Re: quick look question
See QLPreviewPanel http://developer.apple.com/library/mac/#documentation/Quartz/Reference/QLPreviewPanel_Class/Reference/Reference.html On Wed, Aug 3, 2011 at 6:10 AM, Rick Corteza rickcort...@gmail.com wrote: Hi again, I double-checked but isn't it that this one is just to create thumbnails? Actually running /usr/bin/qlmanage -p is exactly what I'm trying to do because I want to display short video clips not just still images. If only it didn't put [DEBUG] in the title bar it would probably be good enough! What would be the best approach to doing this because I need it to work on 10.5? Thanks! On Tue, Aug 2, 2011 at 7:00 PM, jonat...@mugginsoft.com jonat...@mugginsoft.com wrote: On 2 Aug 2011, at 11:39, Rick C. wrote: Hi all, I want to include this feature in my app and the way I want it to work is if a user selects a file in my table view and clicks a button it will open that file in a quick look panel just the way it works in Finder. Actually if there was only a way to trigger that easily like NSWorkspace or something I'd use it! Now I need to support 10.5 still and unfortunately I see a lot of it became formal in 10.6. But is it possible to achieve what I'm looking to do? I have download QuickLook Sketch sample code am I going in the right direction because I wasn't hoping to make any plugins. Any pointers would be great thanks! Checkout NSImage+QuickLook available at http://mattgemmell.com/source. This doesn't use the formal 10.6+ API but calls /usr/bin/qlmanage instead. Works fine on 10.5, 10.6 and 10.7 Regards Jonathan Mitchell Developer Mugginsoft LLP http://www.mugginsoft.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/rickcorteza%40gmail.com This email sent to rickcort...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Quick Look Preview with Scrollbar
On Sun, Feb 13, 2011 at 5:13 PM, Andrew Madsen and...@openreelsoftware.comwrote: I'm writing a Quick Look Plug-in to generate previews of my application's (Core Data) documents. I've got the generator working ok with one problem. If the document is large/long, and I return an preview that is say 5000 pixels high, Quick Look scales the preview image instead of putting it in a scroll view, making it impossible to see any detail. Is there some way to tell Quick Look to display the preview at its native width, then add a vertical scrollbar? For example, when Quick Looking long Microsoft Word or Excel documents, the Quick Look panel includes a vertical scrollbar. The generator uses QLPreviewRequestCreateContext to get a CGContext, and from there an NSGraphicsContext which it then draws into using a custom NSView subclass, in a similar manner to Apple's QuickLookSketch example code. Is generating HTML or multi-page PDF the only way to get a scrolling preview? Yes. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: QuickLook returning small image
When requesting a thumbnail, it's up to the underlying plug-in to honor the requested size. If icon == NO, Quick Look will return the plug-in's image result as is. If icon == YES, Quick Look will force the requested size (because it adds icon decoration which should match the requested size) - so if you request a 512x512 icon, you should always get a 512x512 image. Many plug-in just use a low-resolution image stored in the file as a Quick Look thumbnail. It's totally acceptable but nowadays, low-resolution should be more toward 512x512 than 128x128. Previews are more complex (generally slower to generate) and are meant to be displayed mainly in Quick Look panel but they also appear in Cover Flow, Column view, Info window and, starting with SnowLeopard, in live previews in Finder's icons. -- Julien On Thu, Jan 13, 2011 at 10:11 AM, aglee ag...@mac.com wrote: On Jan 13, 2011, at 01:06 PM, Kyle Sluder kyle.slu...@gmail.com wrote: Taking a while stab here: icon==YES? I get the same result (size-wise) whether icon is YES or NO. Joar pointed out off-list that Finder might be asking for a *preview* rather than a *thumbnail*. I totally missed that distinction and I bet it's the explanation. Thanks, --Andy ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: A Quick Look contribution and a question
Look at Apple provided code sample at:
http://developer.apple.com/library/mac/samplecode/QuickLookDownloader/Introduction/Intro.html
--
Julien Jalon
On Monday, January 3, 2011, Brad Stone cocoa-...@softraph.com wrote:
I'm submitting this code for anyone who needs a quick hack to get Quick Look
working. It's a hack because it's using AppleScript and the Quick Look
Server debug and management tool. If you send an array of paths this will
bring them up in a Quick Look window.
I don't think the Quick Look documentation is as robust as it can be, I'm
still trying to figure out how to do the same thing the right way.
- (void)quickLook:(NSArray *)srcPathArray {
//build the paths string
NSString *paths = @;
for (NSString *thisPath in srcPathArray) {
NSString *s = @;
if ([paths length] == 0) {
s = [NSString stringWithFormat:@(quoted form of
(POSIX path of \%...@\)), thisPath];
} else {
s = [NSString stringWithFormat:@ \ \ (quoted
form of (POSIX path of \%...@\)), thisPath];
}
paths = [paths stringByAppendingString:s];
}
NSAppleScript *quickLook = nil;
// do shell script (qlmanage -p (quoted form of (POSIX path of
srcPath)))
NSString *command = [NSString stringWithFormat:@do shell script
(\qlmanage -p \ %@),paths];
quickLook = [[NSAppleScript alloc] initWithSource:command];
[quickLook executeAndReturnError:nil];
}___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: respondsToSelector warning: may not respond
Z) ignore the warning On Fri, Nov 26, 2010 at 9:44 PM, Ken Thomases k...@codeweavers.com wrote: On Nov 26, 2010, at 12:27 PM, Mike Abdullah wrote: C) Typecast the object to a class that is known to implement -setOrdinal: D) Typecast the object to id. Regards, Ken ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Designated Initializer
NSWindowController doc states that you should invoke super's initWithWindow: or initWithWindowNibName: In your class’s initialization method, be sure to invoke on super either one of the initWithWindowNibName:... initializers or the initWithWindow: initializer This breaks the designated initializer pattern, which is bad (and indeed initWithWindowNibName: calls [self initWithWindow:]). Not catastrophic though if you look at the usual usage pattern for NSWindowController. -- Julien On Sat, Oct 30, 2010 at 10:32 PM, Richard Somers rsomers.li...@infowest.com wrote: Thanks for the insight. If calling any of super's initializers will work from the designated initializer, why then does Apple specifically say the designated initializer should begin by sending a message to super to invoke the designated initializer of its superclass? There must be a subtle issue here. --Richard Somers On Oct 30, 2010, at 1:12 PM, Dave Carrigan wrote: So just because a designated initializer didn't call super's designated initializer, it doesn't mean that super's designated initialer was not invoked; it was. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Why is NSPopUp broken in drop-down mode?
developers have been able to dictate in other platforms' UIs Welcome to the modern Mac platform where the frameworks try to ensure a coherent user experience, especially for basic controls. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Why is NSPopUp broken in drop-down mode?
Let me rephrase: same control should have the same behavior across applications. Cocoa framework enforce that rule. Disagreeing with Apple choices about how each control should work is an other story that I'm not sure is interesting for this list. If you want to develop for MacOS X, embrace the frameworks and read Apple HIG: http://developer.apple.com/library/mac/#documentation/UserExperience/Conceptual/AppleHIGuidelines/XHIGIntro/XHIGIntro.html%23//apple_ref/doc/uid/TP3894-TP6 -- Julien On Friday, October 1, 2010, G S stokest...@gmail.com wrote: Welcome to the modern Mac platform where the frameworks try to ensure a coherent user experience, especially for basic controls. How is this behavior coherent? As has already been noted, it's inconsistent between drop-down lists and combo boxes. Also, nothing indicates to the developer why the data-handling behavior of the control has changed simply because he chose an up or down style of opening. That's like changing the behavior of a control because you picked a beveled instead of a flat decorative box around it. And modern? You're kidding, right? The transition to OS X was Apple's chance to modernize many things about its UI, and it chose to squander that opportunity. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: autorelease: how does this work!? (if at all)
Is the compiler/runtime being clever enough to retain it because it is
going to be needed in the inner block (if so: very clever!)?
Yes, it is very clever.
When creating the block, the ObjC compiler also specifies the Object stored
into the block metadata. When the block is copied (which is done as soon as
you call dispatch_async), the copied block retains these objects.
Note that it only works for blocks and ObjectiveC objects with some
(logical) special cases (ivars, __block).
See:
http://developer.apple.com/mac/library/documentation/Cocoa/Conceptual/Blocks/Articles/bxVariables.html
--
Julien
On Fri, Jun 18, 2010 at 5:00 PM, Jonny Taylor j.m.tay...@durham.ac.ukwrote:
I've just been looking back at some code that has been working fine for me
for a while, and as far as I can see it shouldn't actually work! I'd be
interested for peoples' comments. The code is as follows:
dispatch_async(queue1,
^{
NSImage *theImage = [frame GetNSImage];
NSData *tiffRep = [theImage TIFFRepresentation];
dispatch_async(queue2,
^{
[tiffRep writeToFile:[NSString stringWithFormat:@
%...@%d.tif,
[[frame Camera] ExportFilePrefix],
[frame FrameNumber]]
atomically:YES];
});
[theImage release];
});
Work running on serial queue queue1 calculates a TIFF representation for
an image, and then schedules work on serial queue queue2 to write that
data to disk. What I can't work out is why tiffrep isn't autoreleased as
soon as the outer block completes. Is the compiler/runtime being clever
enough to retain it because it is going to be needed in the inner block (if
so: very clever!)? If not, am I just getting lucky here with exactly
when/how grand central does its autorelease cleanup? Or maybe the TIFF
representation and/or my frame data is still being retained elsewhere for a
while (possible, depending on how the thread timings work out...).
I'm pretty new to Cocoa so I'm keen to understand this properly - I'd be
interested to hear peoples thoughts on this: is what I am doing ok, or do I
need to add some explicit retain/releasing of tiffRep?
