rplevy · February 27, 2012 00:06
diff --git a/midjefreenode-2012-02-26 b/midjefreenode-2012-02-26
 18:35 *** rob`` JOIN
 18:35 *** NAMES rob`` ckirkendall alexbaranosky cch1 SeanTAllen
                duck1123 @devn 
 18:35 /part
 18:35 *** rob`` PART #midje 
 18:36 *** rplevy JOIN
 18:36 *** NAMES rplevy ckirkendall alexbaranosky cch1 SeanTAllen
                duck1123 @devn 
 18:36 <ckirkendall> alexbaranosky: I haven't thought about
                    reporting at all yet.  Still trying to get my
                    baring with the Midje core code.
 18:37 <alexbaranosky> ckirkendall, I can walk you though it if you
                      want, if you think that would help
 18:37 <rplevy> just finishing watching this:
               http://vimeo.com/36579366 really good stuff
 18:37 <rplevy> sfw
 18:37 <alexbaranosky> I only watched the first 3rd or so
 18:38 <alexbaranosky> safe for work?
 18:38 <alexbaranosky> lol
 18:38 <rplevy> worth watching all of it
 18:38 <alexbaranosky> but everything should be in text
 18:38 <alexbaranosky> videos are for sissies
 18:39 <ckirkendall> alexbaranosky: I have traced through the code
                    and I see how background and fakes are setup
                    through bindings that wrap the call form. I get
                    how the structure of sweet -> semi-sweet ->
                    unprocessed.
 18:39 <rplevy> so ckirkendall what do you think about the
               experimental cljs support as it stands so far
 18:40 <alexbaranosky> ckirkendall, tha sounds about right.
                      Reporting happens in unprocessed after it
                      evaluates the fact inside expect*
 18:41 <alexbaranosky> I guess isnce it happens after the evaluation
                      then reporting should be safe
 18:43 <alexbaranosky> rplevy, https://gist.github.com/1919623
 18:44 <alexbaranosky> that illustrates how fakes won't work
 18:44 <rplevy> fakes?
 18:44 <alexbaranosky> provided
 18:45 *** ckirkendall_ JOIN
 18:45 <rplevy> why doesn't it work?
 18:45 *** ckirkendall QUIT Ping timeout: 245 seconds
 18:45 <rplevy> I don't know how provided works
 18:45 <ckirkendall_> sorry got kicked off again
 18:45 <rplevy> please explain
 18:45 <rplevy> (how provided works)
 18:45 <alexbaranosky> rplevy, it alter-var-roots them essentially
 18:46 <ckirkendall_> rplevy: alex and I were discussing how to
                     expand the scope of what is executed in repl
                     to include the entire fact form.
 18:46 <rplevy> so that the provided is executed in the cljs repl?
 18:47 <ckirkendall_> rplevy: no
 18:47 <rplevy> then what?
 18:47 <ckirkendall_> rplevy: but it needs to be, right now it is
                     setup outside and then you execute the call to
                     repl inside the binding created for fakes.
 18:48 <rplevy> I don't even understand the problem, so that doesn't
               mean anything to me yet
 18:48 <ckirkendall_> rplevy: sorry miss understood
 18:48 <alexbaranosky> I thin rplevy was right
 18:48 <alexbaranosky> we need to set up the provideds from within
                      cljs, no?
 18:49 *** ckirkendall JOIN
 18:50 <ckirkendall> I am having internet issues so I keep kicked
                    off is there a log for midje
 18:50 <ckirkendall> a log for the channel
 18:50 <rplevy> that doesn't seem difficult, if the provided is done
               similarly to how facts work
 18:51 <rplevy> nested facts might be a deeper conundrum, but I
               haven't even gone there conceptually
 18:52 <rplevy> are you logging this channel alexbaranosky?
 18:52 *** ckirkendall_ QUIT Ping timeout: 245 seconds
 18:52 <alexbaranosky> no, what do I need to do to log it?
 18:52 <rplevy> I don't know, I've never run an IRC channel
 18:52 <alexbaranosky> be neither
 18:54 <alexbaranosky> I have to get going, or at least can't be
                      guaranteed to be in front of the computer
 18:54 <alexbaranosky> I think if we put our heads to it we may be
                      able to work around the issues, but it sounds
                      like it would require you guys to delve
                      deeper into the Midje source.
 18:54 <rplevy> so have you looked into the provided thing?
