For the graphics gurus (Fozzy, Quicksilver, etc) as well as anyone else out there, they are brainstorming on what the tracker should "look" like. This is vitally important. Hopefully with a great design it will be super user-friendly.

https://****************/forum/forum....orum_id=819271

Any/all advice should be suggested folks....

i don't know how viable it is at this point...but if it's not too late i think if you write some sort of generic library API that can be accessed by other languages you would get a tremendous boost in support.

for example, if u have a fpdb.dll that has the API to add a new hand parsed from a hand history file, someone could program the hand import routine in c++/c#/java (more well known languages) and just call the API in fpdb.dll.

this will remove the python learning curve.

just a thought.

Actually seems not so nasty, there are some good docs on a wiki from the official python site I've been trying to read last few days.

This sounds like a good idea to me, but may cause the project to become a bit strange for a user to get working.

It's not hard as it's designed to be very high level, but IIRC it has some really wacky scope-indentation stuff that drove me nuts the last time I tried to used it (assuming I'm thinking of the correct language).

Juk

Yeah I thought about doing something like that, but to be honest I reckon the best way for accessing fpdb from another language is by simply going straight to the database. If for example someone wants to create another importer just do it in whatever language you like and make it store into the db, and make it have a command line interface. I could then integrate that into fpdb's main GUI by executing command line commands.

For python I found these 3 pages to be most helpful:
- the python tutorial
- the python library spec
- the pygtk docs

Python is super easy to learn if you have any programming experience and the indentation becomes a non issue if you use a decent editor.

Wholly crap Batman, my brain is about to asplode. I downloaded python and mysql with no problem but trying to setup all the database user stuff with webmin or whatever got me lost.

Forget webmin, install MySQL and make sure you created the root user and set a password when it prompted you during install.
Then do this (these instructions are for Windows, since you didn't specify I assume that's what you use):

Open a shell (aka command prompt aka DOS window):
Click Start, then Run. In the opening window type "cmd" (without the inverted commas) and then click OK. A windows with a black background should open.

Type (replacing yourPassword with the root password for MySQL you specified during installation):
mysql --user=root --password=yourPassword

It should say something like this:
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 4
Server version: 5.0.60-log Gentoo Linux mysql-5.0.60-r1

Type 'help;' or '\h' for help. Type '\c' to clear the buffer.

mysql>

Now type this to create the database:
CREATE DATABASE fpdb;

Then type this (replacing newPassword with the password you want the fpdb user to have - this can, but for security shouldn't, be the same as the root mysql password):
GRANT ALL PRIVILEGES ON fpdb.* TO 'fpdb'@'localhost' IDENTIFIED BY 'newPassword' WITH GRANT OPTION;

This thread should be set sticky

Around these parts, it is likely a good idea to forget MySQL, and default to postgres.

it is pretty much a given that anyone on 2p2 software who has tried any tracking software recently, even if a total computer noob, has struggled thru getting postgresql set up on their windows already, since it is requirement for PT3 and HoldemManager, and recommended for PT2.

I already have Postgres up and usable for PT3 & HM. What steps do I need to do to get user setup for FPDB?

that is a good point _dave_...If it is not too late you should definitely be going with postgres.

I agree, the fact that fpdb uses mysql is more a historical anomally really. It happened because someone who helped at an earlier stage couldn't install postgres due to access restrictions on a corporate machine, but could get mysql working. Long story short, in principle fpdb already supports postgres, I just need to do a few minor things (mostly documentation) and figure out how to make conditional imports work in python. I'll try to do that shortly.

Any todo list? My python is basic but it's a nice looking language so I would like to take the chance to learn it better.

[edit] found the dev readme, ill do some of those boring things listed :P

Well, not so fast dave. The reason PG is the db of choice is because when ptrack Pat was considering a heavier db than Access., he decided (correctly I believe) that the MySQL open source license prevented him from distributing PokerTracker without opening up the source code of PT (or there was some other similar restriction). I remember it was a fairly significant debate at the time and many people argued that MySQL was a better choice.

Although I believe PG has come a long way since then, I have read (but have not confirmed as I have about zero MySQL experience) that MySQL performance may be better. I think that since this is an open source project, having compatibility with both would be a very good thing. Although I certainly agree the focus should be on PG.

ugh unix nl and cr's. Guessing this is pretty important as whitespace is important to python. Anyone know a nice plaintext editor for win that handles newlines same as unix?

idle is pretty frustrating but it might be fine when im used too it

PSpad reads it good.

Yeah I remember the pg vs my debates on PT2 forum years ago. Thankfully the licensing issues of MySQL made the decision much easier. I'd wager there are corners of the internet still around that house frequent flamewars re: MySQL Vs postgres

My comment was not really with the intention of which is "better" (fwiw, I prefer postgres mainly due to it's superior integrity protection and awesome datetime handling) - but more that I think it is far easier to persuade the majority of people to try "yet another tracker beta" if they can easily use their existing posgtresql service on Windows, rather than have the additional obstacle of installing another database server - which even the mere mention of may be enough to turn away users if they have had an ordeal getting postgres installed properly.

Try just changing it to Windows line endings, it might work. If not let me know and I'll change it to be flexible.

A word on the latest commits, in git 38-40 I made some fairly significant changes to the tables. For one there is now a third dedicated Tourney table that is also referenced from the cache table and I updated all the tables to my naming convention so that I don't have to do it later. The tables are now very near finalisation, but there's a couple of things left:
- There's gonna be something to store when a player went all in
- More columns in the HudData cache table, and a stud equivalent of it.
- Since tourney support is only partially implemented some minor changes may happen there
That's it though, I don't think any further changes to the existing fields will be necessary. This is not a promise though.

upupup

Please explain, in a totall noob-friendly way, how one can get these new versions - I only find "git-27" zip on sourceforge.

Despite running gentoo, I have no previous knowledge of git. all I know is it is "better than cvs / svn". what do I need to emerge? then what? etc. After a couple days of emerging various python / postgres stuff, I think I should be able to run it now (lol gentoo)

If I fix / add something, how do I use git to get an updated version to you for review?

Win32 instructions are probably useful too if you can


BTW, I hope you have email notification activated for this thread, it is a handy feature of these forums

Yep, I do have email notification on its brilliant

I will make roughly weekly alpha releases on sourceforge, making a release for every git would take too much time and I want to make at least some regression testing and documentation updates and improvements for each alpha.

On Git:
First of all you can use webgit at http://trac-git.assembla.com/fpdb/browser. However you would have to open each changed file, scroll to the bottom, select original format, then save.

To get git for gentoo just do emerge git -av
To get it for Windows go to http://code.google.com/p/msysgit/downloads/list

There's a bunch of instructions at http://www.assembla.com/spaces/fpdb/trac_git_tool
For example, to pull the git tree just create a new directory (lets say ~/fpdb/ ) and type:
git clone git://git.assembla.com/fpdb.git

then you make your changes and run git-add--interactive from ~/fpdb/

now press u and and enter. It'll list any changed files. Assuming you want to commit all changes just press * and enter twice.

If you added any files press a and enter, then the number of the added file. then press enter again to return to the "main screen".

now press q and enter to quit git-add.

i keep my commit message in since_last_commit.txt (don't add this file with git-add though) so to commit you'd run:
git-commit -F since_last_commit.txt

I haven't checked yet how to use git to "create a patch" so in the meantime theres two options:
- just zip (or rar or tbz2 or whatever) up the entire directory and email it to me
- get your own public git tree to commit to, for example at http://repo.or.cz/ Then I could just pull from your tree.

Any questions just post again, I'll add this to git-instructions at some point

[IMG]My Head Just Exploded.GIF[/IMG]

I'll try and explain it a bit clearer in git-instructions, just wrapping up alpha2 now anyways

well, the "clone" command worked and now I've got a git / fpdb tree in my home folder.

I'm kinda unclear on "make your changes" - I just edit the fies direct from my local tree, and git will figure out what has changed / allow me to revert / create a new "revert point"?

I'm gonna read some git tutorials tomorrow, and play about - I think I've read enough python docs to be dangerous now

Fozzy - this git stuff is cool - you guys could possibly get great benefit running it for version control / backup of all your mods, if it deals with PSD / 3ds as easily as .py files

Sounds interesting. I will have to dig into it sometime soon. My method currently is certainly sub-optimal, but luckily my mods and installers don't begin to compare to the complexity of something like this project.