Cheers
Jonny___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: autorelease: how does this work!? (if at all)
[theImage release] is really suspicious. On Friday, June 18, 2010, Jens Alfke j...@mooseyard.com wrote: On Jun 18, 2010, at 9:20 AM, Mike Abdullah wrote: Aside from your actual question, I urge you to go back and read the Cocoa fundamentals. Your method names are totally wrong, I only saw one error: the method name “GetNSImage” should probably be “image” (method names should be lowercase unless they begin with a common acronym like “URL” or “TIFF”, and the prefix “get” is not used.) and the memory management guide covers all the details of -autorelease. Has it been extended to discuss objects referenced inside blocks? That’s a pretty tricky detail. —Jens___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
autorelease: how does this work!? (if at all)
Suspicious because it clearly does not follow ObjC coding style for memory management. If you have to call release here, it's likely because GetNSImage is incorrect by returning an object that needs to be released. If your code happens not to use C++ (or in very localized places), you should run Build Analyze (in Build menu) and see the static analyzer pointing to where your code might clearly be wrong wrt coding style and memory management. -- Julien On Friday, June 18, 2010, Jonny Taylor j.m.tay...@durham.ac.uk wrote: [theImage release] is really suspicious. Suspicious in what way? Are you saying I am using the wrong sort of implementation for my [frame GetNSImage], and I shouldn't be returning something that requires an explicit release? Because I can certainly confirm that in its current state it leaks memory unless I include the release - the function is returning an internally cached NSImage with its reference count incremented by one. To be fair the release is in an illogical place - it should immediately follow the call to [theImage TIFFRepresentation], but with the code as currently written, the release is definitely necessary... ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: How to Open PPT file in Cocoa Application
You can only use Quick Look through the Preview panel API. Search for QLPreviewPanel No preview view is accessible through API so far. File an enhancement request. -- Julien On Thursday, June 17, 2010, kalpana k kalpana.forum...@gmail.com wrote: Hi All, I am trying to open a ppt (power point) file in my cocoa application (leopard/snow leopard). I want to open it in a view inside my application. Open as a slide show view (ie: no editing is allowed) I don't want to open it as a separate application triggered from my application. Quicklook opens ppt files, So Can we achieve this using Quick look Framework? How to use it? Thanks Regards, K.Kalpana ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: menu madness with retain count
ObjectAlloc instrument is your friend. Configure it to record retains and releases. Much more accurate, much easier to understand what's going on. On Tuesday, April 27, 2010, Gary L. Wade garyw...@desisoftsystems.com wrote: On 04/27/2010 2:12 PM, Bill Bumgarner b...@mac.com wrote: On Apr 27, 2010, at 2:09 PM, Gary L. Wade wrote: On 04/27/2010 1:58 PM, Bill Bumgarner b...@mac.com wrote: Frankly, the -retainCount method should be deprecated and, eventually, removed. I wouldn't go THAT far; after all, when you're tracking a memory leak, checking your influence on the retain count is important to your investigation. Hopefully that's why the original poster is looking at it. The combination of leaks, zombies, heap, and malloc stack logging are much *much* more powerful and effective than trying to debug a leak, over-retain or under-retain with -retainCount. b.bum Yes, but how would you use those to determine why an Apple framework now chooses to retain a delegate (I'm referring to one particular one I discovered), thereby causing a retain cycle? It's not a memory leak in the sense that Instruments or leaks would ever catch it. Calling -retainCount immediately before and after the -setDelegate call is pretty much the only way. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: blocks and autorelease weirdness
I don't see anything wrong with the output being 1/2:
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc] init];* //
retain count is 1*
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain]; *// retain count is 2*
[aString *auto*release]; *// retain count is still 2
- object will be released sometimes in the future*
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);* // you can't assume the future release has already been
called*
});
});
});
}
On Mon, Apr 19, 2010 at 11:35 AM, Henk Kampman
henk.kamp...@secondmove.comwrote:
Have a look at the following code
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc] init];
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain];
[aString release];
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);
});
});
});
}
As expected the output is:
retainCount: 1
retainCount: 1
However when I change [aString release] to [aString autorelease] the output
shows:
retainCount: 1
retainCount: 2
Why?
BTW I first noticed this behavior when my application started leaking the
QTMovie objects I was creating in a block.
-
Henk___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: blocks and autorelease weirdness
I have to add that using dispatch_async on main queue within an NSApp will
automatically drain the autorelease pool.
AppKit drains the autorelease pool upon NSEvents dispatch.
On Mon, Apr 19, 2010 at 12:36 PM, Julien Jalon jja...@gmail.com wrote:
I don't see anything wrong with the output being 1/2:
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc] init];
* // retain count is 1*
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain]; *// retain count is
2*
[aString *auto*release]; *// retain count is still
2 - object will be released sometimes in the future*
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);* // you can't assume the future release has already been
called*
});
});
});
}
On Mon, Apr 19, 2010 at 11:35 AM, Henk Kampman
henk.kamp...@secondmove.com wrote:
Have a look at the following code
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc] init];
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain];
[aString release];
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);
});
});
});
}
As expected the output is:
retainCount: 1
retainCount: 1
However when I change [aString release] to [aString autorelease] the
output shows:
retainCount: 1
retainCount: 2
Why?
BTW I first noticed this behavior when my application started leaking the
QTMovie objects I was creating in a block.
-
Henk___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: blocks and autorelease weirdness
And of course, I need some rest:
On Mon, Apr 19, 2010 at 12:38 PM, Julien Jalon jja...@gmail.com wrote:
I have to add that *you can't assume* using dispatch_async on main queue
within an NSApp will automatically drain the autorelease pool.
AppKit drains the autorelease pool upon NSEvents dispatch.
On Mon, Apr 19, 2010 at 12:36 PM, Julien Jalon jja...@gmail.com wrote:
I don't see anything wrong with the output being 1/2:
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc] init];
* // retain count is 1*
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain]; *// retain count is
2*
[aString *auto*release]; *// retain count is still
2 - object will be released sometimes in the future*
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);* // you can't assume the future release has already been
called*
});
});
});
}
On Mon, Apr 19, 2010 at 11:35 AM, Henk Kampman
henk.kamp...@secondmove.com wrote:
Have a look at the following code
-(void) test
{
testString = NULL;
dispatch_async(dispatch_get_global_queue(0, 0), ^{
dispatch_async(dispatch_get_main_queue(), ^{
NSString* aString = [[NSMutableString alloc]
init];
NSLog(@retainCount: %d,[aString retainCount]);
testString = [aString retain];
[aString release];
dispatch_async(dispatch_get_main_queue(), ^{
NSLog(@retainCount: %d,[testString
retainCount]);
});
});
});
}
As expected the output is:
retainCount: 1
retainCount: 1
However when I change [aString release] to [aString autorelease] the
output shows:
retainCount: 1
retainCount: 2
Why?
BTW I first noticed this behavior when my application started leaking the
QTMovie objects I was creating in a block.
-
Henk___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: IKImageBrowserView IKImageView subclasses not getting called
SnowLeopard introduced a bunch of new public API to make customizing easier: On the view itself: - (void) setBackgroundLayer:(CALayer *) aLayer; - (void) setForegroundLayer:(CALayer *) aLayer; extern NSString * const IKImageBrowserGroupHeaderLayer; /* CALayer */ extern NSString * const IKImageBrowserGroupFooterLayer; /* CALayer */ on the browser cell: - (CALayer *) layerForType:(NSString *) type; -- Julien On Mon, Feb 15, 2010 at 5:42 PM, Jens Alfke j...@mooseyard.com wrote: On Feb 15, 2010, at 7:18 AM, Ashley Clark wrote: For objects that were saved in a NIB file -initWithFrame: is usually not what's called to recreate them (there are some exceptions that I don't remember off the top of my head). Since NIBs are essentially archives most of the views stored within them are recreated via -initWithCoder: Right. Setup code for objects from nibs should usually go in -awakeFromNib. As for the drawRect: override, it was my understanding that the IK*View objects didn't do any of their drawing in drawRect: but instead were done via CALayers. I could be misinformed on this point though. Right again, basically, although I think they use direct OpenGL for drawing. This unfortunately makes it extremely difficult to customize the display of these views, since the OpenGL surface covers up any regular drawing. I tried pretty hard to extend the image-browser a few years ago and eventually gave up. —Jens ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: IKImageBrowserView IKImageView subclasses not getting called
And you have an example: http://devworld.apple.com/mac/library/samplecode/ImageBrowserViewAppearance/index.html -- Julien On Mon, Feb 15, 2010 at 5:42 PM, Jens Alfke j...@mooseyard.com wrote: On Feb 15, 2010, at 7:18 AM, Ashley Clark wrote: For objects that were saved in a NIB file -initWithFrame: is usually not what's called to recreate them (there are some exceptions that I don't remember off the top of my head). Since NIBs are essentially archives most of the views stored within them are recreated via -initWithCoder: Right. Setup code for objects from nibs should usually go in -awakeFromNib. As for the drawRect: override, it was my understanding that the IK*View objects didn't do any of their drawing in drawRect: but instead were done via CALayers. I could be misinformed on this point though. Right again, basically, although I think they use direct OpenGL for drawing. This unfortunately makes it extremely difficult to customize the display of these views, since the OpenGL surface covers up any regular drawing. I tried pretty hard to extend the image-browser a few years ago and eventually gave up. —Jens ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Garbage Collection Docs Puzzle
Except -retain is more efficient under GC. -- Julien from his iPhone Le 26 janv. 2010 à 00:06, Dave Keck davek...@gmail.com a écrit : sending myData -self at the end of the method would make a lot more sense. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: if statement causing 32 Byte leak?
You have Instruments for that (ObjectAlloc instrument or Zombie/leaks templates). Also under certain conditions, you have the static analyzer. -- Julien from his iPhone Le 10 janv. 2010 à 18:13, Glenn L. Austin gl...@austin-soft.com a écrit : On Jan 10, 2010, at 8:01 AM, Scott Ribe wrote: Also, an isAutoReleased message would be worthless. At any point in time, you have no idea how many times library routines that you've called might have retained/released/autoreleased, nor should you care. Actually, it would be nice to have *in a debugging context* since trying to track down a pointer that is in the autorelease pool the number of times it has been retained when you do a release would be very beneficial. Yes, it would slow things down, but it would sure save a lot of time when trying to track down a spurious release! e.g. id pObj = [[[ObjectType alloc] init] autorelease]; ... do something here that doesn't retain the pObj, maybe return the value to a caller... [pObj release];// -- This should assert in a debug context, since it is already in the current autorelease pool. // [pObj autorelease] should also probably assert, since it is now in the current autorelease pool twice... What I ended up doing is progressively creating NSAutoreleasePools, running the code from that routine, then draining that pool until I found the culprit. Oh the joys of working on a large project with programmers of varying Cocoa skill levels. -- Glenn L. Austin, Computer Wizard and Race Car Driver http://www.austin-soft.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: if statement causing 32 Byte leak?
This is the purpose of the zombie instrument. Once you know what object is overreleased, you click on the arrow and voilà, you get the list of retain/release/autorelease so you can look where is the extra one. -- Julien from his iPhone Le 10 janv. 2010 à 19:19, Glenn L. Austin gl...@austin-soft.com a écrit : Zombie told us nothing. The project just didn't crash any more -- it only told us that the autoreleased object was a zombie. Gee, we already knew that... ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: NSApplicationLoad() can't connect to window server in user logged out.
ImageIO lets you load and save image files (See CGImageSource and
CGImageDestination)
CoreText provides text drawing facilities
CoreGraphics will give you bitmap contexts and compositing.
Of course, the CG API is a little bit more hardcore than AppKit but not that
much.
On Sat, Dec 19, 2009 at 3:13 AM, Mr. Gecko grmrge...@gmail.com wrote:
Can I do exactly what I'm doing there in that? I also need to place some
text in there which I have in the full code, just made my code shorter as an
example.
On Dec 18, 2009, at 8:07 PM, Julien Jalon wrote:
NSImage needs a connection to the Window Server in Leopard.
