Page 1 of 4 | Next »
[02:14:15] busterarm: i'd probably set up a polymorphic association 'Commentable' and make comments commentable
[02:32:51] busterarm: there are some great blog posts available about polymorphic associations using this exact use case as an example
[15:08:25] busterarm: i just do it to make sure there isn't a better method I'm forgetting. I know the limitations of my own memory :D
[02:32:08] busterarm: i saw a tech talk today about Postgres and somebody showed a merge statement in their code and I was like "uhhh...Postgres doesn't support MERGE"
[02:28:54] busterarm: don't write posts about how you hate a company's service on the internet. They might contact you out of the blue and ask you to interview with them.
[02:29:34] busterarm: I mean, I have a well reasoned argument about why I dislike their service _here_
[02:30:15] busterarm: which is coincidentally the same place they found out about me (my Who's Hiring post last month)
[02:33:21] busterarm: it might be one of those things where I contact HN staff and be like "hey, this post might cost me my job some day, can you delete it?"
[02:35:02] busterarm: i certainly want to interview with them though so I can leverage an offer with other places I'm interviewing with
[17:14:12] busterarm: httparty seems to be the only one with active development. rest-client is close though
[21:07:10] busterarm: 10:06:31 =!= | irc: reading data on socket: error -54 Error in the pull function.
[21:08:12] busterarm: i'm using sasl. i don't have to identify on signin. It even signs me in as busterarm, just the nick is locked when my client disconnects
[05:39:51] busterarm: but then again, i spike new Rails app to test theories so frequently that it's almost gone
[05:44:33] busterarm: "And if you're havin' data problems, I feel bad for you son. I've got 99 Issues but Postgres ain't one"
[17:08:03] busterarm: problems with rails typically happen when people insist on not following the conventions
[17:09:18] busterarm: and typically if you want to change to a different server, folks do it with gems and the rails app config
[17:10:47] busterarm: the best route through a first rails app is usually Hartl's tutorial (debatably)
[17:18:46] busterarm: SteenJobs: return whatever status you want (success/error/etc) back to your controller action and handle it there
[17:20:05] busterarm: do you have a local browser (likely links/lynx) on the VPS you could browse to localhost:3000?
[17:31:18] busterarm: that's the rules of the channel guys, if you want to debate it, #ruby-community
[17:33:40] busterarm: it is probably the second most common topic discussion after actual questions about ruby...
[21:08:18] busterarm: and ideally the cookie wouldn't allow them to charge the card, it would allow them to create the job
[21:08:34] busterarm: token would be on completion of a successful transaction - card has already been charged
[21:18:40] busterarm: yeah i think unavoidably you're making two separate POST requests from the client
[21:19:59] busterarm: one of those would have to be an AJAX request (if you want a good user experience)
[21:32:04] busterarm: you can just return out of the job controller with some error data to the frontend if the payment fails
[21:41:24] busterarm: i've never had a use case that required speed over reliability...so for anywhere I'd use Redis, there's Postgres
[21:48:58] busterarm: what I'm asking is, is your browser getting that data in a stream or in one big chunk?
[21:51:20] busterarm: then you would need a websocket and also some way for the backend to know how much of the job is complete
[21:51:55] busterarm: you don't have a persistent connection between the client and server unless you've set up a socket.
[21:56:43] busterarm: yeah i'm saying you have no way to get that data to the browser unless there's a connection between them
[22:00:01] busterarm: well, the response to the broser request only gets processed and sent at the end
[22:15:54] busterarm: and corproate intranets are the most common source of machines that aren't up to date
[22:17:10] busterarm: I wish Firefox's dev tools were on par with Chrome so I could stop using this pos
[22:22:29] busterarm: Hates_: you don't want them to close the browser before it's sent the server the request
[22:23:32] busterarm: you just start a background job and if they try and come back to the page just tell them it's not done processing yet
[22:24:59] busterarm: if you tell them they need to wait an hour and it _actually_ takes 5 minutes, they'll be ecstatic
[22:31:26] busterarm: in all honesty, emailing them when the job is done is actually a much better user experience than them sitting in front of their browser waiting for a few minutes
[22:33:47] busterarm: do you know that just putting a single character into cell IV65536 puts the page count up to 38,019?
[22:34:55] busterarm: In Excel 2007 - the bottom right cell reference is XFD1048576 - equates to 17,179,869,184 cells per sheet
[22:34:58] busterarm: (about 2.5 times the number of people on the earth) (6,767,908,320 being approximately the number of people on the earth)