Monday, 1 October 2012< ^ >
arjan has set the subject to: Zotonic - the Erlang Content Management Framework
Room Configuration

[08:54:42] arjan joins the room
[09:00:59] andreas.stenius joins the room
[09:09:18] <arjan> hi andreas
[09:09:31] <arjan> I think I also just merged #404
[09:10:22] <arjan> :)
[09:10:55] <arjan> but not by pressing the button
[09:11:06] <arjan> it looks funny, a merge commit of 2 identical commits
[09:11:29] <arjan> well not entirely identical because I changed the commit message slightly
[09:13:12] <andreas.stenius> Oops. :p
[09:14:15] <arjan> hehe
[09:14:18] <arjan> by the way
[09:14:22] <arjan> how do you edit RST tables?
[09:14:33] <andreas.stenius> using emacs table mode
[09:14:46] <arjan> ok
[09:14:55] <arjan> even those with the === ==== syntax?
[09:15:01] <andreas.stenius> first, I used the simpler version of the tables, but it became a mess as soon as there where more rows (with long columns)
[09:15:11] <andreas.stenius> ah, nope
[09:15:15] <arjan> l
[09:15:17] <arjan> ok
[09:15:20] <arjan> I also use table mode
[09:15:32] <andreas.stenius> those are just hand crafted... I switched after a while...
[09:15:32] <arjan> but dont know how to do table headings with those
[09:15:36] <arjan> heheh
[09:16:04] <andreas.stenius> move to the the line separating headers from data rows
[09:16:06] <andreas.stenius> then
[09:16:13] <andreas.stenius> mark line
[09:16:23] <andreas.stenius> then replace all - -> =
[09:16:45] <arjan> ah
[09:16:57] <arjan> so you get a +=====+===+ line
[09:16:59] <andreas.stenius> it'll be smart enough to keep inserting = for you if you widen a cell etc
[09:17:04] <andreas.stenius> yep
[09:17:07] <arjan> cool
[09:17:12] <arjan> I'll try that
[09:17:16] <andreas.stenius> :)
[09:17:24] <arjan> converting this one now: http://zotonic.com/documentation/726/m-search
[09:17:28] <arjan> :P
[09:17:29] <andreas.stenius> I spent an hour learning table mode...
[09:17:41] <arjan> heheh
[09:17:47] <andreas.stenius> ah! defintely woth table mode :)
[09:17:53] <andreas.stenius> worth*
[09:17:58] <arjan> yep
[09:18:45] <andreas.stenius> I found it easiest to insert markers for where I wanted to split the cells first, then create the data into a table.
[09:19:03] <andreas.stenius> otherwise it was hell to get the cells to match up.
[09:19:42] <arjan> yes
[09:19:59] <arjan> well actually if you copypaste from the site there is already the tab character for the columns
[09:20:19] <andreas.stenius> Oh, dang. I missed that. Nice.
[09:20:24] <arjan> only need extra markers needed for multiline texts
[09:23:12] <andreas.stenius> I thoughy I'd convert Kim Shrier's patch for webzmachine into a pull request...
[09:23:30] <andreas.stenius> what is this, too early to type... ? :P
[09:23:51] <andreas.stenius> not healthy taking the entire weekend off...
[09:24:19] <arjan> :)
[09:25:04] <arjan> did not look yet at the patch
[09:26:08] <andreas.stenius> it's easier to comment and review as pull request.. ;)
[09:43:09] <Marc Worrell> Nice markdown editor: http://oscargodson.github.com/EpicEditor/
[09:43:35] maas.maarten.zeeman joins the room
[09:44:03] <andreas.stenius> Marc Worrell: :)
[09:47:08] <Marc Worrell> indeed - check these resources
[09:47:19] <Marc Worrell> http://ticci.org/some-markdown-links
[09:47:35] <Marc Worrell> Dillinger is also nice, some great work is going on here.
[09:47:49] <Marc Worrell> Need to hook it up into Zotonic - will solve quite an editing problem.
[09:54:38] <arjan> I like the popout of the epiceditor
[09:55:07] <Marc Worrell> me too. makes writing a whole lot nicer
[09:55:16] <Marc Worrell> then you can really concentrate on your text
[09:55:45] <andreas.stenius> arjan: popout? (as in fullscreen?)
[09:56:04] <arjan> yes
[09:57:11] <arjan> it does not have a close button in the popout
[09:57:15] <arjan> thats confusing
[09:57:33] <arjan> for mouse-centric people
[09:57:39] <Marc Worrell> yep, I think we need to add a simple menu
[09:57:46] <Marc Worrell> also with a syntax lookup
[09:57:50] <andreas.stenius> nice use of iframe's though. I thought iframe's where more or less dead. But using it to isolate css styling is a rather neat trick, imho.