If you want to do some offscreen headless rendering, use CoreGraphics +
ImageIO
On Sat, Dec 19, 2009 at 2:21 AM, Mr. Gecko grmrge...@gmail.com wrote:
Ok, so I've done some more research and found this
http://developer.apple.com/mac/library/technotes/tn2005/tn2083.html#SECPRELOGINANDTRUST,
it basically says to run it in the loginwindow, but I can't do that with
apache which is what I'm running my code with. Because this is a personal
application I wouldn't mind doing things the hacking way and some how make
the windowserver trust apache.
On Dec 18, 2009, at 6:06 PM, Mr. Gecko wrote:
Hello I'm running 10.5 and I'm trying to do some image drawing with
NSImage in Terminal via SSH and it seems like every time I try, I get this
error
_RegisterApplication(), FAILED TO establish the default connection to
the WindowServer, _CGSDefaultConnection() is NULL.
2009-12-18 17:54:04.963 weather.mgm[655:10b] *** -[NSRecursiveLock
unlock]: lock (NSRecursiveLock: 0x121b30 '(null)') unlocked when not
locked
2009-12-18 17:54:04.973 weather.mgm[655:10b] *** Break on _NSLockError()
to debug.
2009-12-18 17:54:04.978 weather.mgm[655:10b] An uncaught exception was
raised
2009-12-18 17:54:05.009 weather.mgm[655:10b] Error (1002) creating
CGSWindow
2009-12-18 17:54:05.014 weather.mgm[655:10b] *** Terminating app due to
uncaught exception 'NSInternalInconsistencyException', reason: 'Error (1002)
creating CGSWindow'
2009-12-18 17:54:05.040 weather.mgm[655:10b] Stack: (
2513718512,
2536973548,
2513718272,
2513718328,
2470748316,
2470747564,
2470746864,
2470746400,
2470743808,
2470755712,
2470754620,
2470754184,
2470743272,
8540
)
I know this code I'm using works as it runs perfectly fine on Snow
Leopard via SSH but for some reason it doesn't work on Leopard, what could
be wrong here? How can I get this working on Leopard?
Basic code to this is
int main (int argc, const char * argv[]) {
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];
NSApplicationLoad();
NSImage *image = [[NSImage alloc] initWithContentsOfFile:@
/background.png];
NSSize imageSize = [image size];
[image lockFocus];
NSImage *type = [[NSImage alloc] initWithContentsOfFile:@
/icons/93x93/1.png];
[type drawInRect:NSMakeRect((imageSize.width-[type size].width)/2,
imageSize.height-[type size].height, [type size].width, [type size].height)
fromRect:NSMakeRect(0, 0, [type size].width, [type size].height)
operation:NSCompositeSourceOver fraction:1.0];
[image unlockFocus];
NSBitmapImageRep *imageRep = [NSBitmapImageRep
imageRepWithData:[image TIFFRepresentation]];
NSData *iconData = [imageRep representationUsingType:NSPNGFileType
properties:nil];
[iconData writeToFile:@/drawn.png automatcially:YES];
[image release];
[server release];
[pool drain];
return 0;
}
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: NSApplicationLoad() can't connect to window server in user logged out.
NSImage needs a connection to the Window Server in Leopard.
If you want to do some offscreen headless rendering, use CoreGraphics +
ImageIO
On Sat, Dec 19, 2009 at 2:21 AM, Mr. Gecko grmrge...@gmail.com wrote:
Ok, so I've done some more research and found this
http://developer.apple.com/mac/library/technotes/tn2005/tn2083.html#SECPRELOGINANDTRUST,
it basically says to run it in the loginwindow, but I can't do that with
apache which is what I'm running my code with. Because this is a personal
application I wouldn't mind doing things the hacking way and some how make
the windowserver trust apache.
On Dec 18, 2009, at 6:06 PM, Mr. Gecko wrote:
Hello I'm running 10.5 and I'm trying to do some image drawing with
NSImage in Terminal via SSH and it seems like every time I try, I get this
error
_RegisterApplication(), FAILED TO establish the default connection to the
WindowServer, _CGSDefaultConnection() is NULL.
2009-12-18 17:54:04.963 weather.mgm[655:10b] *** -[NSRecursiveLock
unlock]: lock (NSRecursiveLock: 0x121b30 '(null)') unlocked when not
locked
2009-12-18 17:54:04.973 weather.mgm[655:10b] *** Break on _NSLockError()
to debug.
2009-12-18 17:54:04.978 weather.mgm[655:10b] An uncaught exception was
raised
2009-12-18 17:54:05.009 weather.mgm[655:10b] Error (1002) creating
CGSWindow
2009-12-18 17:54:05.014 weather.mgm[655:10b] *** Terminating app due to
uncaught exception 'NSInternalInconsistencyException', reason: 'Error (1002)
creating CGSWindow'
2009-12-18 17:54:05.040 weather.mgm[655:10b] Stack: (
2513718512,
2536973548,
2513718272,
2513718328,
2470748316,
2470747564,
2470746864,
2470746400,
2470743808,
2470755712,
2470754620,
2470754184,
2470743272,
8540
)
I know this code I'm using works as it runs perfectly fine on Snow
Leopard via SSH but for some reason it doesn't work on Leopard, what could
be wrong here? How can I get this working on Leopard?
Basic code to this is
int main (int argc, const char * argv[]) {
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];
NSApplicationLoad();
NSImage *image = [[NSImage alloc] initWithContentsOfFile:@
/background.png];
NSSize imageSize = [image size];
[image lockFocus];
NSImage *type = [[NSImage alloc] initWithContentsOfFile:@
/icons/93x93/1.png];
[type drawInRect:NSMakeRect((imageSize.width-[type size].width)/2,
imageSize.height-[type size].height, [type size].width, [type size].height)
fromRect:NSMakeRect(0, 0, [type size].width, [type size].height)
operation:NSCompositeSourceOver fraction:1.0];
[image unlockFocus];
NSBitmapImageRep *imageRep = [NSBitmapImageRep
imageRepWithData:[image TIFFRepresentation]];
NSData *iconData = [imageRep representationUsingType:NSPNGFileType
properties:nil];
[iconData writeToFile:@/drawn.png automatcially:YES];
[image release];
[server release];
[pool drain];
return 0;
}
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: App Launches in Finder, Hangs While Launching in Debugger
It's not necessarily a memory management problem. As this happens very early in the application launch, when Launch Services uses your Info.plist to register the application, your problem might also be that an entry supposed to be a string is in fact an array. On Mon, Dec 14, 2009 at 9:09 AM, Joe The Programmer joetheappleprogram...@gmail.com wrote: On Dec 14, 2009, at 12:52 AM, Kyle Sluder wrote: Turning on GC is not the solution to your problem. Fixing your memory management bugs is the solution. On Dec 14, 2009, at 12:54 AM, Joar Wingfors wrote: For a trivial app it might be. For anything else, it would not be. GC and RC are fundamentally different memory management models, and converting an existing code base is never easy. You should not attempt it before you understand how GC and RC differs, before you've read up on how GC is supported in ObjC Cocoa, etc. Understood. I'll have a good read. Thanks. Could this memory management issue have existed even when this project was being built with Xcode 2.5 and the 10.4 SDK, or is it something that was introduced to build with Xcode 3.2.1?___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: App Launches in Finder, Hangs While Launching in Debugger
And indeed, your plist is broken. CFBundleDisplayName should be a string not an array. It seems that what is CFBundleDisplayName should be CFBundleDocumentTypes. -- Julien On Mon, Dec 14, 2009 at 3:29 PM, Joe The Programmer joetheappleprogram...@gmail.com wrote: On Dec 14, 2009, at 4:28 AM, Julien Jalon wrote: It's not necessarily a memory management problem. As this happens very early in the application launch, when Launch Services uses your Info.plist to register the application, your problem might also be that an entry supposed to be a string is in fact an array. Yes, that's why I was thinking resource. As far as I can remember, I didn't touch the plist file going from Xcode 2.5 to 3.2.1. However, I noticed a new key, CFBundleDisplayName, in my 3.2.1 info.plist file. I'm not too familiar with the make up of a plist file. Did the format of the info.plist change between versions? Regardless, there seems to be an erroneous false/ in there. Perhaps something got mangled along the way from 2.5 to 3.2.1? I need to bone up on info.plist. Thanks. keyCFBundleDisplayName/key array dict keyCFBundleTypeExtensions/key array stringtsk/string stringTSK/string /array keyCFBundleTypeIconFile/key stringtsk.icns/string keyCFBundleTypeMIMETypes/key array stringapplication/tsk/string /array keyCFBundleTypeName/key stringTSK Document/string keyCFBundleTypeOSTypes/key array stringtsk /string stringTSK /string /array keyCFBundleTypeRole/key stringViewer/string keyLSTypeIsPackage/key false/ keyNSPersistentStoreTypeKey/key stringJSON/string /dict /array ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: NSURLConnection Problem Inside QuickLook Generator
This discussion would be better done in the quicklook-dev mailing list. And in fact, it already took place there. Please avoid cross posting questions like this. On Tue, Dec 8, 2009 at 7:16 PM, Dalmazio Brisinda dbrisi...@gmail.comwrote: Thanks for that. Okay, I see the QuickLook docs sort of allude to this in regards to using Web-Kit plugins within QuickLook generators. Which gets me thinking: would it then be possible to maybe connect to the server via local distributed objects instead of using the NSURLConnection framework? In otherwords, how deep does the sandbox go? Any ideas on how to possibly bypass this would be most welcome. Best, Dalmazio On 2009-12-08, at 9:48 AM, David Duncan wrote: On Dec 8, 2009, at 2:13 AM, Dalmazio Brisinda wrote: I'm writing a QuickLook plugin that, as part of the preview process, makes an http call to a (localhost) server to obtain data that is then used to generate the preview image. The code below works fine outside the QuickLook architecture, as a standard Cocoa application. But as soon as I move the (exact) same code over to the QuickLook plugin, a call to the server only returns (null) response, and (null) data. The docs indicate that this means the connection is failing. But why o why is it failing? My understanding is Quicklook plugins are not allowed to make network connections due to sandbox restrictions. -- David Duncan Apple DTS Animation and Printing ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: using QLPreviewPanel and supporting 10.5
Replace all calls to [QLPreviewPanel sharedPreviewPanel] and other class calls to weak calls: [NSClassFromString(@QLPreviewPanel) sharedPreviewPanel]. On Sat, Oct 10, 2009 at 3:47 AM, Mitchell Livingston livings...@mac.comwrote: Hello, I want to support Quick Look in my application, but still support 10.5. I'm using the 10.6 SDK and 10.5 deployment target. Launching on 10.5 gives: Dyld Error Message: Symbol not found: _OBJC_CLASS_$_QLPreviewPanel Referenced from: /Users/robertvehse/Desktop/Transmission.app/Contents/MacOS/Transmission xpected in: /System/Library/Frameworks/Quartz.framework/Versions/A/Quartz Obviously this is from using objects only available in 10.6. I won't be using the QL classes on 10.5. Is there a way around this? Thanks, Mitch ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: an app that never quits
Note that when filing this kind of bug, it's always better to explain more precisely your use case (not just let me have background processes!) as it might help design a suited solution within the constraints of the device (same way Push notifications was proposed to solve many use cases) -- Julien On Wed, Sep 23, 2009 at 12:32 AM, Erick Calder e...@arix.com wrote: On Sep 21, 2009, at 11:31 PM, Kyle Sluder wrote: On Mon, Sep 21, 2009 at 11:30 PM, Erick Calder e...@arix.com wrote: wow. ok. I guess that's not going to help much then. I must say that without daemons the range of applications possible seems rather quite narrow to me. Background processing is a much-desired feature. You can add your voice by filing a bug at http://bugreport.apple.com. done. I hope somebody reads this stuff since it's obviously a matter of (poor) policy, rather than of technical difficulty. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: _NSAutoreleaseNoPool missing in Snow Leopard
It's __ (2 underscores) and not _ -- Julien from his iPhone Le 30 août 2009 à 23:00, Seth Willits sli...@araelium.com a écrit : In Snow Leopard I started seeing this: *** __NSAutoreleaseNoPool(): Object 0x100a49ec0 of class NSCFString autoreleased with no pool in place - just leaking I have no idea where this is happening so I tried to break in _NSAutoreleaseNoPool just like the Apple documentation at http://developer.apple.com/mac/library/documentation/Performance/Conceptual/ManagingMemory/Articles/FindingLeaks.html says to. However, it's not resolving. GDB claims there is no _NSAutoreleaseNoPool defined. I used nm on Foundation in Snow Leopard and there is no function. On 10.5, though, there is. So it seems it's gone and I can't find anything similarly named. Does anyone know what I *should* be breaking on? I'd obviously like to track this down. -- Seth Willits ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Reasons why QL would fail to create thumbnail?
