btw I refuse to accept acknowledgement for Piglet's Big Movie, that's for my little sister but was on a disc of kids movies I added.

whilst on teh subject, i refuse to watch these modern winnie the pooh programmes, i can't believe they caved in to the crappy style of animating these days. I caught a glimpse of that new advert for the new one, and it looked mightly crappy....


anyway, that's an aside

Nah, it fits nicely right in between Cheerleader Autopsy and Krasue.

But in all seriousness, that looks like a neat program. Can't wait to see how it'll turn out.

lol, well i hope its functionality meets my plans <s>for world domination</s>.

@John, you know how i suggested Edonkey hashing as a feature in EMM ages ago?

I see your md5, EMM, and I raise you ed2k:


That's thanks to a library from GNU and a library from a program called JackSum

Wow! Looks like it's really coming along!!!

It's getting me more and more excited every day.



hehe, what about movies that acutally has a year in the title? Like '2019 - After The Fall Of New York (1983)'

I like the first GUI the most btw.


Might be nice with a 'personal rating' field, where you can give a value from 1-10 or something, instead of the EMM 'Seen' checkbox.

you have a rating box, you think there should be two? one for imdb and one for your personal choice?

reading avi info maybe quite difficult to wrap my head round but i've been reading up and if i use the JMF libraries i can take random or predefined screenshots.

It just grabbing avi info efficiently thats a problem.

An update for those that care,

Playing with/In the system
Still thinking about fields, and where information should be stored... I mean seen? Should that be stored in media? I guess it is, since you can then do a filter, and say what movies have i not seen, and then use the media to check the edition, and then the movies (reverse tree navigation). And likewise can see what editions you haven't seen of a movie, and then ultimately, which media hasn't been seen.

Importing, already can import basic movie information from an EMM output, so I guess that's just lots of lines of code to do each kind of information. Lots of parsing.

I'm using a very sweet filtering method, so basically you can filter pretty much any bit of info stored in the database to any degree. I need to find tune it and implement regular expressions.

Screenshot coding, i don't think that's going to be a problem, I just need to sit down and read the APIs for the JMF.

Hashing, perfect. Using jacksum libraries so its pretty much nothing to do but use their hash, and make some links.

Interfaces, they shouldn't even exist at this stage, all the ones I've been showing are all 'just a bit of fun'. They will of course look a lot neater than this rough stuff.

Media reading; having trouble with this. I really need some kind of exact white paper that tells me the header structures. Can't find one that isn't at a very high level of abstraction. Something that says, this byte here will contain the codec information, this byte has the resolution data, etc. I know its there in the header, just not the exact way of getting at it.

Planned, but a long a way away (not at the stage to implement yet)

Intelligent filename parsing, Title.year.info.dvdrip.xvid-group , you already have all the info you need there, beyond comments you wish to make about it. I mean that's the sole purpose of that structure (that i personally wish everyone would use, or at least omit info and group if they wish)

Searching of imdb, its parsing. I just need to come up with some plugin options. It might be a case of writing a script will be:


So it would just search for "/title/"> and then everything after that to the start of the end, will be read as the title. Nice and simple but there's always complications. The HHAH app just does the same thing, downloads a page, looks through the source code for what it wants.

Intelligent searching of imdb is an aim too, you should be able to specify title and then an approximate year... say you might put star wars, and then enter 1970s, and it would say, right order the results favouring any star wars films in the late 1970s. This would be tied in with the filename reading. If you get title, year; why should you have to double click an entry when the file already states the movie and the year.

Suggesting is another nice thing, if the computer can do the work, then it really should be, if it see 29.9fps, suggest possible nuke flag.

Linking with the nuke flag, you will have an export; you have these films, but these versions need replacing, so you can print out your list, but also a 'but if i can get a better version of this then try'. I end up with rips of the same films, and normally because they are repeatedly shitty or I just don't know I have them.

Another way off but feature that really sparked this whole thing, and will be linked with an emule modification, custom stop lists! Like a known file, but a 'hey you already have a movie that has this title and year! it is not marked as nuke, do you wish to download this movie again?". That will save me so much dvdrs and bandwidth **

Collections is already on its was, very simple to do (thx John), and so its a logical step to allow outputing in the fh collection manager format. I'll have to have a think about export types

Not sure about
Some of the guis, look and feel.

User control, lots of features, but I'm thinking its important for people to omit them if they choose to not take advise by the program (might be good to have newbie ripping advise mode, which will explain why a movie is messed up? + the yes/no nuke with reason).

Recommendation system, an advanced pick a movie, but more one that's based on mood and ratings. So if you rate highly zombie movies, and kung fu movies, then select your in the mood for a horror movie it would pick something like Versus... or if its extremely simple, it can use imdbs recommended movie system that's implemented. The other alternative is a site just for the program, that parses in all these details that aren't really stored anywhere, but also grabs recommendations and what not. Its all ideas I have, requires a lot of resources and interaction/management from users, and I can tell you one thing... even if its in a users best interest they will never, ever play ball Distant future we're talking of now, but I got a lot of things that I'd really want from a program, not just a movie manager, but a real resource your going to actually use would be ideal.