[09:57:58] <Marc Worrell> it is MIT - we can change it :p
[09:58:05] <andreas.stenius> :)
[09:58:21] <Marc Worrell> need to test it on my phone — one moment
[09:58:36] <andreas.stenius> toolbars are nice, especially when you encounter unknown/new territories...
[09:59:03] <Marc Worrell> some people wrote markdown editors with toolbars - will help starters a lot
[09:59:16] <arjan> also how are you planning to do embeded images?
[09:59:57] <Marc Worrell> maybe something like: [my image](image:1) ?
[10:00:17] <Marc Worrell> the first image/media of the page you are editing
[10:00:29] <Marc Worrell> [my image](media:1) might be better
[10:01:01] <arjan> mm
[10:01:08] <arjan> can you work on this in a branch? :P
[10:01:14] <arjan> it seems like a major thing
[10:01:40] <arjan> and would be nice if we could review your approach
[10:01:45] <andreas.stenius> and for media of other resources? ... [other image](324/media:1) ... ?
[10:02:16] <arjan> and alignment, link, cropping, size...
[10:02:39] <Marc Worrell> better use the media styles we added
[10:03:53] <arjan> I still dont know how those work...
[10:04:08] <Marc Worrell> on iphone that thing is a mess
[10:04:33] <arjan> lets see on nexus7
[10:04:39] <Marc Worrell> @arjan maybe we need a little chapter in the docs on that…. just maybe :p
[10:05:02] <arjan> uhu
[10:05:16] <arjan> im porting over some more docs right now actually
[10:05:19] <arjan> a lot is outdated
[10:05:24] <arjan> and a *lot* is missing
[10:05:24] <Marc Worrell> how strange :p
[10:07:17] <andreas.stenius> lol :p
[10:07:25] <arjan> :(
[10:13:46] <andreas.stenius> yeah, lot's of work :/
[10:14:06] <andreas.stenius> but it's gonna shine! ;)
[10:20:35] <andreas.stenius> for review: https://github.com/zotonic/webzmachine/pull/5 <https://github.com/zotonic/webzmachine/pull/5> :)
[10:37:24] <andreas.stenius> merging webzmachine pull #5...
[10:53:41] <arjan> wow, emacs has a hard time converting this one: http://zotonic.com/model-m-rsc
[10:54:35] <andreas.stenius> haha... that's a BIG one :)
[10:54:53] <andreas.stenius> table mode isn't always 100% stable (I've read)...
[10:55:36] <andreas.stenius> got to spend some time on my "other" work... :/
[10:58:14] <arjan> me too soon
[11:00:51] <andreas.stenius> I mailed to travis support regarding the virtualbox error messages for the builds running R14B03.
[11:09:55] <arjan> fixed model docs
[11:10:06] <andreas.stenius> :D
[11:10:46] <arjan> Lot of work
[11:10:53] <arjan> but emacs rst-mode is helpful
[11:11:22] <andreas.stenius> good ol' emacs ;)
[11:12:17] <andreas.stenius> I took the time to pick it up and learn it when I was consulting for Ericsson... and that has paid off ever since :)
[11:13:14] <arjan> :)
[11:13:34] <arjan> I use it for everything except android development
[11:52:06] <andreas.stenius> From Travis-CI support:
Hi Andreas,
I am sorry about the VM issues, I am working on a fix as I type which hopefully should be deployed later this week.
I have restarted the misbehaving worker and everything should be back to normal now.
Have an awesome day,
[11:54:01] <arjan> cool :)
[14:21:21] <maas.maarten.zeeman> Cool. Does this mean they now have smp support too?
[14:40:33] <andreas.stenius> maas.maarten.zeeman: uhm... not sure. This was a reply I got regarding the recent errors on travis while building zotonic... don't think it's smp related..
[14:40:50] <andreas.stenius> gotta run...
[14:41:40] <maas.maarten.zeeman> Too bad, the sqlite nif neads a erlang vm with smp support in order to function.
[14:42:06] <arjan> hmmmm
[14:42:33] <arjan> most cpus are smp these days right
[14:42:40] <arjan> but probably not on vm's :)
[14:42:54] <maas.maarten.zeeman> right it is probably running on a vm smp.
[14:44:18] <maas.maarten.zeeman> oops without smp. Then the erlang vm misses some import nif api calls to send messages.
[14:44:37] <maas.maarten.zeeman> :-/
[14:45:23] <maas.maarten.zeeman> And calls to lock c structures.