Still, trying qlmanage -t /path/to/file on failing machine might help a lot (qlmanage -t -d4 /path/to/file on SnowLeopard).Also, check that the path you give is correct even on a case-sensitive FS. -- Julien On Fri, Aug 28, 2009 at 2:01 AM, Graham Cox graham@bigpond.com wrote: On 28/08/2009, at 5:12 AM, Keith Duncan wrote: What is your preview being generated from, a QuickLook generator or the Preview and Thumbnail files in your bundle's QuickLook folder? The bundle's QuickLook folder - I do not have a QL Generator, I just write an image to /QuickLook/Thumbnail.jpg when the document (package) is saved. I can also load this image directly and I am now doing so if the thumbnail creation fails, though it doesn't look as nice. I haven't found out yet whether this fallback succeeds in the cases where the thumbnail creation fails. --Graham ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Reasons why QL would fail to create thumbnail?
Might be interesting to ask your user to use qlmanage -t on the file to see if anything relevant appears here. Also what kind of file is it? Maybe the user has some plugin installed there. On Thursday, August 27, 2009, Graham Cox wrote: I'm using the following code to generate thumbnail images in one of my user interfaces. It works fine on my machine, and on all the ones I've tested, but a couple of users have reported that they don't see the thumbnails. One user experiencing this problem helped me to log some results which shows that the QLThumbnailImageCreate function is returning nil. The docs state that this will return nil if it can't read the file, but it's the exact same file on every system, loaded from my app's resources. So why would it fail in a few percent of cases, but mostly work? All affected users are running the latest system version. CFURLRef url = CFURLCreateWithFileSystemPath( NULL, (CFStringRef)[self path], kCFURLPOSIXPathStyle, NO ); // logging the URL/path at this point shows that the correct path is being returned from [self path], a path to app internal resources CGSize maxImageSize = CGSizeMake( 512, 512 ); NSDictionary* options = [NSDictionary dictionaryWithObjectsAndKeys:(id)kCFBooleanTrue, kQLThumbnailOptionIconModeKey, nil]; CGImageRef qli = QLThumbnailImageCreate( NULL, url, maxImageSize, (CFDictionaryRef)options ); CFRelease( url ); // on a very few systems, qli is nil here Is there a possibility that QuickLook can somehow be turned off? Any ideas? --Graham ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)http://lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: UUID method Garbage Collection safe?
Correct code is:
+ (NSString *)stringWithUUID{ CFUUIDRef uuidObj = CFUUIDCreate(nil);
NSString *uuidString =
(NSString*)NSMakeCollectable(CFUUIDCreateString(nil,uuidObj));
CFRelease(uuidObj); return [uuidString autorelease];}
Or you will leak in GC.
--
Julien
On Wed, Jul 22, 2009 at 6:45 PM, Squ Aire squ...@live.com wrote:
I've been using this method to make UUID strings:
+ (NSString *)stringWithUUID{ CFUUIDRef uuidObj = CFUUIDCreate(nil);
NSString *uuidString = (NSString*)CFUUIDCreateString(nil,uuidObj);
CFRelease(uuidObj); return [uuidString autorelease];}
How do you all like it? Is it ok? It's based on some code I Googled a while
back. In particular, is this code perfectly garbage collection safe?
_
Show them the way! Add maps and directions to your party invites.
http://www.microsoft.com/windows/windowslive/products/events.aspx___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: Best 'Cocoa' File Preview
Hi David, Glad you got it to work. On Tue, Jul 14, 2009 at 3:34 AM, David Blanton aired...@tularosa.netwrote: Furthermore why so convoluted and difficult? I should be able to say call me for files of type'.pes' and be done with it. That's what you need to do. Except file extensions is not a robust way to designate file types (what about mime types or other way to identify types?). That's why MacOS X uses an intermediate abstraction for type naming (the content type) - You have to declare the content type to the OS (UTTypeIdentifier) and how to sniff files of this type (file extension, mime type, HFS type and whatever might come up in the future under UTTypeTagSpecification). You can also declare other information as well (like its type description, conformance, etc.) - At this point the OS is able to identify your files, now you can associate your file type to applications or plug-ins in CFBundleDocumentTypes: -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Best 'Cocoa' File Preview
On Monday, July 13, 2009, David Blanton aired...@tularosa.net wrote: Is it not the case that the UTI is defined in the info.plist CFBundleDocumentTypes? (which I have done). Nope, CFBundleDocumentTypes associates your type to the plugin. You should use UTExportedTypes or UTImportedTypes look for declaring UTI in info.plist in google, you'll find exactly what you need. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Best 'Cocoa' File Preview
On Mon, Jul 13, 2009 at 12:58 AM, David Blanton aired...@tularosa.netwrote: I am traveling down the QuickLook path ... I have followed all directions but my code is not being called by qlmange as evidenced here: 7/12/09 4:46:08 PM qlmanage[855] [QL] Thumbnailing /users/davidblanton/Aibnb18.pes. Content type UTI: dyn.ah62d4rv4ge81a3px. Generator used: None 7/12/09 4:46:08 PM qlmanage[855] [QL] Thumbnailing for /users/davidblanton/Aibnb18.pes done (No image created). Is there some known secret sauce not divulged in the docs? Your file type (UTI) should be declared somewhere (in your app or in the plug-in). Still, during development, the file type registration might not be done when you want to you can use: qlmanage -t -c com.mycompany.mytype /path/to/file to override type identification. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Memory management in QuickLook plugin
Le 20 mai 2009 à 13:17, Georg Seifert georg.seif...@gmx.de a écrit : Hi, I’m developing a QuickLook plugin for a custom binary file format. As long as I did not cared about memory leaks everything was fine. Then I fixed the leak with the help of Instruments. Everything runs fine from xCode and Instruments but now it crashes if I run the plugin from the finder. Do you mean quicklookd crashes but qlmanage does not? I have set up a custom executable with /usr/bin/qlmanage and the argument -p path/to/my/file. Any advice is very welcome. We lack information here. We'd need to know where it is crashing. -- Julien from his iPhone ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: How do I find this compile error?
I think Bill was meaning QuartzCore framework (for CoreAnimation) and not CoreAudio :-) -- Julien from his iPhone Le 23 avr. 2009 à 20:17, Bill Bumgarner b...@mac.com a écrit : On Apr 23, 2009, at 11:09 AM, James Cicenia wrote: Undefined symbols: _kCATransactionDisableActions, referenced from: _kCATransactionDisableActions$non_lazy_ptr in AudioViewController.o _OBJC_CLASS_$_CALayer, referenced from: __objc_classrefs__d...@0 in AudioViewController.o _OBJC_CLASS_$_CATransaction, referenced from: __objc_classrefs__d...@0 in AudioViewController.o ld: symbol(s) not found collect2: ld returned 1 exit status That isn't a compiler error, that is a linker error! It means that you are using functionality in your application for which you haven't added the appropriate framework. Given the errors, I'm guessing you didn't include CoreAudio.framework in your project? b.bum ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: canBeVisibleOnAllSpaces added in 10.5 and deprecated in 10.5?
I think the how is obvious. I suspect you're more interested in the why :-) -- Julien from his iPhone Le 21 avr. 2009 à 23:34, Sean McBride s...@rogue-research.com a écrit : Hi, Can someone explain how a method added in 10.5 can be deprecated in 10.5? [sic] http://developer.apple.com/documentation/Cocoa/Reference/ ApplicationKit/ Classes/NSWindow_Class/DeprecationAppendix/ AppendixADeprecatedAPI.html Cheers, -- Sean McBride, B. Eng s...@rogue-research.com Rogue Researchwww.rogue-research.com Mac Software Developer Montréal, Québec, Canada ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Getting the message text of a mail
The problem is not really that your plugin might be broken by a new version of mail but that it might break mail without the user understanding it's caused by your plugin. -- Julien from his iPhone Le 21 avr. 2009 à 23:47, geoff...@fileflow.com a écrit : On 21 Apr 2009, at 22:53, Kyle Sluder wrote: On Tue, Apr 21, 2009 at 4:17 PM, Scott Anguish sc...@cocoadoc.com wrote: You should know that there is no public API for writing a Mail.app plugin. And please file a bug regarding this glaring omission. --Kyle Sluder Good point. I know all this but I'm willing to take the chance and update version of the plugins when new Mail.app is coming. But I really would need some help getting the body message as a String. --Geoffrey Rekier ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Difference between SEL and const char* when sending a message?
64bit runtime selectors are not char* as far as I know. In general, you should use NSSelectorFromString/NSStringFromSelector the more low-level objc functions. On Wed, Apr 8, 2009 at 9:21 AM, Sherm Pendley sherm.pend...@gmail.comwrote: On Wed, Apr 8, 2009 at 3:17 AM, Mark Ritchie mark.ritc...@mac.com wrote: You can not simply cast a C string to SEL. is from: http://developer.apple.com/documentation/Cocoa/Reference/ObjCRuntimeRef/ObjCRuntimeRef.pdf Note that while it's be convenient to cast SEL to char *, especially when debugging in gdb, there's no guarantee it will be that way in the future. In fact, in the Cocotron libobjc, and I believe in the GNU libobjc as well, it's not that way in the present. :-) sherm-- -- Cocoa programming in Perl: http://camelbones.sourceforge.net ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Why isn't NSString's caseInsensitiveCompare working correctly?
if( [fn caseInsensitiveCompare:s] ) break;
NSLog(@%@ != %@, fn, s);
compare methods return NSOrderedSame (==0) if the strings are the same so
your test here seems incorrect.
