Monday, 8 October 2012< ^ >
arjan has set the subject to: Zotonic - the Erlang Content Management Framework
Room Configuration

[08:27:36] arjan leaves the room
[08:48:58] arjan joins the room
[09:21:54] andreas.stenius joins the room
[09:35:43] maas.maarten.zeeman joins the room
[10:17:11] arjan leaves the room
[10:27:19] arjan joins the room
[10:43:26] Ilya Rezvov joins the room
[11:54:04] <andreas.stenius> phew. scomps migrated. :) I see there's still a rather long list of todo's in the ref section... :/
[11:54:25] <arjan> andreas.stenius: nice work!
[11:55:20] <andreas.stenius> :)
[11:55:36] <andreas.stenius> gotta run. have a meeting at the kids school...
[12:01:54] <arjan> :D
[12:38:21] <maas.maarten.zeeman> Nice
[14:38:28] <andreas.stenius> arjan: check out this little defun: http://zotonic.com/docs/dev/documentation.html#heading-styles ;)
[15:02:52] <andreas.stenius> are the early zotonic days from svn available somewhere?
[15:03:32] <Marc Worrell> yep, on google code :p
[15:06:08] <andreas.stenius> oh, can't find it there...
[15:07:04] <andreas.stenius> looks like the source part has been removed/hidden.. ? or is it just in the web gui?... but then one would need to know the svn url for it...
[15:16:49] <andreas.stenius> well, yay! I guessed it right.... got, thanks :p
[15:17:04] <andreas.stenius> got it*
[15:18:00] <andreas.stenius> uhm, spoke too soon.... it was empty... :/
[15:20:52] <andreas.stenius> so, googlecode answers on zotonic.googlecode.com/svn/ but, it's nothing more than an empty repo there...
[15:25:37] Ilya Rezvov leaves the room
[16:44:01] Ilya Rezvov joins the room
[17:24:41] Ilya Rezvov leaves the room
[17:37:33] maas.maarten.zeeman leaves the room
[17:49:06] Marc Worrell leaves the room: Disconnected: connection closed
[17:59:36] Marc Worrell joins the room
[18:06:06] Marc Worrell leaves the room: Disconnected: session closed
[18:06:11] Marc Worrell joins the room
[18:14:39] <arjan> andreas.stenius: https://code.google.com/p/zotonic/source/list
[18:14:51] <arjan> the source code tab is hidden
[18:16:15] arjan leaves the room
[18:36:03] arjan joins the room
[18:37:45] <andreas.stenius> arjan: yeah, thanks. I found that. But that is the hg stuff only. And the first commit is adding the stuff fron svn r1392...
[18:38:27] <arjan> maybe not all history is online
[18:39:16] <andreas.stenius> yeah, I was looking for those first 1392 revisions of svn history, which I couldn't find... :p
[18:39:51] <andreas.stenius> someone took the easy path, not committing the entire history ;d
[18:40:05] <andreas.stenius> *nudge nudge*
[18:40:36] <andreas.stenius> do you have it locally somewhere?
[18:44:50] arjan leaves the room
[18:55:28] <Marc Worrell> hmmm, I don't think so, it was stored in an external vsn hosting service
[18:55:39] <Marc Worrell> and migrated from there
[18:56:23] <andreas.stenius> so no svn checkout on any of your machines floating around, then...? :}
[18:56:53] <andreas.stenius> oh, right... svn doesn't include the history in the checkout, right
[18:57:16] <Marc Worrell> the machine I used back then is decommissioned (aka stolen)
[18:57:26] <andreas.stenius> oh :/
[18:57:29] <Marc Worrell> let me check a backup disk
[18:59:12] <andreas.stenius> :)
[19:01:32] <Marc Worrell> i have a tgz of r587
[19:02:42] <andreas.stenius> what I'm looking for this time, is the root of erlydtl, and what changes has been made after the fork from emiller's version....
[19:03:06] <andreas.stenius> so perhaps 587 is closer to that than 1392... ? ;)
[19:03:07] <Marc Worrell> hmmm, that is a loooong history :p
[19:03:58] <andreas.stenius> I looked at erlydtl today, and see that they've been adding and improving stuff over the years. Wouldn't it be nice to be able to stay somewhat in sync... ?
[19:04:15] <Marc Worrell> I have also an old design for the website :p
[19:04:19] <andreas.stenius> I realize it's a lot of work to get upto speed with these years in between, though...
[19:04:27] <Marc Worrell> stuff was called zophrenic back then
[19:04:47] <andreas.stenius> I've noticed :) and "pragmatic innovation" :d
[19:06:23] <andreas.stenius> I think that maybe the mods needed to erlydtl can be made less intrusive with todays version of erlydtl and what was available back then...