 18:55 *** ckirkendall QUIT Ping timeout: 245 seconds
 18:55 <alexbaranosky> I haven't looked into it, other than seeing
                      Creighton's gist
 18:55 <rplevy> it would be good if we could make a list of
               everything that doesn't work and create issues
 18:55 <alexbaranosky> it soudns like Midje is "alter var rooting"
                      the var from CLJ, but the code is run in
                      CLJS, and so the vars aren't the same
 18:56 <alexbaranosky> well that gist shows oe thing that doesn't
                      work
 18:56 <rplevy> I think the experimental support can be merged into
               main branch as soon as that bug is fixed though, it
               doesn't need to support all features right away, no?
 18:57 <alexbaranosky> it could be incorporated as a test/alpha
                      concept
 18:57 <rplevy> provided might be low-hanging fruit though
 18:57 <alexbaranosky> but Midje without provideds is pretty weak
 18:57 <rplevy> if it's merged in it doesn't mean it's released yet,
               how often are releases done
 18:58 <alexbaranosky> it soudns like we need to wrap evaluate fact
                      calls in cljs instead of doing it way down in
                      sem-sweet
 18:58 <alexbaranosky> SNAPSHOTS? or say 1.3.3 ?
 18:58 <rplevy> dunno
 18:58 <alexbaranosky> there hasn't been a release ina  while
 18:58 <rplevy> ok
 18:58 <alexbaranosky> SNAPSHOTS happen all the tie though
 18:59 <rplevy> the thing about loading the cljs file before every
               assertion might be a problem too
 18:59 <alexbaranosky> but as long as our stuff doesn't break
                      anythign its ok to have new fancy half
                      working stuff in a SNAPSHOT I think
 18:59 <alexbaranosky> well we could load once per namespace easily
                      enough right?
 18:59 <rplevy> that makes sense
 18:59 <rplevy> how would you recommend doing that?
 19:00 <alexbaranosky> an atom??
 19:00 <alexbaranosky> any other ideas?
 19:00 <alexbaranosky> set the atom, and if its already set don't
                      load it?
 19:01 *** ckirkendall_ JOIN
 19:01 <alexbaranosky> ok I've got to go, but I'll be around , just
                      not glued to my seat
 19:01 <ckirkendall_> rplevy: you still on I think I am on a better
                     connection now.
	18:35 *** rob`` JOIN
	18:35 *** NAMES rob`` ckirkendall alexbaranosky cch1 SeanTAllen
	duck1123 @devn
	18:35 /part
	18:35 *** rob`` PART #midje
	18:36 *** rplevy JOIN
	18:36 *** NAMES rplevy ckirkendall alexbaranosky cch1 SeanTAllen
	duck1123 @devn
	18:36 <ckirkendall> alexbaranosky: I haven't thought about
	reporting at all yet. Still trying to get my
	baring with the Midje core code.
	18:37 <alexbaranosky> ckirkendall, I can walk you though it if you
	want, if you think that would help
	18:37 <rplevy> just finishing watching this:
	http://vimeo.com/36579366 really good stuff
	18:37 <rplevy> sfw
	18:37 <alexbaranosky> I only watched the first 3rd or so
	18:38 <alexbaranosky> safe for work?
	18:38 <alexbaranosky> lol
	18:38 <rplevy> worth watching all of it
	18:38 <alexbaranosky> but everything should be in text
	18:38 <alexbaranosky> videos are for sissies
	18:39 <ckirkendall> alexbaranosky: I have traced through the code
	and I see how background and fakes are setup
	through bindings that wrap the call form. I get
	how the structure of sweet -> semi-sweet ->
	unprocessed.
	18:39 <rplevy> so ckirkendall what do you think about the
	experimental cljs support as it stands so far
	18:40 <alexbaranosky> ckirkendall, tha sounds about right.
	Reporting happens in unprocessed after it
	evaluates the fact inside expect*
	18:41 <alexbaranosky> I guess isnce it happens after the evaluation
	then reporting should be safe
	18:43 <alexbaranosky> rplevy, https://gist.github.com/1919623
	18:44 <alexbaranosky> that illustrates how fakes won't work
	18:44 <rplevy> fakes?
	18:44 <alexbaranosky> provided
	18:45 *** ckirkendall_ JOIN
	18:45 <rplevy> why doesn't it work?