On Sat, Mar 21, 2009 at 7:26 PM, Wesley Spikes wesley.spi...@gmail.comwrote:
Odds are, it's a bug that is based entirely in my code, but it's bugging me
now, and I've hit a road block mentally. Any help provided would be greatly
appreciated.
Please excuse the crypticness of this message, but I'm only able to
disclose
the parts of the code in question.
The basic overview of what's happening in the code is [classInst
findIndexOfFilename:inData:] is being called, with the filename being
passed
in from a directory enumeration. A manual browsing through the data makes
it
look like all aspects of it appear like it should work.
If anything further needs to be known, just let me know, I'll likely be
able
to provide.
At the moment, I'm pretty sure it should work by the looks of it. What's
really bothersome about it to me is that the output of the first error
shows
that the entry at 10:28:17.401 is the only entry that is outputted by
findIndexOfFilename:inData:, and none of the subsequent entries are
presented as having been checked. *scratches head*
First failed output record in the XCode console
[Session started at 2009-03-21 10:28:17 -0700.]
2009-03-21 10:28:17.400 IX2[52319:10b] Contents
2009-03-21 10:28:17.401 IX2[52319:10b] Contents != Contents
2009-03-21 10:28:17.407 IX2[52319:10b] Hmm, it appears that Contents is
not in passed data pointer. pname=cut path/Contents
The code to find the index -- before this is called, the caller is
currently calling NSLog(@%@, fn);
- (int) findIndexOfFilename:(NSString*)fn
inData:(S_KH_DATA*)data
{
int ctr = 0;
NSString *s;
// this would indicate a failure in our build process
assert( data[0].filename != NULL );
do {
s = [NSString stringWithCString:data[ctr].filename
encoding:NSUTF8StringEncoding];
if( [fn length] != [s length] ) continue; // don't go and do a full
compare if diff lens
if( [fn caseInsensitiveCompare:s] ) break;
NSLog(@%@ != %@, fn, s);
} while( data[++ctr].filename != NULL );
if( data[ctr].filename != NULL )
return ctr;
return -1;
}
The structure that is the data[x]
typedef struct _S_KH_DATA
{
uint8_t fs_type; // 1 is directory, 2 is plain
file, 3 is executable file, 0 is unset/ignore
char *filename; // filename is the name of the
filesystem
uint8_t hash[SHA512_DIGEST_LENGTH]; // sha2-512
} S_KH_DATA;
The data being passed (first entries only, abbreviated)
S_KH_DATA kh_data [] = {
{ KHD_TYPE_DIRECTORY, , 0 },
{ KHD_TYPE_DIRECTORY, Contents, 0 },
{ KHD_TYPE_DIRECTORY, Contents/MacOS, 0 },
{ KHD_TYPE_PLAIN_FILE, Contents/Info.plist, { 0x82, ..., 0xe3, } },
...
{ 0, NULL, 0 },
};
--
Wesley Spikes
wesley.spi...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to jja...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: using QuickLook from an application?
-- Julien Le 12 mars 09 à 22:29, Bill Janssen jans...@parc.com a écrit : I'd like to be able to invoke the QuickLook server from my application to obtain PDF or RTF preview versions of MS-Office Documents before opening them. The capability exists; you can see it in Finder, for instance. But I can't see how to invoke it from a client. The only useful call I can make is QLThumbnailImageCreate(), which unfortunately is for single-page data. Symmetry tells me there should be something like QLPreviewDocumentCreate(), but I don't see it anywhere. What am I missing? Surely Apple wouldn't keep such a useful facility out of the hands of developers? As of Leopard, Apple does. Please file a radar and explain precisely what you want. Bill ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: UTIs and type codes and extensions
On Wed, Jan 14, 2009 at 2:14 AM, Randall Meadows cocoa-...@not-pc.comwrote:
I'm just making sure I'm not missing something...
There's no system-defined way to go from a UTI to an old-style file type
code (which the UTI docs seem to refer to as a tag) and back again, am I
correct?
extern CFStringRef
UTTypeCopyPreferredTagWithClass(
CFStringRef inUTI,
CFStringRef inTagClass)
AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
extern CFStringRef
UTTypeCreatePreferredIdentifierForTag(
CFStringRef inTagClass,
CFStringRef inTag,
CFStringRef inConformingToUTI) /* can be NULL */
AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
extern CFArrayRef
UTTypeCreateAllIdentifiersForTag(
CFStringRef inTagClass,
CFStringRef inTag,
CFStringRef inConformingToUTI) /* can be NULL */
AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER;
For instance, my app can export files as TIFF, JPEG, or PNG. When the user
makes the selection, I convert the selection popup menu to the appropriate
UTI using the defined constants in UTCoreTypes.h (kUTTypeTIFF, et al). I'd
like to generalize the subsequent code to take that UTI (public.tiff for
example) and convert it to the appropriate type code ('TIFF'). The next
logical step would be producing a file name extension based on that as well
(public.tiff - tif).
*All* this information is contained in the table at
http://developer.apple.com/documentation/Carbon/Conceptual/understanding_utis/utilist/chapter_4_section_1.html#//apple_ref/doc/uid/TP40001319-CH205-CHDIJFGJ,
but I could discern no way to actually do these types of conversions in a
general manner.
History: The reason I need to do this is, Photoshop is brain-dead when it
comes to opening image files that happen to have no extension in the
filename (and no file type code). Preview, QuickTime Player and
GraphicConverter (and sometimes Safari) have NO problem identifying and
opening the exact same files, sans extension, but PS simply says Could not
complete your request because Photoshop does not recognize this type of
file. I am working around this by attaching type and creator codes to all
the image files I export. (Yes, my client could avoid all this by simply
including an extension to the image name, but old dogs, new tricks, and all
that...)
You always can teach new tricks to old dogs... it's significantly more
difficult for human beings.
--
Julien
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: Security With Show Package Contents?
New Pages format is the same but zipped. On Mon, Jan 12, 2009 at 4:27 PM, Devon Ferns dfe...@devonferns.com wrote: Have you checked if the new Pages file format is now binary instead of a package? That would be my guess. I don't see how you can stop anyone from listing a directory structure. Devon Chunk 1978 wrote: so i was a little put off after purchasing iWork '09, because i could no longer access Show Package Contents of my pages files. i generally used this to swap out images of the same size, or to color balance, etc. anyway, i started thinking about security of applications based on showing package contents. as far as i know the only way for someone to crack an application is to have access to the package contents which lists the Unix Executable File in the Mac OS folder. i guess there's also the possibility to swap out frameworks (particularly Aquatic Prime framework if the framework is installed instead of the Aquatic Prime library)... since apps are really just folders with a .app extension, wouldn't it be possible to disable Show Package Contents, as with the new .pages files, so that it would make the app more secure (if not impossible to crack)? couldn't Apple implement some sort of password protection or optional block on viewing package contents with XCode so that apps are impossible to crack? this post is totally just me thinking out loud. i personally believe that if someone is going to download a cracked version of an app then either they wouldn't have bought a license anyway, or they don't have the money... i'm not trying to make an app of mine more secure. but i'd like to hear your thoughts about this. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/dferns%40devonferns.com This email sent to dfe...@devonferns.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to jja...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: mydoc.myext/QuickLook/Preview.html and full screen
On Fri, Jan 2, 2009 at 5:05 PM, Gerd Knops gerti-cocoa...@bitart.comwrote: On Dec 31, 2008, at 8:43 PM, Julien Jalon wrote: On Wed, Dec 31, 2008 at 10:37 PM, Gerd Knops gerti-cocoa...@bitart.com wrote: Seemingly arbitrary limitation, when according to the documentation Java applets and Flash are supported. Web Plug-ins are not supported (if this is in QL documentation, it is a bug). http://developer.apple.com/documentation/UserExperience/Conceptual/Quicklook_Programming_Guide/QLDynamicGeneration/chapter_7_section_3.html#/ /apple_ref/doc/uid/TP40005020-CH15-SW2 Last Paragraph claims: Although the code listing uses an img HTML element for the cid-scheme reference to the image attachment, you can also use the object element for all kinds of attachments (images, audio, videos, Java applets, and Flash animations). It is not recommended that you use Web Kit plug-ins in enriched HTML passed back to Quick Look. Please file a bug as this is incorrect. Would have been nice to have some client-side javascript to produce a nice functional QuickLook document, instead of being limited to 'no nib allowed' C code. Why no full Objective-C support for Quicklook seems rather odd. Security, stability and other reasons. QL is not meant to fully replace an application. It's not meant to be some sort of Active X or OpenDoc either. Its goal is much more limited (but in the end much more useful ;-) ). For more complex document formats the ability to provide a UI is necessary. For example to explore archives, or in my case a hybrid document that contains a number of other documents (each of which do have PDF and gif previews), and where the document state changes dynamically because it is of a collaborative nature. That's a misunderstanding. Quick Look is not meant to explore files but to provide a quick look of the file. Goals of QL generators is to provide a simple standard representation of the files' contents regardless of the UI (QL is then responsible to provide the best UI possible depending on the context, be it the Quick Look panel, Cover Flow or any other place in the system it might make sense, today or in the future). If you feel the proposed representations for your documents can't really suit your needs, please file a bug describing exactly what your document is all about. In general, QL could be made a lot nicer for Cocoa Programmers than that limited C API. Once you have a CGContext, you have a NSGraphicContext and you can use whatever Cocoa drawing API (you can even use printing APIs or -[NSView displayInContext...]). Note that there is a more suited mailing list to discuss Quick Look ( quicklook-...@lists.apple.com) -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: mydoc.myext/QuickLook/Preview.html and full screen
On Wed, Dec 31, 2008 at 10:37 PM, Gerd Knops gerti-cocoa...@bitart.comwrote: On Dec 31, 2008, at 1:22 PM, Gerd Knops wrote: I am looking to add QuickLook functionality to a fairly complex document. A static html file, using some javascript to interact with the document contents would be ideal for a number of reasons. So I experimented by adding a QuickLook/Preview.html file to my document bundle. That works fine, and in 'normal' QuickLook mode that html can be resized to cover most of the screen. But in full screen mode the html always appears to be sized to cover only about 1/4 of the available space. Is there some trick to allow the html document to use the entire screen? Closer examination shows that apparently javascript is not supported inside QuickLook. Security reasons. In particular, QL is a little more paranoid if it gets the preview directly from the disk (e.g. in MyDoc.docextension/QuickLook/Preview.html) Seemingly arbitrary limitation, when according to the documentation Java applets and Flash are supported. Web Plug-ins are not supported (if this is in QL documentation, it is a bug). Would have been nice to have some client-side javascript to produce a nice functional QuickLook document, instead of being limited to 'no nib allowed' C code. Why no full Objective-C support for Quicklook seems rather odd. Security, stability and other reasons. QL is not meant to fully replace an application. It's not meant to be some sort of Active X or OpenDoc either. Its goal is much more limited (but in the end much more useful ;-) ). -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Asynchronous timers (without a run loop)
The question itself does not make a lot of sense for me. A timer API is tightly bound to the underlying framework that is idling your thread. If the idling API is CFRunLoopRun(), you'd use CFTimers, if your idling API is select() or kevent(), you'd use their timeout parameters. So, for what you seem to say, it's your framework responsibility to provide some asynchronous timer API. If it does not, your only choice is to simulate the timer using other means: - the easy one: one thread that sleeps() for each timer then messages back to your threads - a more complex one is to use a dedicated thread that will effectively run a CFRunLoop and bridge your timers over CFTimers. - you could imagine using signal() +SIGALRM bug using Unix signals here is really not something you should try to use except if you really know what you are doing. -- Julien On Thu, Dec 4, 2008 at 4:14 PM, Påhl Melin [EMAIL PROTECTED] wrote: 2008/12/4 Joseph Kelly [EMAIL PROTECTED]: I've used both the CF and NS timer apis in many different situations without a hitch. They're fairly reliable. Perhaps you could explain your can't use a runloop restriction -- e.g. if you are writing a daemon or a kext, this is not the list you want to post to. I'm not writing a daemon or a kext. I'm porting a framework that I've designed and used on .Net before that will form the foundation of my architecture for my future Mac OS applications (I'm new to Mac OS programming). The framework is used for easy communication and synchronization between threads, processes and computers (with the same syntax) and make it possible to design an application in small modules that communicates via messages and it's straightforward to start with the modules as threads in a single application and then split into several processes on a computer or ever distribute certain modules to other computers without changing much (or any) code at all. But mainly it makes it easier to write efficient multithreaded applications that work. I'm very happy with the messaging framework and would like to use it on Mac OS X as well. Run loops are not compatible with this approach since my threads need to run forever or block in my framework when they are waiting for messages (the timers are used to generate timer messages internally in the framework). The threads are not event driven and will not return to the run loop. You might find if you refactor your design a bit, that it will take to a runloop model quite well. In fact, most of the user space IOKit async routines require that you use a runloop. I don't want to refactor my design. :-) To do that, I would have give up the framework design completely. I'm generally happy with the design just want to make a good port to Mac OS X. Also, the Mach APIs are not strictly private -- the headers are publicly available. They are subject to change between releases, so if you do start using them, you will need to stay on top of things -- check your binaries on pre-release OS seeds etc. Okay, thats great to know. But I suppose it's still begging for trouble, if you can avoid it. Joe K. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Quick Look with native types?
