Need testers - updated 10-1

[golgoj4]

the guesses don't work with me, but apart from that, its all peachy :-) Good job, buddy.

can you provide a few filenames? i want to try and improve the regex's a bit. Im currently trying to write a c++ based version that does this automatically.

-golgoj4.

[maverick0815]

Sorry for the delay...I had a little less time at my hands for a while, but here you got some filenames:
Star.Trek.Enterprise.2x02.Carbon.Creek.avi
Paradox.1x01.avi
Two.And.A.Half.Men.2x03.A.Bag.Full.Of.Jawea.avi

just a few examples.

[golgoj4]

Sorry for the delay...I had a little less time at my hands for a while, but here you got some filenames:
Star.Trek.Enterprise.2x02.Carbon.Creek.avi
Paradox.1x01.avi
Two.And.A.Half.Men.2x03.A.Bag.Full.Of.Jawea.avi

just a few examples.

cool thanks!
im posting my progress here http://wiki.linuxmce.org/index.php/MetaData_Handling#Automatic_File_tagging

[jthodges]

cool thanks!
im posting my progress here http://wiki.linuxmce.org/index.php/MetaData_Handling#Automatic_File_tagging

I extended UpdateMedia to execute a set of rules for new and existing media files quite a while ago.  Reading that wiki entry, I think that what you're doing could just plug into that.  Details are in this thread: http://forum.linuxmce.org/index.php?topic=7116.0.  Trak ticket: http://svn.linuxmce.org/trac.cgi/ticket/158.

It's sat for a long time because I was told no new features would be considered until 0810 was released.  I'm currently setting up a new build environment and will make sure the patch still works with the latest source.  Then I was planning to see if I can generate any more interest in it by making it easier to add to a live system, create some more useful rules, etc.  I'd be really interested in discussing how we could plug your automatic component in to this system as a rule or a set of rules.

[golgoj4]

cool thanks!
im posting my progress here http://wiki.linuxmce.org/index.php/MetaData_Handling#Automatic_File_tagging

I extended UpdateMedia to execute a set of rules for new and existing media files quite a while ago.  Reading that wiki entry, I think that what you're doing could just plug into that.  Details are in this thread: http://forum.linuxmce.org/index.php?topic=7116.0.  Trak ticket: http://svn.linuxmce.org/trac.cgi/ticket/158.

It's sat for a long time because I was told no new features would be considered until 0810 was released.  I'm currently setting up a new build environment and will make sure the patch still works with the latest source.  Then I was planning to see if I can generate any more interest in it by making it easier to add to a live system, create some more useful rules, etc.  I'd be really interested in discussing how we could plug your automatic component in to this system as a rule or a set of rules.

My approach is slightly different in that it doesn't really have anything to do with rules save that the file is video and named a certain way. I considered a directory based approach, but thinking from an appliance standpoint i want interested in having people follow specific directory structure AND filename convention. Basically my goal is that anywhere linuxmce finds a video file that meets the criteria, and tags it based solely on how its named. I also intend for it to hit multiple sources for data so that I can fill out the metadata on things as much as possible.

Rules would still be usefull I think in helping users who wanted to establish how their files are organized if they choose to do it that way. But i still have much work to do so once i get a working proto-type ill have better idea.

-golgoj4

[jthodges]

I'm not sure I explained it very well, or I'm not understanding your goals correctly.  I'm not sure where the description of it as a directory-based approach comes from, but it's not necessarily directory based.  You can attach rules to a directory, but you can also put them at the top level and just execute the rule based on a regular expression match.  It's really just an engine wired into UpdateMedia that executes custom logic against media that it matches via regular expression. Isn't that what you need? 

In your case I was thinking a rule for each data source (e.g., "IMDBRule", "TVDB"), which you could apply at the top level (or lower if folks did happen to have a tv directory they wanted to limit it to) for an array of regular expressions to match the various naming schemes you wanted to support.  Any media matching would cause your rule to execute. 

Am I way off base here?  It still seems like a good fit to me.

[golgoj4]

I'm not sure I explained it very well, or I'm not understanding your goals correctly.  I'm not sure where the description of it as a directory-based approach comes from, but it's not necessarily directory based.  You can attach rules to a directory, but you can also put them at the top level and just execute the rule based on a regular expression match.  It's really just an engine wired into UpdateMedia that executes custom logic against media that it matches via regular expression. Isn't that what you need?  

In your case I was thinking a rule for each data source (e.g., "IMDBRule", "TVDB"), which you could apply at the top level (or lower if folks did happen to have a tv directory they wanted to limit it to) for an array of regular expressions to match the various naming schemes you wanted to support.  Any media matching would cause your rule to execute.  

Am I way off base here?  It still seems like a good fit to me.

jthodges: i've made quite a bit of progress and ill want to see if i can get a hold of you on irc later this week. Im at the stage of writing the sql code to insert stuff, so we are approaching that point of convergence

I wasn't really sure of the best way to make the code available for others to look at / make suggestions / improve, so i signed up for this gitorious business as I wasn't interested in bothering the devs until its further along. this should let us collaborate a little easier.

http://gitorious.org/linuxmce-autotagger
http://linuxmce.pastebin.com/vvm6DvEZ <--console output, basically to debug and see its determinations.

the code atm will compile and run (its built on qt). Throw some video files in 'home/public/data/videos/' and it will scan and give you output in the console. this is basically a lot of debug info to let me know what its identified and so on to help refine th initial process. its to the point where it retrieves the data online (for tv atm) but doesn't insert it yet. I figured this was a good point to start sharing my progress.

-golgoj4

[jthodges]

Fantastic!  I've had work stuff taking over the last week or so unfortunately, so I still need to dig out and test that patch.  I'm hoping to be freed up in the evenings starting tomorrow, and I'll make that a first priority to support any integration discussions we have this week.  Also looking forward to checking out what you've uploaded... thanks!

[golgoj4]

Continuing the topic over here: http://forum.linuxmce.org/index.php?topic=10884.0

Long story short, have a basic stand alone version.

Look like ill need to setup a builder to compile the necessary changes to updateMedia.

In the mean time if you want to test it out, go ahead. i would just recommend using a small set of test files. Details are on the last commit.

-golgoj4

[jeangot]

Hello,

without the autotagger being integrated into UpdateMedia, what is the best way to have it run automatically on new files added to the system? If I would set a cron to run autotagger once daily, would it retag previously tagged files, or would it be smart enough to only look at new files? (Assuming I'm passing directories to it, and not individual filenames)

Jean

[golgoj4]

Hello,

without the autotagger being integrated into UpdateMedia, what is the best way to have it run automatically on new files added to the system? If I would set a cron to run autotagger once daily, would it retag previously tagged files, or would it be smart enough to only look at new files? (Assuming I'm passing directories to it, and not individual filenames)

Jean

Im currently coding changes to it so that it will run in the background, checking for new media files to tag. As far as existing stuff, it checks to see if they are missing anything or if there is more info to add, but it does this pretty quickly so a re-scan of existing stuff doesnt take very long at all.

As far as it being 'smart enough' to skip already tagged media. I made a choice for it not to specifically because cast lists expand, synopis becomes more filled out, etc.

So yes, run your cron job, i dont see that it would cause any problems

-golgoj4