[19:06:45] <andreas.stenius> and -> than*
[19:07:05] <andreas.stenius> time to put kids to sleep.... be back in 45, or so....
[19:07:58] <Marc Worrell> it is in your e-mail
[19:09:58] <Marc Worrell> I like the slogan "pragmatic innovation" :p
[19:12:33] <Marc Worrell> tadaaaa!
[19:12:35] <Marc Worrell> SVN-fs-dump-format-version: 2
UUID: 39ac7d6a-fe55-4cd0-aea4-b13bf11f1711
Revision-number: 0
Prop-content-length: 56
Content-length: 56
V 27
[19:42:40] arjan joins the room
[19:43:00] <arjan> nice nice
[20:05:26] <andreas.stenius> yay! :D
[20:12:00] <Marc Worrell> it is not the oldest though
[20:12:08] <Marc Worrell> but can't find older on my systems here
[20:12:27] <Marc Worrell> oldest is from 1.5 months before that one
[20:13:06] <andreas.stenius> it's so neat and smallish (yet, a lot of todays functionality is there, rather intact)... ;)
[20:13:20] <Marc Worrell> :p
[20:13:25] <Marc Worrell> no vhosts etc :p
[20:14:10] <andreas.stenius> do you know what date or version of erlydtl you forked from?
[20:14:35] <Marc Worrell> google code version, must be from that janurai
[20:14:38] <Marc Worrell> januari
[20:14:47] <andreas.stenius> 2009-01?
[20:14:53] <Marc Worrell> guess so
[20:15:09] <Marc Worrell> it was fairly static back then
[20:15:15] <andreas.stenius> nice, will look into it. See how far away current erlydtl have come...
[20:15:25] <Marc Worrell> quite far.....
[20:15:47] <Marc Worrell> they tried to catch up with us :p
[20:15:53] <andreas.stenius> lol
[20:16:03] <Marc Worrell> copied our expression syntax
[20:16:15] <andreas.stenius> thing is, they have more ppl contributing to the template stuff...
[20:16:21] <andreas.stenius> nice
[20:16:54] <Marc Worrell> we can check what they do
[20:16:57] <andreas.stenius> so I thought it could be nice to be able to pull in from them occassionally...
[20:17:05] <Marc Worrell> our stuff regarding translations is much better though
[20:17:30] <Marc Worrell> and we added an extra argument for our #context
[20:17:42] <Marc Worrell> they didn't have anything like that, a bit strange
[20:17:44] <andreas.stenius> and for model, too right
[20:18:06] <andreas.stenius> I think they have a way to pass a user context to their callbacks now...
[20:18:12] <Marc Worrell> model, scomp, all include, include args, include tpl expression
[20:18:16] <Marc Worrell> caching
[20:18:23] <Marc Worrell> for .. empty
[20:18:25] <Marc Worrell> elseif
[20:18:33] <Marc Worrell> {_ … _}
[20:18:35] <andreas.stenius> they have elif now...
[20:18:38] <Marc Worrell> arg=_"aaaa"
[20:18:51] <Marc Worrell> elif? :p
[20:19:10] <Marc Worrell> we can make an alias :p
[20:19:11] <andreas.stenius> elseif :p, think it's elif in DTL...
[20:19:22] <andreas.stenius> yeah
[20:19:25] <Marc Worrell> I though I didn't see it in DTL
[20:19:47] <Marc Worrell> then we make it elif and alias elseif
[20:20:08] <andreas.stenius> https://docs.djangoproject.com/en/1.4/ref/templates/builtins/#if
[20:20:41] <Marc Worrell> aha, new in March 23, 2012
[20:22:00] <Marc Worrell> they have now also:
[20:22:01] <Marc Worrell> {% include "name_snippet.html" with person="Jane" greeting="Hello" %}
[20:22:26] <Marc Worrell> we have the same without he "with" maybe add it as an optional element in the syntax
[20:25:23] <arjan> yes, would be nice :)
[20:25:29] <arjan> keep it in sync a bit
[20:26:19] Marc Worrell adding elif (+elseif)
[20:30:43] <arjan> dont forget the documentation :P
[20:41:43] <andreas.stenius> aha, most likey based on erlydtl 0.5.1, judging by the readme, since 0.5.2 mentions a out_dir option, which isn't present in the readme in the version in zotonic ;)
[20:47:32] <Marc Worrell> yeah :)
[20:47:54] <Marc Worrell> ok, I have elif and an optional 'with' keyword :p
[20:48:19] <andreas.stenius> nice :d
[20:48:39] <andreas.stenius> how come you didn't keep the erlydtl template tests?