On Fri, Nov 28, 2008 at 5:50 AM, Rob Keniger [EMAIL PROTECTED] wrote: There's no public API to do this at present. You'd need to create a temporary file and use the qlmanage -p /path/to/thefile tool via NSTask. I strongly discourage using qlmanage to display the Quick Look panel. qlmanage -p is meant to debug plug-ins (as stated by the man page) and is really not suited to provide a user interface. In fact, it will give you a very poor user experience: 1) it executes the plug-in directly (and does not use quicklookd), it can't protect itself from bad plug-in behavior. 2) it does not provide all Quick Look panel controls 3) since it's a simple terminal tool it does not integrate properly with a calling application (does not get any key events, can't correctly zoom in/out, etc.) Using qlmanage that way will be even more discouraged in the future. Of course, I also strongly discourage trying to use any private framework or private API to work around that problem as you'll have 99% chance to see your application broken in the next OS release. The best you can hope is to wait for Apple to provide a public API (and you are welcome filing a report to Apple). Of course, you'd need to write a QuickLook generator in order for the temporary file to be previewed correctly. If it were me, I'd just use a custom preview class as you have already done. Exactly, an image view (IKImageView or NSImageView) + a HUD window is not that hard to do. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: LaunchServices/Finder refuses to recognize my NSDocument subclass
If your document type is a package, it does not conform to public.data but to com.apple.package. -- Julien On Sat, Nov 1, 2008 at 5:12 PM, John Pannell [EMAIL PROTECTED]wrote: Hi all- I'm working on a document-based application, but having trouble getting the Finder to match my document to my application. On disk, the document is a package. I am successfully writing the directory and the contents to disk, and the directory is written with the proper extension (.pflow). On the un-successful side, the directory appears as a regular folder in the finder. Here are the relevant entries to my app's Info.plist... do you see anything amiss? ?xml version=1.0 encoding=UTF-8? !DOCTYPE plist PUBLIC -//Apple//DTD PLIST 1.0//EN http://www.apple.com/DTDs/PropertyList-1.0.dtd; plist version=1.0 dict keyCFBundleDevelopmentRegion/key stringEnglish/string keyCFBundleDocumentTypes/key array dict keyCFBundleTypeExtensions/key array stringpflow/string /array keyCFBundleTypeIconFile/key string/string keyCFBundleTypeName/key stringProtoFlow Document/string keyCFBundleTypeRole/key stringEditor/string keyLSItemContentTypes/key array stringcom.positivespinmedia.pflow/string /array keyNSDocumentClass/key stringPFDocument/string /dict /array keyCFBundleExecutable/key stringProtoFlow/string keyCFBundleIconFile/key string/string keyCFBundleIdentifier/key stringcom.positivespinmedia.ProtoFlow/string keyCFBundleInfoDictionaryVersion/key string6.0/string keyCFBundlePackageType/key stringAPPL/string keyCFBundleShortVersionString/key string2.0/string keyCFBundleVersion/key string48.1/string keyNSMainNibFile/key stringMainMenu/string keyNSPrincipalClass/key stringNSApplication/string keyUTExportedTypeDeclarations/key array dict keyUTTypeDescription/key stringProtoFlow Document/string keyUTTypeConformsTo/key array stringpublic.data/string /array keyUTTypeIconFile/key string/string keyUTTypeIdentifier/key stringcom.positivespinmedia.pflow/string keyUTTypeTagSpecification/key dict keypublic.filename-extension/key array stringpflow/string /array /dict /dict /array /dict /plist Any help you can give is much appreciated! John Positive Spin Media http://www.positivespinmedia.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Leaking CGColor objects
On Sun, Oct 26, 2008 at 4:20 PM, Antonio Nunes [EMAIL PROTECTED]wrote: On 26 Oct 2008, at 14:25, DKJ wrote: 1. Is there such a thing as a CGColor class? I don't see it in the documentation; but CGColor is what Instruments lists as the leaked objects. You are creating CGColor objects to which you receive pointers of type CGColorRef. You are at C level here, not Objective-C. 2. Are the CGColor objects being leaked when I remove MyCALayer objects from myArray? Should I set foregroundColor to nil when doing this? You need to make sure in the MyCALayer dealloc method to release the color you created during init using something like CGColorRelease(self.foregroundColor). Bad idea: you don't know for sure self.foregroundColor instance will be the one created in the first place (the one that is leaking). The flaw is at the beginning: self.foregroundColor = CGColorCreateGenericRGB( 1.0, 1.0, 0.9, 1.0 ); should be CGColorRef color = CGColorCreateGenericRGB( 1.0, 1.0, 0.9, 1.0 ); self.foregroundColor = color; CFRelease(color); -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Leaking CGColor objects
self.foregroundColor = nil in your -dealloc is totally useless (and in this case, it's even logging some error log!). Taking care of foregroundColor is you superclass responsibility. You are only responsible of the object YOU create or retain (that's why the temp pattern is indeed the right one) In the documentation, the foregroundColor property is defined like this: @property CGColorRef foregroundColor There's no retain parameter. because CGColorRef is not an ObjC type so it would confuse the compiler, but you can consider it's there. For Charles: Because otherwise there won't be a release to balance the retain in the in setForgroundColor:. See the docs on Objective-C memory management: since you don't do that retain you are clearly not the one responsible to do any retain... as you mentioned: see the docs. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: NSNotificationQueue EXC_BAD_ACCESS problems
Try your code by enabling NSZombies (Add NSZombieEnabled=YES to the
environment before launching your tool/app).
This should show you what object is supposed to get messaged. I suspect some
object observes your notification but is deallocated.
--
Julien
On Tue, Oct 7, 2008 at 4:28 PM, Karan, Cem (Civ, ARL/CISD)
[EMAIL PROTECTED] wrote:
I'm having a great deal of trouble getting my notification queue to work
correctly. It always causes a EXC_BAD_ACCESS
whenever I post using NSPostASAP or NSPostWhenIdle, but it works when I
post using NSPostNow. Here are the relevant lines of code:
-(void) testMethod
{
NSNotification *tempNotification = [NSNotification
notificationWithName:@DummyName object:self userInfo:nil];
[[NSNotificationQueue defaultQueue] enqueueNotification:tempNotification
postingStyle:NSPostNow
coalesceMask:(NSNotificationCoalescingOnName |
NSNotificationCoalescingOnSender) forModes:nil];
}
If I change NSPostNow to NSPostASAP, then it immediately crashes. In the
last crash, the stack looked like this:
#0 0x964e6688 in objc_msgSend
#1 0x931731da in _nsnote_callback
#2 0x90722aba in __CFXNotificationPost
#3 0x90722d93 in _CFXNotificationPostNotification
#4 0x931a071c in -[NSNotificationCenter postNotification:]
#5 0x931d3121 in postQueueNotifications
#6 0x9073f9c2 in __CFRunLoopDoObservers
#7 0x90740d1c in CFRunLoopRunSpecific
#8 0x90741cf8 in CFRunLoopRunInMode
#9 0x94902480 in RunCurrentEventLoopInMode
#10 0x949021d2 in ReceiveNextEventCommon
#11 0x9490210d in BlockUntilNextEventMatchingListInMode
#12 0x9580b3ed in _DPSNextEvent
#13 0x9580aca0 in -[NSApplication
nextEventMatchingMask:untilDate:inMode:dequeue:]
#14 0x95803cdb in -[NSApplication run]
#15 0x957d0f14 in NSApplicationMain
#16 0xa7fe in main at main.mm:13
The precise line I get the EXC_BAD_ACCESS is:
0x964e6688 +0024 mov0x20(%edx),%edi
and this seems to happen at the end of the runloop. I tried removing the
coalesceMask arguments, but no luck. Note that NSPostNow causes all the
code to work completely correctly, its just the other two modes that it
doesn't work. Anyone have any clue what is going on?
This is on a OS X 10.5.5 machine with MacFUSE 1.7 installed. The code is
being compiled using Xcode 3.1.1.
Thanks,
Cem Karan
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to [EMAIL PROTECTED]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
Re: How do I guarantee that an object is dealloced on the main thread?