	18:45 *** ckirkendall QUIT Ping timeout: 245 seconds
	18:45 <rplevy> I don't know how provided works
	18:45 <ckirkendall_> sorry got kicked off again
	18:45 <rplevy> please explain
	18:45 <rplevy> (how provided works)
	18:45 <alexbaranosky> rplevy, it alter-var-roots them essentially
	18:46 <ckirkendall_> rplevy: alex and I were discussing how to
	expand the scope of what is executed in repl
	to include the entire fact form.
	18:46 <rplevy> so that the provided is executed in the cljs repl?
	18:47 <ckirkendall_> rplevy: no
	18:47 <rplevy> then what?
	18:47 <ckirkendall_> rplevy: but it needs to be, right now it is
	setup outside and then you execute the call to
	repl inside the binding created for fakes.
	18:48 <rplevy> I don't even understand the problem, so that doesn't
	mean anything to me yet
	18:48 <ckirkendall_> rplevy: sorry miss understood
	18:48 <alexbaranosky> I thin rplevy was right
	18:48 <alexbaranosky> we need to set up the provideds from within
	cljs, no?
	18:49 *** ckirkendall JOIN
	18:50 <ckirkendall> I am having internet issues so I keep kicked
	off is there a log for midje
	18:50 <ckirkendall> a log for the channel
	18:50 <rplevy> that doesn't seem difficult, if the provided is done
	similarly to how facts work
	18:51 <rplevy> nested facts might be a deeper conundrum, but I
	haven't even gone there conceptually
	18:52 <rplevy> are you logging this channel alexbaranosky?
	18:52 *** ckirkendall_ QUIT Ping timeout: 245 seconds
	18:52 <alexbaranosky> no, what do I need to do to log it?
	18:52 <rplevy> I don't know, I've never run an IRC channel
	18:52 <alexbaranosky> be neither
	18:54 <alexbaranosky> I have to get going, or at least can't be
	guaranteed to be in front of the computer
	18:54 <alexbaranosky> I think if we put our heads to it we may be
	able to work around the issues, but it sounds
	like it would require you guys to delve
	deeper into the Midje source.
	18:54 <rplevy> so have you looked into the provided thing?
	18:55 *** ckirkendall QUIT Ping timeout: 245 seconds
	18:55 <alexbaranosky> I haven't looked into it, other than seeing
	Creighton's gist
	18:55 <rplevy> it would be good if we could make a list of
	everything that doesn't work and create issues
	18:55 <alexbaranosky> it soudns like Midje is "alter var rooting"
	the var from CLJ, but the code is run in
	CLJS, and so the vars aren't the same
	18:56 <alexbaranosky> well that gist shows oe thing that doesn't
	work
	18:56 <rplevy> I think the experimental support can be merged into
	main branch as soon as that bug is fixed though, it
	doesn't need to support all features right away, no?
	18:57 <alexbaranosky> it could be incorporated as a test/alpha
	concept
	18:57 <rplevy> provided might be low-hanging fruit though
	18:57 <alexbaranosky> but Midje without provideds is pretty weak
	18:57 <rplevy> if it's merged in it doesn't mean it's released yet,
	how often are releases done
	18:58 <alexbaranosky> it soudns like we need to wrap evaluate fact
	calls in cljs instead of doing it way down in
	sem-sweet
	18:58 <alexbaranosky> SNAPSHOTS? or say 1.3.3 ?
	18:58 <rplevy> dunno
	18:58 <alexbaranosky> there hasn't been a release ina while
	18:58 <rplevy> ok
	18:58 <alexbaranosky> SNAPSHOTS happen all the tie though
	18:59 <rplevy> the thing about loading the cljs file before every
	assertion might be a problem too
	18:59 <alexbaranosky> but as long as our stuff doesn't break
	anythign its ok to have new fancy half
	working stuff in a SNAPSHOT I think
	18:59 <alexbaranosky> well we could load once per namespace easily
	enough right?
	18:59 <rplevy> that makes sense
	18:59 <rplevy> how would you recommend doing that?
	19:00 <alexbaranosky> an atom??
	19:00 <alexbaranosky> any other ideas?
	19:00 <alexbaranosky> set the atom, and if its already set don't
	load it?
	19:01 *** ckirkendall_ JOIN
	19:01 <alexbaranosky> ok I've got to go, but I'll be around , just
	not glued to my seat
	19:01 <ckirkendall_> rplevy: you still on I think I am on a better
	connection now.