR\resources\src\gbpvr\home.py", line 152, in refreshOnInit
                                 self.statusData = self.gbpvr.GetGBPVRInfo(self.settings.GBPVR_USER, self.settings.GBPVR_PW)
                               File "Q:\scripts\myGBPVR\resources\src\GBPVR_Connect.py", line 139, in GetGBPVRInfo
                                 self.channels = self.getChannelList(userid, password)
                               File "Q:\scripts\myGBPVR\resources\src\GBPVR_Connect.py", line 162, in getChannelList
                                 ret_soap = client.service.getChannelListObject(soapheaders=authObj)
                             WebFault: Server raised fault: 'Server was unable to process request. ---> Padding is invalid and cannot be removed.'
18:57:52 M: 20209664  NOTICE:

I've looked at the logs for EWA (from the statistics page) but it makes no mention of even an attempt to connect ? Is there somewhere to change the logging level for EWA?

Marcus

Hi Martyn / Marcus,



Just tried it (copied the aes.py in the resrources/src directory and modifying the function in GBPVR_Connect.py), and it works like a charm. It's a little slower (on my laptop) then using PyCrypto, but not all that much....