[20:48:51] <Marc Worrell> we have them
[20:48:56] <andreas.stenius> oh...
[20:49:06] <Marc Worrell> src/tests/
[20:49:13] <andreas.stenius> ah, right
[20:49:30] Ilya Rezvov joins the room
[20:50:01] <andreas.stenius> I have two concerns. To not break existing functionality, and to if not improve speed, at least not make it worse :p
[20:50:30] <andreas.stenius> about the second concern, do we have any benchmark style tests?
[20:55:04] <andreas.stenius> Marc Worrell: the svn dump, it start with "Initial import" at rev 1, so if this isn't the oldest, I guess the older stuff wasn't in this svn repo... :p
[21:01:46] <Marc Worrell> yep, in some local svn thing that is not alive anymore
[21:01:50] <Marc Worrell> tried to find it
[21:02:29] <Marc Worrell> we don't have benchmark style tests
[21:02:56] <andreas.stenius> this goes far enough back in history.. :)
[21:02:58] <andreas.stenius> thanks
[21:03:41] <andreas.stenius> I would like to have isolated unit tests. and perhaps it would be nice to have benchmark tests too, so we can keep an active eye on performance...
[21:04:52] <Marc Worrell> yep, I agree
[21:05:01] <Marc Worrell> it is too easy to slow down things
[21:05:36] <Marc Worrell> maybe we should splt our erlydtl in erlyzdtl - I think that Arjan was doing just that
[21:05:45] <arjan> yep
[21:05:56] <arjan> although that repo is not published yet
[21:06:02] <andreas.stenius> aha, how far you'd get with that?
[21:06:06] <arjan> and I commented out all zootnic-specific stuff :P
[21:06:20] <Marc Worrell> uh oh :p
[21:06:28] <Marc Worrell> that is a lot :p
[21:06:28] <arjan> I can push it though
[21:06:30] <andreas.stenius> ah, I think I'm about to tackle it the other way around... :[
[21:06:34] <andreas.stenius> :p
[21:06:44] <Marc Worrell> Andreas, that might be hard
[21:06:45] <arjan> andreas.stenius: what do you mean?
[21:07:14] <Marc Worrell> we did generate other functions and have many constructs that ErlyDTL doesn't support
[21:07:26] <andreas.stenius> by adding the zotonic stuff to a new fork of erlydtl...
[21:07:36] <andreas.stenius> but that was just an idea I'm looking into
[21:07:44] <Marc Worrell> it might be easier to go from ours to theirs and then let them take ours :p
[21:07:44] <andreas.stenius> Marc Worrell: I know...
[21:07:56] <arjan> I think our erlydtl has more features
[21:08:00] <Marc Worrell> we have also gems like #id
[21:08:14] <arjan> or maybe they have similar features but the projects are diverged
[21:08:25] <arjan> it would be nice to somehow sync and contribute back
[21:08:42] <Marc Worrell> they actually didn't do that much since we left off
[21:08:56] <Marc Worrell> lately a bit more, because of Chicago Boss
[21:08:59] <andreas.stenius> well, my idea is since I have the history (well, most of it), it would be managable to merge them "hand-in-hand"....
[21:09:20] <arjan> that's a challenge..
[21:09:28] <Marc Worrell> a big one :p
[21:09:41] <Marc Worrell> could be hundreds of edits
[21:09:56] <Marc Worrell> the thing is very different by now
[21:10:14] <Marc Worrell> maybe you can diff without white spaces, then you will see
[21:10:29] <andreas.stenius> I may give up, but not until I've looked at it a little closer :p
[21:10:57] <Marc Worrell> at least you will learn more about the diffs and what the beast does
[21:11:09] <andreas.stenius> yep ;)
[21:11:12] <Marc Worrell> it is quite interesting
[21:11:27] <andreas.stenius> yeah, I like spending time in the template engine...
[21:11:28] <Marc Worrell> and there were some obvious omissions in the original ErlyDTL
[21:12:22] <andreas.stenius> Marc, btw, completely unrelated, did you glance over the rbac stuff?
[21:12:49] <andreas.stenius> It feels like I could pull it off without extra columns in the rsc table (for better or worse?)
[21:13:10] <arjan> I did actually look at it
[21:13:19] <arjan> you want to do it all with edges and categories?
[21:13:40] <arjan> maybe it is good to label the edges in your diagram with the predicates youre gonna use
[21:13:44] <andreas.stenius> feels more flexible...
[21:13:51] <arjan> yes, I like that idea
[21:13:54] <andreas.stenius> ah, right. I forgot to do that.
