How to end all import issues (an IDEA)

[whateverer]

Not sure if its the right place to post, but right now i'm doing interesting thing, and i thought that it would be both easy to implement and very profitable.

I'm importing all of my hands into pt3 DB in 100k chunks, as it is faster (checked and verified - 150k takes 30 min to import, while 50k takes 6) and less likely to get hangups.

After all of my hands are imported i will just convert that DB to PT4 and after all that caching, housekeeping. etc ill get a nice DB.
Why is it good? Because at the same time i can use my PT4 to play and i don't have to wait hours or days for the import to do its thing.

So my idea is a simple one and i like it very much.

Create a standalone (or not) import program, that will create a new DB and import large HH files into it (hopefully auto divide big files into chunks, as it gets faster, and have no downside as far as im aware). After importing It copies configuration from PT4 and builds cache, vacuums and do all that mysterious things as if it were a normal DB. After it finishes it asks for permission to overwrite/rename/whatever you do with an old DB.... And VOILA!

Ideally it should be able to stop and restart any given task at the same point so user can turn it off and go to sleep/watch a movie/play a game.


It deals with 2 most common complaints about imports.
"I cant play while it is importing" - now you can.
"It takes too long" - now it doesnt matter, just import most recent hands normally and let the program import for a few days, as it does not affect nothing.


What do you think? Am i awesome? (on top of that you get the job halfway done, you can copy code from existing importer and just add things i mentioned)

[WhiteRider]

When PT4 "converts" your PT3 database it is actually exporting all of the hands from your PT3 database and importing them into PT4, so it will actually be much faster to import your 100k chuck files directly into PT4. You can do that import in batches if you want to break it up to allow you to play in between.

[whateverer]

Thanks for replying. I don't know if it would be faster or not, and this is not the reason why i was writing this.
Actually i don't mind if it was 2x or 4x slower if i import into pt3 db first and then convert, because i still can play and have the hud working.

I'm not sure if you get what i was about in the previous post, so ill go step by step.

1. Program copies chosen original HH database with configuration files. (say PT4DB and makes PT4DBCOPY)
2. Program imports hands from any given file/directory into PT4COPY. I still can use PT4DB, i can play, use hud etc.
3. After all those housekeeping things, program waits for me to stop autoimporting and just disconnects PT4DB and connects PT4DBCOPY in its place.
4. Last part is importing hands that might be still on the original PT4DB but not on PT4DBCOPY when program was running and player got anything from autoimport.
5. Done.

Why i think it'sa good idea? Because all of the long work is going on in the background. Even if it runs for week straight i can still play, use hud, populate and load graphs. Pokertracker is still usable.

What i hate about current situation is that i'm blocked from even viewing my stats when import is going on.


I really think that is a good idea, but i don't know if i was clear enough before.

[WhiteRider]

I'll pass on your suggestion to the development team, but since this type of import is a one-off process I'm not sure something like this would be a priority.