IAs a side note, you always can remove statements like:[NSObject cancelPreviousPerformRequestWithTarget:self ...]; in -dealloc as they are always useless (delayed performs retain the target so if your object is dealloc'ed, there are no outstanding performs for this object). Also, if some object participates in some run loop sources, it should be somehow retained until the source is removed from the loop (or the object stops participating) - in a thread safe manner. It's especially true in a multi-threaded architecture (you have to avoid the source being triggered while the object is dealloc'ed). -- Julien On Sun, Oct 5, 2008 at 11:41 PM, Brian Stern [EMAIL PROTECTED] wrote: On Oct 5, 2008, at 4:52 PM, Jim Correia wrote: On Oct 5, 2008, at 4:43 PM, Brian Stern wrote: This is the question I really wanted to ask: I have an object X that is alloc/inited on the main thread. This object creates some NSThreads with detachNewThreadSelector:toTarget:withObject, which have object X as their target. At a later point I want to release object X and have it be dealloced on the main thread. How do I guarantee that object X is dealloced on the main thread? The problem is this: the NSThread objects retain object X. I have a stop method in object X that sets a 'cancelled' global that the thread objects inspect and they then exit when it turns true. So I send stop and then release on the main thread. The thread objects exit and eventually release object X but the last release, which causes the dealloc, can be on any thread. The real question is, why do you care what thread the -dealloc method runs on? What non-memory management side effects does your -dealloc method have which much be run on the main thread? My main reason was just cleanliness. Looking more closely at the code however, most of which I didn't write, there are some runloop sources that are removed and calls to NSObject cancelPreviousPerformRequestsWithTarge:selector:object. Most of the code that set those things up will have run on the main thread so should be cleaned up on the main thread. There's some other networking cleanup that I don't know if it's threadsafe or not. It's conceivable that I can rework the cleanup so that most of it happens in the stop method and the dealloc method is mostly empty but if there's a way to guarantee that dealloc runs on the main thread then I'd probably prefer that. I guess from your question you don't have a one sentence answer to my question. -- Brian Stern [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Abusing targetForAction: with non-action selectors
It's not safe... once the target is found (using respondsToSelector:, you're right), the action is performed with:[target performSelector:actionSelector withObject:sender]; which means that anything but an object as a sender will potentially crash. The only safe thing you can image is to have an action selector with no paramater at all. -- Julien On Mon, Sep 15, 2008 at 10:13 PM, Dave Dribin [EMAIL PROTECTED] wrote: Hello, Is it safe to use -[NSApplication targetForAction:] with non-action selectors? For example, selectors that have more than one argument, non-id first argument, or return values? The documentation for targetForAction:to:from: seems to imply it just does a respondsToSelector: on each object in the responder chain, and it seems to behave this way in my tests. I'd just like to make sure it is safe to rely on this behavior, or if we should find some other approach. Thanks, -Dave ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: brain-dead NSThread question ...
sleep() is just blocking the thread, so no event is processed. Use run loops instead. Try to replace sleep() with [[NSRunLoop currentRunLoop] runMode:beforeDate:] -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Linker error: symbol(s) not found
In the header declaring DGBoardRepFromGame, make sure to declare it extern
C.
#ifdef __cplusplus
extern C {
#endif
extern DGBoardRepFromGame(DGGame*, char*);
#ifdef __cplusplus
}
#endif
On Wed, Sep 3, 2008 at 10:40 PM, Joachim [EMAIL PROTECTED] wrote:
Dear list,
I'm getting a linker error, and it's not a missing inclusion of a
framework. I suspect it has to do with mixing .m and .mm files, but I'm
actually pretty clueless.
This is the output from the build:
DGBoardRepFromGame(DGGame*, char*), referenced from:
-[DGTinyEngine setPositionFromGame:error:] in DGTinyEngine.o
ld: symbol(s) not found
collect2: ld returned 1 exit status
In my otherwise Objective-C only project, DGTinyEngine.mm is using a C++
engine, hence the need to make it an Objective-C++ file. It imports a
DGFunctions.h file that amongst other general C functions declares the
void DGBoardRepFromGame (DGGame *game, DGBoardRep buffer) function. The
function is implemented in a .m file.
It is the call in DGTinyEngine.mm to this function that causes the linker
error. If I comment out the call to DGBoardRepFromGame, the project links
fine. DGGame is an Objective-C class that is used several places in
DGTinyEngine.mm without causing any problems.
What causes the linker error, and what can I do to avoid it?
Thanks in advance,
Joachim
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to [EMAIL PROTECTED]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
Re: Quick look preview multipage rich text
On Fri, Jul 11, 2008 at 8:00 PM, Philip Dow [EMAIL PROTECTED] wrote: That's right, I realized after posting the message that I should have boiled it down to the following: + (NSPrintOperation *)PDFOperationWithView:insideRect:toData:printInfo: This operation produces a single page of PDF no matter what print settings I pass in, one very long page. + (NSPrintOperation *)printOperationWithView:printInfo:aPrintInfo This operation when set to write the data to disk correctly produces a paginated document. It's like the NSPrintOperation methods do not allow you to create a multi-page PDF in memory. You have to write it to disk first. I did take a look at the output in both operations, and that's exactly what happens. The fist op gives you one long page of pdf, which is unreadable when scaled to fit inside the quicklook preview. The second op gives you what you expect. I suppose I could use the second operation and then read back in the correctly paginated pdf, but then I'm writing out to the hard disk any time the user wants to quicklook my document. This seems like a terrible waste of resources. Please, don't do that as your plug-in really should avoid doing anything but reading stuff on disk. Maybe someone from the AppKit team will be able to find the source of your problem. -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Quick look preview multipage rich text
If I understand correctly your point, it seems that in the first case, you use the direct to data print operation and in the second case, you use the direct to file print operation then read back the file in memory and send it to Quick Look. Both methods use QLPreviewRequestSetDataRepresentation(). Am I right? What do you mean when you say does not display the content as multipage. What does it display? You should try to dump on disk the data produced in the first case and take a look at it. -- Julien On Wed, Jul 9, 2008 at 4:40 AM, Philip Dow [EMAIL PROTECTED] wrote: Hi all, I am trying to generate multipage pdf data for display in a quick look preview from rich text attributed string data. Attributed strings don't know anything about pages, so it seems to me that I'll have to go through the OS printing architecture to generate the multipage pdf content. Frustrating thing is, this works if I write the data to a file but not if I send it to quicklook. I can generate the pdf data with a custom print info specifying the page attributes, but quicklook does not display the content as multipage. If I use the same settings but a different print operation, writing the data to a temporary file instead, I get a correctly paginated document. Code follows. I understand that PDF content can be created with CGPDFContextCreate and the associated begin and end page calls, but the attributed string doesn't know about pages, so that seems like a dead end to me. If there's a method for doing it that way, I'm all for switching. You might also suggest just sending RTF to Quick Look, but I'd like the text attachments to display. Converting it to html seems like even more work. === GenerateMultiPagePDFPreviewForURL(...) NSAttributedString *attrString = ...; NSTextView *textView = ... (filled with attrString); NSMutableData *pdfData = [NSMutableData data]; NSPrintInfo *printInfo = [NSPrintInfo sharedPrintInfo]; [printInfo setPaperSize:NSMakeSize(612,792)]; [printInfo setHorizontalPagination: NSFitPagination]; [printInfo setVerticalPagination: NSAutoPagination]; [printInfo setVerticallyCentered:NO]; NSPrintOperation *po = [NSPrintOperation PDFOperationWithView:textView insideRect:[textView bounds] toData:pdfData printInfo:printInfo]; [po runOperation]; QLPreviewRequestSetDataRepresentation(preview, (CFDataRef)pdfData, kUTTypePDF, NULL); === Quicklook does not present a multipage preview with that code. But the following code writes a multipage document... === NSAttributedString *attrString = ...; NSTextView *textView = ... (filled with attrString); NSPrintInfo *printInfo = [NSPrintInfo sharedPrintInfo]; NSMutableDictionary *printInfoDict = [NSMutableDictionary dictionaryWithDictionary:[printInfo dictionary]]; [printInfoDict setObject:NSPrintSaveJob forKey:NSPrintJobDisposition]; [printInfoDict setObject:@someLocation.pdf forKey:NSPrintSavePath]; printInfo = [[NSPrintInfo alloc] initWithDictionary: printInfoDict]; [printInfo setHorizontalPagination: NSFitPagination]; [printInfo setVerticalPagination: NSAutoPagination]; [printInfo setVerticallyCentered:NO]; po = [NSPrintOperation printOperationWithView:textView printInfo:printInfo]; [po setShowPanels:NO]; [po runOperation]; === In the first case I'm using [NSPrintOperation PDFOperationWithView:insideRect:toData:printInfo:] and in the second [NSPrintOperation printOperationWithView: printInfo:]. Is the PDFOperationWithView method not capable of producing multipage pdf data despite the custom printInfo? ~Phil ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Building against 10.5 SDK, link error on open
The basic POSIX functions have changed a bit in Leopard (to achieve POSIX conformance). This means that POSIX libraries are different in 10.4 SDK vs. 10.5. You can not mix them. Also, if you change the SDK, it's generally a good idea to clean all your targets. -- Julien On Tue, Jun 10, 2008 at 7:34 AM, Dale Jensen [EMAIL PROTECTED] wrote: I'm trying to get my project to build against the 10.5 SDK, but when I switch it from $(DEVELOPER_SDK_DIR)/MacOSX10.4u.sdk to 10.5, I get a couple of errors: Undefined symbols: _open$UNIX2003, referenced from: _CreateEventForPath in editorBridge.o _close$UNIX2003, referenced from: _SimplePing in SimplePing.o _CreateSocketForCommunicationWithHost in SimplePing.o ld: symbol(s) not found collect2: ld returned 1 exit status SimplePing.o is from the Apple provided example. editorBridge is mine, an implementation of kQueue, which uses the open found in fcntl.h I don't know enough about UNIX flavoured C to figure out what settings I might need to change to solve the link error. Suggestions? Thanks! dale ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: invoking quicklook via code
In fact, this is not entirely true... there is no public way to activate Quick Look. qlmanage is for debugging purpose only. What's public is: 1) QLThumbnailImageCreate() in QuickLook framework 2) ImageKit usage of Quick Look -- Julien On Wed, Jun 4, 2008 at 3:29 PM, Charles Steinman [EMAIL PROTECTED] wrote: --- Memo Akten [EMAIL PROTECTED] wrote: Is it possible to somehow just launch the quicklook window for a quicktime file? The publicly available way to activate Quick Look is to call the qlmanage command line tool. Why they didn't make the QuickLookUI framework public is beyond me (I spent hours looking for a public API before I realized it just wasn't there), but you can do something similar with qlmanage -p. http://developer.apple.com/documentation/Darwin/Reference/ManPages/man1/qlmanage.1.html Cheers, Chuck ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: File's Owner
