Repository navigation
Delegate conflicts with exprecs #3
Description
I tried to use exprecs with the Delegate annotation and exprecs fails when executing code. For example:
-module(test).
-compile({parse_transform, exprecs}).
-export_records([myrecord]).
-include_lib("annotations/include/annotations.hrl").
-include("records.hrl").
-delegate([{delegate, ["id", "name"]},
{args, ['$1']},
{arity, 1}]).
get(Property, #myrecord{} = Rec) when is_atom(Property) ->
'#get-'(Property, Rec).
When I try to do test:id(Rec), I get:
** exception error: undefined function test:'#exported_records-'/1
'#exported_records-'/0 exists but not /1 and when you change the args in the delegate tag it affects this value. If you pass no args, no matter what you put in the get function yields '#exported_records-'/0.
Metadata
Metadata
Assignees
Labels
No labels
Activity
hyperthunk commentedon May 23, 2012
Shouldn't that argument spec be more like:
That way, the property (name) gets passed correctly. Although on my machine, I was unable to make this generate any delegate code at all, but then I have just rebuilt a new laptop.
Also, I'm a little hazy on whether or not the
$narguments array is indexed from zero or one - you'll have to check the code to make sure.If this is still bothering you comment back and I will get a look at it some time this week.
Cheers,
Tim
aberman commentedon May 23, 2012
Thanks for the reply, Tim. That doesn't fix the issue. If you pass two values in the args param you then get "** exception error: undefined function test:'#exported_records-'/2." No matter what you pass in the args gets passed to the exported_records function. I'm unclear why the delegate is not attached to the "get" function. Looking at the delegate code I notice that you don't have an -annotation('function') at the top of the file. I don't know if that's by design or not.
Thanks again!
Andrew
aberman commentedon May 23, 2012
I was just now able to fix it by moving the exprecs code below the annotations include. This seems like a band-aid fix to me, but it's working now, so I'll leave it up to you if you want to look at this or not. Close at will :-) Thanks again for your hard work on this stuff!
hyperthunk commentedon May 23, 2012
Thanks for keeping me in the loop. I do believe that based on what you've said in that last comment, that there is a clash between the two parse transforms. I will leave this issue open until I've had time to investigate properly, but either way I'm glad you've got it working! :)