<    April 2017    >
Su Mo Tu We Th Fr Sa  
 2  3  4  5  6  7  8  
 9 10 11 12 13 14 15  
16 17 18 19 20 21 22  
23 24 25 26 27 28 29  
00:19 glennpratt joined
00:40 GitHub136 joined
00:40 <GitHub136> [13sequel] 15jeremyevans pushed 8 new commits to 06master: 02https://git.io/vSReR
00:40 <GitHub136> 13sequel/06master 14492e21f 15Jeremy Evans: Allow optimized lookups and deletes for models using SQL::Identifier and SQL::QualifiedIdentifier values as the FROM table
00:40 <GitHub136> 13sequel/06master 142c3179a 15Jeremy Evans: Deprecate splitting symbols to look for qualified/aliased identifiers...
00:40 <GitHub136> 13sequel/06master 149ca96d4 15Jeremy Evans: Add note to from_block extension for integration in Sequel 5
00:40 GitHub136 left
01:05 glennpratt joined
03:28 filterfish_ left
03:50 glennpratt joined
05:07 mmun_ joined
05:08 JJonah__ joined
05:09 glennpratt joined
05:12 Caesium_ joined
06:25 glennpratt joined
07:17 ta_ joined
07:41 aidalgol joined
08:10 glennpratt joined
08:11 glennpratt joined
08:32 <Bish> :table__something will also be removed ;;
08:52 glennpratt joined
09:33 irclogger_com joined
09:33 Topic for
14:33 banditron joined
15:09 glennpratt joined
16:15 mkaito_ joined
16:15 mkaito joined
16:16 mkaito left
16:26 ta_ joined
18:02 banditron joined
19:20 jaequery joined
19:21 <jaequery> in before_save and after_save, is there a variable that differentiates your previous values and after save values?
19:31 <adam12> Maybe you could use around_save with a yield
19:31 <adam12> and some sort of assignment
19:31 <adam12> in AR I've used the "changes" property
19:31 <adam12> I'm not sure what Sequel uses to store changes.
19:34 <adam12> I'd say dirty plugin
20:02 <jeremyevans> jaequery: Sequel::Model by default does not stored changed values, it only stores which columns have been changed. You do probably want the dirty plugin
20:49 <jaequery> ok thanks!
20:51 <lopex> jeremyevans: that row_proc= deprecation should give an info to use with_row_proc now
20:52 <lopex> jeremyevans: just like that quoting advice info
20:57 <jeremyevans> lopex: The deprecation message is sent at a lower level. I think it would require an API modification to come up with a suggestion
20:57 <lopex> jeremyevans: so it does differ for quoting ?
20:58 <jeremyevans> lopex: Not sure. quote_identifiers= gives a similar deprecation warning that row_proc= does
20:58 <lopex> jeremyevans: the one with identifier_mangling extension
20:58 <jeremyevans> lopex: but you may be referring to a different deprecation warning
20:58 <lopex> yeah that different
20:59 <lopex> but helpful
20:59 <jeremyevans> lopex: that's handled at a different level. It's probably possible to improve the deprecation warnings by changing the argument used
21:00 <lopex> no problem
21:00 <jeremyevans> lopex: But in general mutating -> not_mutating is not a simple switch, if you are writing methods that mutate arguments and don't return them
21:00 <lopex> cool datasets are going to be immutable
21:02 <lopex> jeremyevans: yeah, that was the only discrepency wrt warnings
21:03 <lopex> jeremyevans: so will sequel 5 have more breaking changes ?
21:03 <jeremyevans> lopex: I'll try to improve the deprecation warnings for the dataset mutation methods before the next release
21:04 <lopex> jeremyevans: but for now updating was easy enough
21:04 <jeremyevans> lopex: Yes. Not sure if you've been following the commit log and Google Group since the last release, but yes, there are more deprecations since 4.45.0, and there are more changes planned
21:06 <lopex> will take a look
21:07 <lopex> quite a lot of that
21:08 <lopex> ah so lots of global thingies like that identifier_input_method are going to be db wise
21:09 <lopex> jeremyevans: btw that old mysql ruby dirver is already deprecated ? (non sequel related)
21:10 <lopex> afaik I had some build issues for ruby 2.4
21:10 <lopex> so all goes to mysql2 ?
21:11 <jeremyevans> lopex: I don't believe it is officially supported, but Sequel support for it will continue
21:11 <jeremyevans> lopex: mysql2 doesn't support everything that mysql supports currently
21:12 <lopex> ok
21:13 <lopex> jeremyevans: that old mysql driver build is affected by 2.4 lacking fixnums btw
21:13 <lopex> and hah
21:13 <jeremyevans> lopex: yep, but it's simple to fix. I know I patched it in the OpenBSD port :)
21:14 <lopex> yeah
21:14 <lopex> but removing rbFixnum is going to have an impact on exts for some time too
21:14 <lopex> jeremyevans: I've been using squirrel for some time
21:15 <lopex> jeremyevans: but now discovered dbeaver
21:15 <lopex> http://dbeaver.jkiss.org/
21:15 <lopex> not as good as intellij datagrip but it's really good
21:16 <lopex> downloads jdbc driver on it's own
21:16 <lopex> supports tunnelling in connection conf
21:16 <lopex> I'm thrilled
21:18 <lopex> jeremyevans: it also allow exports inbetween different dbs
21:18 <lopex> and full text search for db is a killer
21:40 Guest60 joined
21:41 <Guest60> Tests were w/out fixtures.. but I miss the simplicity of them...
21:50 glennpratt joined