Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:21566
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Reply-To: <Jared.Williams1@ntlworld.com>
To: "'Sara Golemon'" <pollita@php.net>,
	<internals@lists.php.net>
Date: Wed, 18 Jan 2006 10:55:02 -0000
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
In-Reply-To: <002301c61c05$626f66c0$7d051fac@stumpy>
thread-index: AcYcBRWimt7zFg7/RcCBoa4WiNBUAgAFkbMg
Message-ID: <20060118105324.JTIT10196.aamta09-winn.ispmail.ntl.com@win2ks>
Subject: RE: [PHP-DEV] Naming Arguments
From: jared.williams1@ntlworld.com ("Jared Williams")
References: <002301c61c05$626f66c0$7d051fac@stumpy>

 
> From the perspective of providing reflection in general and 
> giving something for editors with code completion to use in 
> specific, would there be an interrest in backfilling the 
> arg_info structs for internal functions (both core and 
> otherwise) with argument naming and type hinting even where 
> it's not technically needed?
> 
> On the con side, this does add memory usage and processing 
> time for no direct production benefit.  Any utility which 
> wanted this degree of introspection *could* parse the XML 
> protos in phpdoc and get the same (or
> better) information, so it doesn't really gain that much for 
> what it costs.

Can you produce the arg_info structs from parsing the XML protos?

> 
> Do please forget about the work required to fill in this 
> information.  There are more than enough volunteers to do the 
> grunge work(and frankly it'd be tough for 'em to get it 
> horribly wrong), this is just a question of: "Would there be 
> an interrest in folding this data into the core?" and secondly: 
> "Should new extension writers be encouraged to fill this 
> information in?".

Perhaps some  tool to generate boiler plate docbook from introspection would be enough encouragement?

Jared