[14:46:23] <arjan> then the nif should just refuse to load on non-smp archs, I guess
[14:47:24] <maas.maarten.zeeman> Yes. Bumped into that one when running the tests on travis. Didn't know those calls where smp only.
[16:02:09] arjan leaves the room
[16:16:48] arjan joins the room
[16:18:16] arjan leaves the room
[16:19:58] arjan joins the room
[16:39:22] <Marc Worrell> sometime for me to learn from :p http://blog.8thlight.com/kevin-liddle/2012/09/27/mind-your-git-manners.html
[17:42:32] Ilya Rezvov leaves the room
[17:42:58] maas.maarten.zeeman leaves the room
[17:47:22] Ilya Rezvov joins the room
[18:16:14] arjan leaves the room
[18:34:22] arjan joins the room
[18:36:14] <arjan> Marc Worrell: nice read
[19:21:55] arjan leaves the room
[20:11:30] arjan joins the room
[20:11:43] <andreas.stenius> yeah, good basic stuff.. :)
[20:11:58] <andreas.stenius> arjan: nice you got the link in the docs, too :)
[20:12:06] arjan leaves the room
[20:13:02] arjan joins the room
[20:13:57] <arjan> andreas.stenius: yup why not set an example
[20:13:59] <arjan> :)
[20:14:53] <andreas.stenius> would be nice with some guidelines to what to put into the summary of the commit description too...
[20:15:18] <arjan> you mean like this? http://zotonic.com/docs/dev/codestyle.html#writing-commit-messages
[20:15:19] <andreas.stenius> maybe that's in the rebar doc already? haven't looked, actually... Lp
[20:15:26] <arjan> "The summary should be less than 50 characters, and tell what was changed. Use the imperative present tense (fix, add, change). For example: Add ‘foobar’ filter, Fix bug in media upload service."
[20:15:31] <arjan> oh no wait
[20:15:36] <arjan> I mean,
[20:15:38] <arjan> "The description should explain the intention and implementation of your approach, in the present tense."
[20:15:41] <arjan> second bullet
[20:15:46] <andreas.stenius> bah. right :P
[20:16:13] <Marc Worrell> :p not "Work in progress, might work, will break, am now on holiday, see you in a year!" ?
[20:16:30] <arjan> eh yeah
[20:16:42] <andreas.stenius> I tend to like to prefix the summary with a tag indicating which are is being affected...
[20:16:44] <arjan> regarding those, let's professionalise a bit :P
[20:16:56] <arjan> andreas.stenius: yes
[20:17:22] <arjan> I mostly already do that
[20:17:30] <arjan> maybe we can prefix it always
[20:17:37] <arjan> either with doc: or core: or mod_xxxx:
[20:17:41] <arjan> depending on what you dol
[20:17:44] <arjan> *do
[20:17:52] <Marc Worrell> sound good
[20:17:58] <Marc Worrell> reads a lot faster
[20:18:03] <andreas.stenius> would be nice. Perhaps update the docs with that, and could be worth mentioning to limit commits to single logical change/update.
[20:18:21] <andreas.stenius> per commit
[20:18:52] <arjan> andreas.stenius: you mean this? http://zotonic.com/docs/dev/codestyle.html#atomicity
[20:18:56] <arjan> lol
[20:19:16] <arjan> but yes
[20:19:21] <arjan> prefixing is good
[20:19:26] <arjan> lets make it official
[20:19:39] <andreas.stenius> I'm sure I read that when you first committed it! :p
[20:19:53] <andreas.stenius> +1 :)
[20:19:59] <arjan> k
[20:39:43] <arjan> yay,alias gpr='git pull --rebase'
[20:39:44] <arjan> :)
[20:41:03] <arjan> pushed codestyle doc.... waiting for rebuild
[20:41:08] <arjan> done
[20:41:09] <arjan> http://zotonic.com/docs/dev/codestyle.html
[20:43:48] <arjan> this is a nice point, ### explain how we do this and how we indent HTML vs. template tags? ### :P
[20:44:11] <arjan> anyone wanna do a proposal for that one? :P
[20:52:52] arjan leaves the room
[21:13:13] <Marc Worrell> about the indenting - tabs are smaller than spaces - which is interesting when indenting HTML
[21:13:34] <Marc Worrell> everything should be 4 spaces per tab though (so set your tab to length 4)
[22:17:21] andreas.stenius leaves the room
[22:30:14] <Marc Worrell> just pinged Loïc about cowboy - thinking of removing the lists and going binary for the requests.
[22:30:38] <Marc Worrell> (and replacing our websocket stuff with cowboy - less code to maintain)
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!