Hopbot log for 2008-04-29 - Helma IRC channel: #helma on irc.freenode.net

2008-04-29:

[9:08] <bslivka> Whew, I have a voice
[9:08] <bslivka> I accept your minor IRC honor with gratitude
[9:08] <zumbrunn> ;-)
[9:10] <bslivka> So it turns out, my bug can be fixed, totally, by using object() instead of mounpoint()
[9:10] <bslivka> because as far as I can tell, the two are exactly the same except for one thing. object() is persisted, and mountpoint() isn't
[9:11] <bslivka> So it looks like the fact that in my test application, that a mountpoint isn't persisted, is perhaps intended behavior.
[9:11] <zumbrunn> maybe that's because mountpoint isn't supported with the built-in db
[9:12] <zumbrunn> but at least object mappings seem to do what the docs say
[9:12] <zumbrunn> even with the built-in db
[10:01] <bslivka> mountpoint works fine with the built-in db, it just isn't persisted, nor is it persisted with the SQL DB
[10:01] <bslivka> but it doesn't matter as much with the SQL DB because child collections do dynamic queries to get their objects
[10:01] <bslivka> with the built in DB the full graph must be in tact or helma won't be able to reconstruct it
[10:02] <bslivka> mountpoints not being persisted means it forms a break in the chain
[10:02] <bslivka> which is a problem with the builtin db, but you wouldn't notice in SQL land
[10:02] <bslivka> everything is bulit dynamically with the SQL mappings
[10:04] <bslivka> but so far as I can tell mountpoints behave exactly the same whether it's SQL or DB, it's just the consequences of that behavior that is different
[10:05] <bslivka> But it seems that some mountpoints are persisted with the built in DB
[10:05] <bslivka> So I suppose I don't fully understand what is happening there
[10:07] <zumbrunn> lol
[10:07] <zumbrunn> if we would ever figure out how exactly all this behaves and what is intended vs a bug, we could actually properly document this stuff
[10:09] <bslivka> that would be nice
[10:10] <zumbrunn> maybe a wiki page describing the behavior as seen would be a good start
[10:11] <bslivka> right. I've never been able to figure much logic in the Helma wiki. It seems to be used more like a chat board than as an attempt at documentation
[10:11] <bslivka> so finding useful information in it has always seemed like a lost cause
[10:12] <zumbrunn> more a scratchpad for future docs, yes
[10:13] <zumbrunn> at least newly added functionality is properly documented there
[10:14] <bslivka> I have just registered into the wiki
[10:14] <bslivka> Are there any protocols or can I just add a page and start typing?
[10:15] <zumbrunn> in the wiki area you can just add pages as you like
[10:16] <zumbrunn> the idea is to later collect the different bits from the /wiki/ area and collect the infos in a more structured way under /docs/
[10:16] <bslivka> that's pretty ambitious
[10:17] <bslivka> Perhaps I should file a bug first, if only to ask Hannes what the intended behavior is.
[10:17] <zumbrunn> yes, that's fine as well
[18:45] <zumbrunn> http://dev.helma.org/Sites+using+Helma/
[18:45] <zumbrunn> feel free to add whatever is missing
[18:49] <zumbrunn> http://dev.helma.org/wiki/Sites+using+Helma/
[18:49] <zumbrunn> (moved it to the wiki area)

 

 

In the channel now:

Logs by date: