rurabe/drafthacker

drafthacker's Issues

SimpleCov and pushing

It looks like every time we use simplecov and then pull, we will get merge conflicts. I don't know if there is an easy way to get by it but I have been using git rm on the coverage file before every commit, because every time you run the tests it gets regenerated.

Links slots to players

Set up player cron job

24 hours?

Depends on if we want news/etc to be up to date

Links players to picks

Add fields to player model

Get the JSON from the API

Come up with a basic formula

Move the pick-team link functionality from the link class to the team class

For efficiency's sake- In pick it will make (14*number of teams) DB calls to find its team individually. If it were in the team class, it would only make (number of teams) calls to the DB and return the entire array of picks.

I would also love to make it an after_commit callback so that it's not dependent on the order of the megahash, but when it was an after_commit in the pick class it basically crashed rails server.

Links picks to slots

Might need to edit the schema and associations to do this

Pull the data from the API

Get rid of the embedded hashes in the model attributes

Starting with the creation an owner object

Replace search logic with regex

Change the logo attribute of Teams to 'text' in the db

Long urls cause the system to crash

Error Handling for before the draft starts

Currently, the @players variable is nil before the draft starts so we need some type of error handling to stop it from crashing when it does @players.each

Design a UI to display this info

D3!!!!!!!!!!!!!!!!!!!!11

Drafted players should have a 0% chance of being available at your next turn

self explanatory.

Make draft update more efficient

Update should check and see if there is new data, and only return the new players

Get the JSON and convert it

Set up draft cron job

Maybe every 5 seconds when draft_status == active? It might be more efficient if it can be scheduled for the draft date/time.

We can tune it after we get it working

Upserts

Allow the data to update or insert if not found

Chart pick probability distributions based on ADP

Based on a players average pick, high and low, we can make certain statistical assumptions about when that player will get selected. The distribution can be mapped to picks and thus can we can calculate the probability that a player will be drafted at any given pick in the draft.

upsert

finish fixing upsert module

fix slots
fix picks
fix rounds!

Build it

Connect it to the backend

Make league update more efficient

League should check the api to see if the data is new, and should update only if the data is new. This is dependent on the upsert process getting worked out.

Ajax it

Update the existing player objects with the ADP data

Get cron jobs working with audit logs

Chances should update after every pick

Search fields

Maybe combined with the select boxes?

Update the players from the Cbs module

Assign pick probabilities to the players

Select boxes

Use this -- http://harvesthq.github.com/chosen/

Batch inserts

Nested attributes inserts atomically, which is bad for us. Get batch inserts working.

next_pick.number

in the last round, next pick is nil, which means that the number method errors out

Moar Ajax