Fail to handle string[]

[Steffan-Ennis]

I'm not sure if this is intentional behaviour. If it is it would be a good feature to be able to compare arrays and not just their items

Found a bad condition check on line 7 of conditionsMeet.js
line 7-8
if (!isObject(conditions) || !isObject(formData)) { toError(Rule ${conditions} with ${formData} can't be processed); }

line 21
return refVal.some(val => conditionsMeet(refCondition, val));

can call with bad arguments

Example condition, targets is a string[]
"conditions": { "targets": "_empty" }

[Steffan-Ennis]

ps I really like the library. Good work

[mavarazy]

Thanks, a lot, sorry for getting back late to this.
Can you clarify something

1. Condition
   "conditions": { "targets": "_empty" }
_empty is not part of predicate js library, that is used.
   It should display appropriate error when you try to use it this way

Try with this one
"conditions": { "targets": "empty" }

2. By referencing line 7 and related checks, do you mean you want to feed something other, than object to the rules engine?

I've opened an issue - #7, which should allow you to specify any conditional logic you want.

[Steffan-Ennis]

Hey, thanks for getting back to me.

I like your solution for extending the predicate logic. I took a very similar approach by extending the import predicate object.

I referenced line 23 because it would be nice to have the empty predicate ran against the array and its elements.

something like this.
if(Array.isArray(refVal)) { return refVal.some(val => conditionsMeet(refCondition, val)) || checkField(refVal, refCondition) }
and in checkField line 9
return fieldVal.some(val => checkField(val, rule)) || predicate[rule](fieldVal)

The feature could be very useful for associating rules against json-schema-form arrays. We have a use case for hiding an entire section of the form until one element has been added to the array.

[mavarazy]

@Steffan-Ennis Can you give me an example? So I would see how to better address this?

[mavarazy]

@Steffan-Ennis merged your request

[Steffan-Ennis]

Awesome, thanks.

[mavarazy]

@Steffan-Ennis I've updated your fix, should still work