How does can-connect do the equivalent of can.Model.parseModel
nvm
Unsure what the best/correct way is to define an edit button on a model.
(in a #list)
I could do ($click}="edit(id)" but that implies I'd need to search for the desired model in the list. (Or is this normal procedure?)
StealJS training on Wednesday: https://plus.google.com/u/1/b/117481170193954630503/events/c94t89oatm16nkk79o3ent4lfeo
I would however need to use %scope._parent , is is that ok ? or is there a correct wait to look into the parent ?
ok "../"
curious what people do about setting up test data? can-fixture seems to be tuned to ajax calls. But what about for non-ajax functions. For example, I am writing a unit test for a post-processing function, so it's not making the ajax call, rather it gets passed the response data (or a portion thereof). Is just using a direct call to
store.get recommended? Or is there a better way?
you mean in your unit test, using the
store that is exported from your fixture?
that works, but you could also just import the data from a json file
import it into your fixture and into your test
I don’t know that either way is drastically better
import myJson from ‘my-json-file.json’;
do recommend keeping the json in the same folder as the test code or in the fixtures folder? I am guessing fixture makes more sense
I put it in the modlet folder sometimes to use in demo files
also I have noticed that you (I think it's you) have been using mocha. So far I have just used qunit since it's the default and I am used to that style from other languages. What do you find preferential about mocha?
there’s honestly not a huge difference
I mostly prefer mocha because of how they do async tests
the
QUnit.start() stuff has always confused me
but you don’t even have to do that in qunit anymore I don’t think
qunit / jasmine / mocha all basically have the same API now
you can use mocha directly in node
so that’s a huge benefit if you’re doing js on the server also
since you can use the same test style for both