[21:13:59] <arjan> labellng explains the relationships between the categories
[21:14:19] <andreas.stenius> yep, I had them in my pencil version (not scanned), and in the text
[21:14:21] <arjan> like I did here, http://zotonic.com/docs/manuals/datamodel/example.html
[21:14:23] <arjan> ah
[21:14:38] <andreas.stenius> exactly
[21:14:40] <arjan> the domain model as I proposed to call it
[21:14:48] <andreas.stenius> yeah, funny that
[21:14:53] <arjan> :)
[21:15:00] <andreas.stenius> we came up with using domain on the same day... :d
[21:15:09] <arjan> uhu
[21:15:22] <arjan> not entirely the same tho :P
[21:15:36] <arjan> is a domain (in your sense) comparable to a "content group"?
[21:15:51] <andreas.stenius> uhm... maybe?
[21:16:02] <andreas.stenius> or, "resource group"... ?
[21:16:05] <arjan> exactly
[21:16:18] <Marc Worrell> Andrea, it is on my list to do!
[21:16:19] <andreas.stenius> or is "content group" some form of terminology that is commonly known... ?
[21:16:22] <arjan> that was the only part that was not clear to me
[21:16:24] <arjan> nah
[21:16:25] <arjan> :P
[21:16:52] <andreas.stenius> in the first sketch I simply called it Group.
[21:16:57] <arjan> marc would remember the term "axo"
[21:16:58] <arjan> :P
[21:17:00] <andreas.stenius> then it becam domain...
[21:17:06] <andreas.stenius> lol
[21:17:17] <arjan> domain is a nice term
[21:17:18] <andreas.stenius> I found axo, aro, aco and whatnot confusing
[21:17:26] <andreas.stenius> indeed
[21:17:28] <arjan> yeah
[21:17:34] <arjan> lets not introduce them in zotonic :P
[21:17:41] <andreas.stenius> agreed :p
[21:17:58] <arjan> im off for the night now, gonna watch boardwalk empire :p
[21:18:03] <arjan> talk to you tomorow
[21:18:19] <andreas.stenius> yeah, see you :)
[21:18:22] arjan leaves the room
[21:24:48] <Marc Worrell> pushed the elif and with
[21:24:52] <Marc Worrell> (and docs)
[21:24:59] <andreas.stenius> :)
[21:25:45] <Marc Worrell> a bit closer again to the real DTL :p
[21:26:52] <Marc Worrell> they still have all their filters in the same repo
[21:27:05] <Marc Worrell> (looking at https://github.com/evanmiller/erlydtl/commits/master/)
[21:27:54] <andreas.stenius> in a single file, by the looks of it...
[21:28:02] <Marc Worrell> uh oh
[21:28:17] <andreas.stenius> https://github.com/evanmiller/erlydtl/blob/master/src/erlydtl_filters.erl
[21:29:01] <andreas.stenius> a hefty 1200 line file with filters.... :p
[21:29:02] <Marc Worrell> they have ifchanged now
[21:29:07] <andreas.stenius> yep
[21:29:11] <Marc Worrell> maybe a good candidate to copy
[21:29:29] <Marc Worrell> and regrou[
[21:29:32] <Marc Worrell> regroup
[21:31:37] <Marc Worrell> yeah, we should move over some stuff from the original erlydtl
[21:31:48] <Marc Worrell> will be a day session or so, then we should be ok again
[21:31:49] <andreas.stenius> Do you monitor their commits for bugfixes that might apply to our version too...? or have we come too far away for that to be relevant?
[21:32:27] <Marc Worrell> we are not that far, only lots of extra stuff and the #context (ZpContext) support
[21:32:37] <Marc Worrell> and many fixes on our version, that can't be ported back
[21:35:50] <andreas.stenius> ok
[21:37:05] <andreas.stenius> I'll have a read-through of both lines of history so I get a better picture of it all... as it is I have a pretty good picture of the zotonic side, since that is the one I've been reading on and off during my zotonic development...
[21:38:02] <Marc Worrell> for some other project we are working on Arjan actually made the erlyzdtl, because erlydtl was missing too many nice features.
[21:38:02] <andreas.stenius> uhm, pretty good overview picture, at least :p (many of the deatils are still in the unknown.. )
[21:38:32] <andreas.stenius> ah :)
[22:10:14] <andreas.stenius> time for bed....
[22:10:48] andreas.stenius leaves the room
[22:16:19] Ilya Rezvov leaves the room
[23:42:33] <Marc Worrell> Maybe we should do something with Server-Side Events, as they work better than Comet
[23:42:51] <Marc Worrell> But then, all browsers supporting SSE also support Websockets….
[23:43:59] <Marc Worrell> http://en.wikipedia.org/wiki/Server-sent_events
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!