2009 -- the year of November. <november-wiki.org> <github.com/viklund/november> <irclog.perlgeek.de/november-wiki> <nopaste.snit.ch>
Set by moderator on 27 May 2009.
01:15 lichtkind_ joined 05:35 ihrd joined 06:58 ihrd joined 06:59 ihrd left 07:01 p6eval joined 07:57 viklund joined 09:21 masak joined 09:27 lichtkind joined 09:40 sri_kraih_ joined
viklund masak: have you checked out mom? 09:51
masak just saw the email about it.
viklund (i'm not particularly happy with it, but it does the job)
masak I should say that I'm having a different intuition about solving this problem (I'd be happy with not having an intermediate DOM-like format and just solving everything with text), but I'm willing to try this path and see where it leads. 09:52
seems this is something that can lead to quite powerful plugins.
viklund masak: yes, it could, and as mentioned, toc will be a simple plugin 09:53
masak yes.
viklund it will increase the time to render a page quite a lot though
luckily we have caching
and maybe we can plug grampa into it somehow as well ;) 09:55
making it more general
masak I had the same thought.
grampa will likely end up being a general XPath search tool for hierarchical structures (maybe in its separate project) plus an adapter for traversing Matchers. 09:56
er, Matches.
viklund yes 09:57
masak we should have november commits reported on the channel. 10:11
viklund yes, hook up dalek? 10:12
(or is it delek)
masak it's dalek. who runs it?
viklund no idea, is it in the pugs repo? 10:13
masak don't think so. 10:14
viklund moritz_++ might know 10:17
12:15 moritz_ joined
moritz_ ehlo november-folks 12:15
I somehow forgot about that channel after my last reboot
masak not acceptable. :) here's where all the secondary action is. :P 12:16
moritz_ action = integral langrange-function dt 12:17
(I know, physicists are just sick) 12:18
sri_kraih_ throws a shoe at moritz_
moritz_ tunes is surface to have a reflection coefficient of 1 12:19
masak sri_kraih_: I'd like to discuss more about scaffolding at some point. 13:06
sri_kraih_ masak: i don't like scaffolding much
masak that's a good start for a discussion.
I know next to nothing about them, but I've heard obra_ saying he doesn't like 'em either. 13:07
sri_kraih_ the only actual useful implementation i have ever seen was the django admin interface
masak and why is that one useful, unlike all the others?
sri_kraih_ the problem is they are never enough and hard to extend
masak ok.
sri_kraih_ because it's mostly independent from your app, and you can use it as a web interface to your database, edit raw tables and stuff 13:08
masak sounds nice.
sri_kraih_ it's like phpmyadmin
masak I'll have to have a good look at Django.
aha. 13:09
sri_kraih_ but when you start your real app with scaffolding you tend to build your app around that
which nealry always ends up in bad code
and wasted time
masak ok.
what are the alternatives?
sri_kraih_ not using scaffolding or having it separate like django 13:10
masak hm. defined 'separate'. 13:11
s/defined/define/
sri_kraih_ a miniapp in your webapp
it exists right next to your real app 13:12
/admin/
masak ah, ok.
yes, that does sounds cleaner and more extensible.
sri_kraih_ it's not bad 13:13
masak how did Merb do it? you mentioned that it didn't have scaffolding.
sri_kraih_ they don't do scaffolding mostly
thing about scaffolding and admin interfaces is that you are bound to a single ORM 13:14
it's a huge tradeoff
sure, you can do cool things like auto generated REST apis to your tables, but you have to let your framework make choices like which ORM to use 13:15
masak good to know.
right now, we don't even have a working database connection in Perl 6, but it's good to think ahead. :) 13:16
sri_kraih_ :)
masak November still uses a file backend.
sri_kraih_ would like to get Mojo::Transaction ported to Perl6
masak sri_kraih_: what are you blocking on? 13:17
sri_kraih_ last time i checked rakudo didn't have working sockets
masak it does have sockets. 13:19
moritz_ it has basic sockets now.
sri_kraih_ good to know
masak but I don't know what you mean by 'working' :)
sri_kraih_ xD
sri_kraih_ remembers the netcat hack for the first perl6 web server 13:20
masak mberends++ and I did the bare minimum to get server-side sockets working during the post-NPW hackathon.
sri_kraih_: that hack is now replaced by Rakudo sockets.
sri_kraih_ select() working? 13:21
actually there is not that much i need to port 13:22
github.com/kraih/mojo/blob/9c7e1509...akefile.PL
zarah sri_kraih_'s link is also tinyurl.com/nby2c5
sri_kraih_ if apis similar to those core modules are available i can get everything from http1.1 client/server, fastcgi, template engine to routes working 13:23
masak sri_kraih_: have a look for yourself at what's implemented: github.com/rakudo/rakudo/tree/d95de.../IO/Socket
zarah masak's link is also tinyurl.com/n4wclq
masak so essentially open, bind, listen and accept. 13:24
sri_kraih_ thats enough if it's working with select 13:26
masak I don't understand what that means.
sri_kraih_ dunno if and how the select syscall is implemented currently 13:27
doesn't look like theres a fancy IO::Select module yet 13:28
masak if it's at all possible, it would probably be PIR code.
you can inline PIR into Perl 6 code. there are plenty of examples in the Rakudo setting. 13:29
masak reads up on XPath 1.0 13:46
one of my favourite standards of all time.
13:49 ruoso joined 17:50 ruoso joined 20:43 masak joined 22:13 ruoso joined