... and all this leads me to the conclusion, there's a long, long, long way to version 1.0
*Cue Peter Kaye singing 'The Road to version 1.0"*


** emule functionality, a content manager side is really what the program will eventually offer. Just something to manager and make movies easier to get at

oh and feature requests more than welcome, and partifularly exactly the bits of information you may want to store on a movie, an edition of the movie, a copy of the movie and in collections and what not. Fields or processes you would like to use on your data...

Yeah, I would like that. Something like: IMDB Rating. and Personal Rating.

Would be nice to see a list sorted by what films you enjoyed the most, etc. That feature could of course be extended into a 'recomend' feature again. "I like these films... what other's should I see?'. I remember I once bought the Corel All Movie Guide, which had a system like that, and I was very often amazed at how accurate it turned out to be! I had entered my personal ratings for more than 1000 films, and asked it to check what others I should see, and it would pick out a film (or more) that I actually had seen (but forgotten to rate) and enjoyed very much! I think the recomendation used mostly Plot Keywords, and possibly also director preference. Ie, if you'd entered a high rating for alot of films that had the keyword 'Zombie', it would find another movie with that keyword and recomend it.



eh, I don't know if every emule mod has it, but Neomule has that exact feature implented, at least. It's saved me MANY times!! It only recognizes file-hash's though...


Sadly I can't help with the file-recognition, but I do hope that once you get it done, it will also include MPEG, MKV, OGM and DVD. Very anoying that EMM don't do that.
Maybe it would be possible to use a pseudonym and ask Alessio (the EMM guy) how he does the file-reading? Say you're working on a g-spot kind of program or something....
Or perhaps ask the g-spot guys?

Yup, that's probably a good idea, just trying to avoid a lot of the uselss clutter you see. Now maybe imdb rating under the movie, and personal on the editions?

The reason I say this, is because in my opinion, I like Brazil very much, but i hate the "love conquers all edition" it drops loads. So the imdb version is basically a stat on the film, whereas you rate an edition?

I like recommendations, sometimes you wonder whether you'll like a movie, or just aren't sure what to watch.

Like the TiVo systems, you never know what your missing, and you get the illusion that all tv is shit when sometimes there is the odd programme that's really kind of neat.

I am working on filtering, I want it to be smart, but not restrictive on the user... you know just be helpful.



yeah, indeed that's what got me going... it stores your known files, whereas the other known file only keeps them 'known' whilst your actually sharing, which is really, very, very silly.

Avi file reading is a bit of a blag, you need to decode the avi headers and I just can't say i'm knowledgable about file header information. I'm just going to ask MKV if they can help me with the allocation when i get my head round avi file spec capture, then I'm going to go round lots of Java sites once I've got it figured and release the filereading code.

I think its quite literally going to end up as open the file, read in xyz bytes, drop those bytes, read the next byte and compare it to a list of codec byte codes, read in z bytes, drop those, read the next byte as an video bitrate value, etc.

but how would you let your db-app notice what emule is downloading? Scan the .part.met files?



I once suggested on the official emule forum, that emule should remember what site the link was clicked from. That way you could later go back easily, if you found you had missed a cd, wanted to make a comment about quality, tell people that the file was a fake, nuke, etc. I thought it would be really handy, since it always takes at least a day or two to complete a file, and you don't always remember where you clicked it from. I was flamed for that proposal. =(

look in the root emule folder, then check out the file downloads.txt , I thought all emule versions made this:



if it isn't I'll probably only implement it in mods that do this. I think morphxt and neomule do.

[ Add all 5 links to your ed2k client ]

oh.. forgot about that one.

So what's the idea? It'll scan for similar titles every time you start the db-app?

You got a promo name for it yet btw? just something to call it in the pre-alpha stage I mean..?

No, i thought it would be better to give the manual option, click create stoplist.

And then emule watches to see if its already 'known' or a possible owned movie everytime a new download is added.

My current train of thought is nothing should happen at load time of the app, it should just display the options. This is to give the quickest possible route to adding new movies and what not, then it loads in and caches the first time it needs info. Maybe i'll just tell it to try and get everything ready at bootup of the app later on though.

ATM i'm calling it JMMM Java Movie Media Manager

hmm.. how will you get that to work? wont you need to actually make a emule mod for that? Just trying to understand...



How about inserting an 'I' for internet or whatever, and call it 'Stuttering Jim' =)

oh yes, I know a bit of c++ and have visual studio .net so i'll have a look at the source code sometime.

LOL! Betting it wont be called Neo-mule or anything like it. hehe

question, if you were to search for "Brain Dead", would you want it to match "braindead"?

The reason I ask is that i'm playing with the filters, and I'm not sure whether its best ot be accurate, or have a widescope.

Currently, when i search for "Brain Dead", it matches that exactly. If i search for Braindead, it matches the other film of that name.

What's your 2 cents on the matter?

a check-box like "wider search" near the search field. if not checked then simple search, if checked then go wide