Re: councilroom.com idea

[Rob Renaud]

I don't admin the councilroom server anymore.  If CR is updating normally (I am not sure if this is currently the case), the log will show up tomorrow after it parses the nightly dump, and you can see the graphs then.

It shouldn't be that hard to change the frontend so that it can pull a single game on demand, parse, and render it.  Basically you'd change the GamePage to take an additional url param that when given, does a request to iso instead of looking it up in the database, and then pass those fresh contents to the annotate game function.  Everything else should work fine.

https://github.com/mikemccllstr/dominionstats/blob/master/frontend.py#L423

You don't have to worry about inserting it into the database, you can just let that happen in its own time when the new file nightly file comes.

But I don't know that there is much demand for such a feature.  Most people are not so excited about a particular game result that they need to see the graph/deck tracker immediately.  Of course, if you want to scratch the itch, I encourage you to try to implement it.  It's actually nice in that most CR development work really requires you get the whole game database running and loaded, but since you don't actually require the backing store, you don't need to get the db running.

On Sun, Feb 24, 2013 at 3:25 PM, The Real Plato <the.rea...@gmail.com> wrote:

Hey man, can I use the councilroom.com deck stats page to view a game by URL?

I want to view stats on http://dominion.isotropic.org/gamelog/201302/24/game-20130224-103117-6c8c2cb8.html 

This is a game I just played today with goons/grand market. I got an early lead by using a cellar to mill my deck so I could buy the first goons.

I was hoping to look at the councilroom graphs showing deck stats over the progress of the game. But it looks like I can only search for games by player names, and it doesn't find this new one. I presume your search indexes cached games from the nightly dumps.

Do you parse all the game logs in a batch every night? Would it be feasible to parse logs on demand (i.e. you haven't seen the game of my pasted URL before), and just skip re-parsing them when you see them later in a dump?

If this isn't coded yet, I might be able to contribute a bit, here is a regex I would use to parse isotropic URLs. 

https://gist.github.com/therealplato/5025421

(i mostly code javascript not python; paste the script into chrome's F12 console or `node gistfile1.js`)

cheers!

--
-plato

(\/) (°m°) (\/)  
Woop woop woop woop woop!