[Openmcl-devel] PDFKit crashes

Ron Garret ron at awun.net
Thu May 28 08:04:00 UTC 2009


That's pretty slick, though I have to confess a certain dismay that  
there seem to be three different conventions for translating lisp  
symbol names into ObjC method names (corresponding roughly to some- 
objc-method, #/SomeOBJCMethod, and <s>ome<objcm>ethod.

I see in the docs where it describes how to use make-instance to  
instantiate ObjC classes it says:

"In some special cases, such as loading an ns:ns-window-controller  
from a .nib file, it may be necessary for you to pass the instance  
itself as one of the parameters to the initialization method."

Can you be a little more specific about the circumstances under which  
these special cases arise?

And as long as I'm asking random ObjC bridge questions, the 1.3  
release notes say:

"The Objective-C bridge no longer converts lisp strings to NSString  
instances automatically."

Why not?  I thought that was a handy feature.  Is it a memory  
management issue?

rg

On May 28, 2009, at 12:15 AM, Gary Byers wrote:

>
>
> On Thu, 28 May 2009, R. Matthew Emerson wrote:
>
>>
>> On May 28, 2009, at 2:06 AM, Ron Garret wrote:
>>
>>> It turns out that I've been forgetting to #/init things all over the
>>> place but things still mostly work for me.  Am I just getting lucky?
>>> Should #/init always be called when #/alloc is called?
>>
>> Yes, either #/init or some more specific initializer, e.g. #/
>> initWithFrame:, etc.
>>
>> There's also the #/new method, which combines #/alloc and #/init into
>> one, but using it is apparently not considered stylish any more.
>
> I personally consider MAKE-INSTANCE to be at least somewhat stylish.
>
> (MAKE-INSTANCE class) == (#/new class) == (#/init (#/alloc class))
>
> (MAKE-INSTANCE class {:with-pseudo-initarg-keyword value}+)
>
> uses the provided set of keywords to construct an #/init... function
> name (there's some caching involved ...) and does
>
> (#/thatInitFunction: (#/alloc class) , at values)
>
> It's not as high-level as it might look - a set of keywords that map
> to an init function name have to all be present and present in the
> canonical order for this to work - but things like:
>
> (make-instance window-class
>               :with-content-rect rect
>               :style-mask style-mask
>               :backing backing-type
>               :defer nil)
>
> may (or may not) look more readable than:
>
> (#/initWithContentRect:styleMask:backing:defer: (#/alloc window-class)
>                                                rect
>                                                style-mask
>                                                backing-type
>                                                nil)
>
>
> though (modulo typos) the two forms should have equivalent effect.
>
>>
>>
>>> If not, how do
>>> you know when you need to call #/init?  (Feel free to treat those as
>>> rhetorical questions.  I'm sure there's a Cocoa memory management
>>> primer somewhere that I can go read.)
>>
>> <http://developer.apple.com/documentation/Cocoa/Conceptual/CocoaFundamentals/CocoaObjects/CocoaObjects.html#/ 
>> /apple_ref/doc/uid/TP40002974-CH4-SW17
>> >
>>
>>
>> _______________________________________________
>> Openmcl-devel mailing list
>> Openmcl-devel at clozure.com
>> http://clozure.com/mailman/listinfo/openmcl-devel
>>
>>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clozure.com/pipermail/openmcl-devel/attachments/20090528/d1fbe1fc/attachment.html>


More information about the Openmcl-devel mailing list