So for my place of summer employment, one of my jobs is to organize their digital library which consists mainly of PDFs, but also .docs, .txts, some jpgs, etc.

I just went through and renamed them all and sorted them into a folder hierarchy that makes sense (file names like R4252tra.pdf.... in a folder called "Things to remember".... not too descriptive), but I'm unsatisfied.

There are a lot of files that can/should go in multiple folders, seeing as they relate to many different topics. There are a lot of files that are very unique in their subject matter, and would need their own top level folder, which would just clutter up the entire organizational system.


SO BASICALLY. What I need is a program that can index/tag files, and then search for those tags and return the results. My dad was showing me Picasa last night (which is really awesome), and it would work perfectly except that it doesn't work with pdfs. So I'm looking for something that works similarly to Picasa (doesn't need to be nearly as complex though).

Any ideas? Anyone have any experience with this sort of problem, or working with any programs like what I need?

I'm trying a few from Sourceforge (Adigres and PDM), but I don't think they're what I need.

Help me Massassiwon Kenobi. You're my only hope.

(Also, before anyone makes some snide-@55 remark, I HAVE searched google, and I HAVE searched sourceforge, and I am TRYING several of the things I found, I just want to know if anyone has any personal experience with anything like what I'm looking for.)

I don't have any known ones to offer, but you could write a simple tagging system.

.

I was thinking about it, but I don't have much (read: hardly any) experience in file manipulation. I'm also not sure how one would store the tag information and associate it with the file.

I'll do some research, see if I can find some tutorials. That would be a good last resort scenerio. (This isn't one of my primary tasks, they just wanted the library cleaned up/renamed/etc, what I've done in the past two days. This indexing/searching is what they NEED, but they don't realize that yet.

Hm. Question...

Does anyone have knowledge in how a tagging system works? I'm thinking about it, and do you even modify the files? Or could you just have a database that contains the file name and the associated tags.

Because if that's all it is....


*schemes*

You shouldn't need to modify the files at all. A database can associate tags with filenames.

Windows Vista/XP does this using the NTFS index.

I don't understand how that helps. I did a google search on NTFS Index and... nothing seemingly relevent came up.

Just map tags to filenames and store the info somewhere yourself. With Java or C# that's as simple as serializing the class containing the map.

Rather than file names I'd generate a hash for each file and use that. That way if a file is renamed but the contents are the same you can still track it.

But once you edit a file, it would break. You can't rescan all the files because that would be clunky and too slow.

The most efficient way to do it is run a service that monitors the target folder structure, which is kind of stupid. If you have Windows' indexing service enabled, you should be able to reference its index of a particular folder instead of doing it yourself.

You could upload all the articles to your webspace and use del.icio.us to tag them, or make an http server to do it. (You may even be able to do it with nfs). It's kind of a reach around for the problem, as I don't know much about your silly fatface windows file system.

Oops, I thought this was for ebooks and that sort of (unchanging) thing.

More than likely, the contents of the files would be unchanged, but it is possible someone might update or edit some of the .docs or whatever, and I think it would be more robust to play with the filenames, as those should stay constant.

Fish- an interesting idea, but we'd prefer to keep it local. Eventually we do want these files to be downloadable from other locations, however.


I think I'm going to write a java app to do this. I could make it an applet to make it web accessable.

Thanks for the help. Any further advice is still welcomed, but I've started to plan out what I'm going to do.

They better be paying you well. Technology people who do more work than they should (Than I would) are very under appreciated.

"Yeah, I just wrote a scalable java applet to store and organize our entire system of folders, files, and documents."
"Good work, now can you possibly keep my coffee warm? I know you're working on all of that computer stuff, but you still have to make sure my coffee is warm."

Yeah, if it turns into more than a few days work, they should be paying you more. Even that might be a bit much.