Display error on valid JSDoc syntax

[hexode]

Invalid JSDoc @param at ./js/components/validator.userphone.js
429 | * @params {...*} [args]
---------------^
This is message is incorrect because this is syntax allowed by jsdoc.

http://usejsdoc.org/tags-param.html - search Allows a parameter to be repeated

[mdevils]

I can accept pull-request for this behavior.

[qfox]

@mdevils I'll do it if you assign it to me ;-)

[mdevils]

If you've got a good idea on implementing this, I would be glad to accept a pull-request ;-)

[qfox]

Actually, I think we could parse it more accurately and store result in Function* nodes. Is it possible?
Probably, using jsdoc3/jsdoc or OrgaChem/jsdoctypeparser or something else.

[mdevils]

Not sure about storing in Function* nodes, but I'm totally agree that we need better JSDoc parsing approach.

Code at #75 looks messy, makes me unhappy to accept pull-request)). We need a good jsdoc parser with a good object-representation. With or without external libraries.

[qfox]

Okay, I'll to use some jsdoc parsers tonight. I think we should accept all kinds of jsdoc types? http://orgachem.github.io/jsdoctypeparser/

[mdevils]

I'm not sure about type parsing. Do we need it?

[qfox]

I think it should be pretty useful to check it up as fast as possible. Afais lint-like tools pretty useful when they can check as much as they can to prevent buggy behaviour of software. Probably JSCS is not a place to check doctypes and type of resulting value but it's another question.

[mrjoelkemp]

What's the verdict on this?

@mikesherov @zxqfox Should we deprecate this rule within jscs and inform users to use https://github.com/jscs-dev/jscs-jsdoc instead? I can create another issue for that task and put a PR together if you guys agree.

[qfox]

@mrjoelkemp probably yes. I've fixed this in plugin and I'm looking for issues.

[mikesherov]

yes, we should deprecate in 1.7, remove in 2.0

On Thu, Sep 18, 2014 at 11:37 PM, Alex Yaroshevich <notifications@github.com

wrote:

@mrjoelkemp https://github.com/mrjoelkemp probably yes. I've fixed this
in plugin and I'm looking for issues.

Reply to this email directly or view it on GitHub
#59 (comment).

Mike Sherov