On Sat, May 24, 2008 at 12:30 AM, Hamish Allan [EMAIL PROTECTED] wrote: On Fri, May 23, 2008 at 11:17 PM, Steve Weller [EMAIL PROTECTED] wrote: The hang up that I see is that this documentation give no clue as to the reason for File's Owner's existence. From http://developer.apple.com/documentation/Cocoa/Conceptual/CocoaFundamentals/CoreAppArchitecture/chapter_7_section_8.html (the first hit on Google for cocoa file's owner): You use the File's Owner object as the conduit for connections between objects in the nib file and objects outside of it. which then point to *Further Reading:* You can find out more about nib files (including how to dynamically load them) by reading *Resource Programming Guidehttp://developer.apple.com/documentation/Cocoa/Conceptual/LoadingResources/index.html#//apple_ref/doc/uid/1051i *which explains the anatomy of a nib file: About the File's Owner One of the more important concepts to understand when building your nib files is what the File's Owner object represents. In Interface Builder, the File's Owner object that appears in the nib document window is a proxy for an object that your application will provide at runtime. When you subsequently load a nib file into your application, your code can specify an actual object to take the place of the File's Owner proxy. When you do, any nib file connections to File's Owner are automatically made to the object you provide (assuming it has the correct outlets and actions). The File's Owner object is one of the primary connections your application has to the contents of a nib file. File's Owner can be any object in your application but is most often set to the controller object that manages the corresponding user interface objects in the nib file. By using File's Owner in this way, you can isolate the details of your user interface to your controller object and let it manage the interactions with other parts of your application. Indeed, there is no clue ;-) -- Julien ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: adding Quicklook preview to my app
If your document's format is a document bundle, just make sure to save your PDF preview inside your document's bundle in a folder names QuickLook. The preview should be named Preview.pdf The thumbnail should be named Thumbnail.png (or jpg, or whatever format suites you) If you can't do that, you will have to write a Quick Look plug-in. -- Julien On Thu, Apr 17, 2008 at 10:07 PM, Victor Bovio [EMAIL PROTECTED] wrote: Hi, I have a document based Cocoa app (CAD-like), to which I like to add the Leopard quicklook plugin feature. I tried first looking at the Documentation, but looks overwhelming to me, is there a simple tutorial or examples somewhere I can look at ?? Basically I just want to embed a snapshot view (either PDF or Bitmap) to the saved documents and use that for the quicklook... I appreciate any hints on where to start with something as simple as possible. Thanks! ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Unable to access an instance properties and methods
I'd bet the firstNote object is not a DBNNote instance but an NSString instance. On Wed, Mar 19, 2008 at 5:33 PM, Davide Benini [EMAIL PROTECTED] wrote: Hello folks. I receive an exception message and my application exits whenever I try to access properties or methods of an instance of DBNNote, a custo made class. Here is the code: DBNNote *firstNote= [[tune body] objectAtIndex:0]; NSLog(@%@, firstNote); NSLog(@%@, [firstNote accident]); And this is the exception: 2008-03-19 17:19:30.957 cocoabc[10832:10b] Note: A flat Octave: 6 Duration: 3.00 2008-03-19 17:19:30.957 cocoabc[10832:10b] *** -[NSCFString accident]: unrecognized selector sent to instance 0x33b040 2008-03-19 17:19:30.958 cocoabc[10832:10b] An uncaught exception was raised 2008-03-19 17:19:30.958 cocoabc[10832:10b] *** -[NSCFString accident]: unrecognized selector sent to instance 0x33b040 2008-03-19 17:19:30.959 cocoabc[10832:10b] *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '*** - [NSCFString accident]: unrecognized selector sent to instance 0x33b040' 2008-03-19 17:19:30.959 cocoabc[10832:10b] Stack: ( 2505978443, 2451210491, 2506007626, 2506000972, 2506001170, 13259, 8495, 2506037845, 2490420930, 2490380524, 2490378829, 2490377872, 2490377678, 2490376830, 8372, 8298 ) The Debugger has exited due to signal 5 (SIGTRAP).The Debugger has exited due to signal 5 (SIGTRAP). As you see the first line of the exception message correctly logs [firstNote description]. The second NSLog should print flat. So it appears the object is correctly instantiated, yet I cannot access its methods. I have tried to call both methods and properties accessors, but no call works so far. I have a feeling I am making a very basic mistake, but I can't tell which one. Any suggestion? Davide ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: warning: assignment from distinct Objective-C type
1) All init methods should return (id) not a specific class2) your initWithDelegate: is likely too generic as a name and its signature conflicts with an other one. Since [XMPPStream alloc] is typed id, the compiler might not be sure of what method signature you want to use for -initWithDelegate: In the second case, the compiler knows you are using -[XMPPStream initWithDelegate:] I tend to remember that the compiler warning was more specific before Leopard, maybe a regression. Nonetheless, good coding style should lead you to rename your method to: - (id)initXMPPStreamWithDelegate:(id)aDelegate; instead of: - (XMPPStream *)initWithDelegate:(id)aDelegate; -- Julien On Tue, Mar 11, 2008 at 9:19 AM, Stuart Malin [EMAIL PROTECTED] wrote: I have a line of code: xmppStream = [[XMPPStream alloc] initWithDelegate:self]; That when compiled, receives a warning: warning: assignment from distinct Objective-C type Now, what's odd to me, is if I change the source code to this: xmppStream = [XMPPStream alloc]; [xmppStream initWithDelegate:self]; It compiles without any warning. The interface for the XMPPStream initializer is: - (XMPPStream*) initWithDelegate:(id)initialDelegate; And, in the file that is allocating and initializing, the xmppStream variable is defined as: XMPPStream *xmppStream; So I don't see any reason that I should get the warning. Any clues about why this happens would be appreciated. I'm sure I must be overlooking something obvious... This is happening with Xcode 2.4.1 [I have posted to both Cocoa and Xcode because I have no idea which list would be better for this] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: warning: assignment from distinct Objective-C type
It depends on what you mean by bad... it is clearly bad Cocoa coding style
(*). It also lead to a lot of typing problems when subclassing you classes.
If the compiler can't decide which methods your code is referring to, it can
end up use the wrong signature producing incorrect code.
For example, suppose you have the two following methods:
- (id)initWithValue:(float)value;
- (id)initWithValue:(int)value;
If the compiler uses the wrong method, the underlying method implementation
will be called with a float parameter while expecting a int or vice-versa.
--
Julien
Note that this coding style exists mainly to work-around the ObjC
declaration limitations.
On Tue, Mar 11, 2008 at 7:03 PM, Stuart Malin [EMAIL PROTECTED] wrote:
Brilliant Julien! Thank you!
I had always been curious why initializers returned id, since their return
could be made specific to the class.
I've been doing what I've done in many classes, and some give me grief,
and not others.
Both of your assessments are correct.
To validate #1, if I change the offending line of code to do a cast:
xmppStream = [(XMPPStream*)[XMPPStream alloc] initWithDelegate:self];
the warning goes away. Of course, that's ugly and I won't retain that code
change because the deeper problem is exactly what you refer to in item #2 --
I do have more than one class with the same initializer signature. I'd
thought the compiler would choose the right one because I'd
naively presumed that [className alloc] returned an instance in that
class, not an object of type id.
This is probably why I had to cast in my initializer:
- (XMPPStream*) initWithDelegate:(id)initialDelegate
{
self = (XMPPStream*)[super init];
If I change my initializer method name as you suggest so that it has a
unique signature,
- (XMPPStream*) initXMPPStreamWithDelegate:(id)initialDelegate;
and the creation of an instance accordingly:
xmppStream = [[XMPPStream alloc] initXMPPStreamWithDelegate:self];
all now compiles cleanly.
Now, the question is: is it bad to continue doing what I've been doing in
having the initializer interface defined to not return id but an instance of
the actual class? Or would my continuing to do this merely be a
non-conventional coding style?
--Stuart
On Mar 10, 2008, at 11:16 PM, Julien Jalon wrote:
1) All init methods should return (id) not a specific class2) your
initWithDelegate: is likely too generic as a name and its signature
conflicts with an other one.
Since [XMPPStream alloc] is typed id, the compiler might not be sure of
what method signature you want to use for -initWithDelegate:
In the second case, the compiler knows you are using -[XMPPStream
initWithDelegate:]
I tend to remember that the compiler warning was more specific before
Leopard, maybe a regression.
Nonetheless, good coding style should lead you to rename your method to:
- (id)initXMPPStreamWithDelegate:(id)aDelegate;
instead of:
- (XMPPStream *)initWithDelegate:(id)aDelegate;
--
Julien
On Tue, Mar 11, 2008 at 9:19 AM, Stuart Malin [EMAIL PROTECTED]
wrote:
I have a line of code:
xmppStream = [[XMPPStream alloc] initWithDelegate:self];
That when compiled, receives a warning:
warning: assignment from distinct Objective-C type
Now, what's odd to me, is if I change the source code to this:
xmppStream = [XMPPStream alloc];
[xmppStream initWithDelegate:self];
It compiles without any warning.
The interface for the XMPPStream initializer is:
- (XMPPStream*) initWithDelegate:(id)initialDelegate;
And, in the file that is allocating and initializing, the xmppStream
variable is defined as:
XMPPStream *xmppStream;
So I don't see any reason that I should get the warning.
Any clues about why this happens would be appreciated.
I'm sure I must be overlooking something obvious...
This is happening with Xcode 2.4.1
[I have posted to both Cocoa and Xcode because I have no idea which
list would be better for this]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to [EMAIL PROTECTED]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
Re: [iPhone] xml parsing
I'm pretty sure you read the SDK release notes ;-) If not, I think you should, that's always a good idea. -- Julien On Mon, Mar 10, 2008 at 11:49 PM, Simon Fell [EMAIL PROTECTED] wrote: The iPhone docs point you in the direction of libXML2 for parsing XML, yet the headers for NSXML are included in the SDK headers, and i was able to build and run fine using NSXMLDoc/element/node. (in fact i just dropped my existing NSXML based code into a iPhone project and it worked fine, so it never even occurred to me that this might not be supported). Would this constitute using a private API ? (is there a good definition of private API somewhere?) Tx Simon ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: [iPhone] xml parsing
I can't comment on the iPhone part as this is NDA (and should not be discussed here) but, on Mac OS X, I'd add /usr/include/libxml2 to the header search path in project (or target) build settings. Also, I'd add the libxml2.dylib to my project using Add existing framework or add -lxml2 to the additional linker flags of project (or target) build settings. Of course, this only describes what I'd do for a Mac OS X project in Xcode. -- Julien On Tue, Mar 11, 2008 at 1:33 AM, Jacob Bandes-Storch [EMAIL PROTECTED] wrote: On a similar-ish note, I'm having trouble using libxml. I'm using a slightly modified version of the appropriate sample code. I need to make sure that the library is in my project. Since I couldn't find the library in a convenient place (where are you supposed to get those anyway?), I just dragged it from the sample project to mine. However, when I try to build the project, it says it cannot locate libxml or libxml/tree.h or whatever I import. I can't find any significant differences in the projects' properties... does anyone know how to get this working? On Mar 10, 2008, at 4:46 PM, [EMAIL PROTECTED] wrote: If you find a function/method/class declaration in a public header (an header that is not in PrivateHeader folder) you can problably considere it as public. Le 10 mars 08 à 23:49, Simon Fell a écrit : The iPhone docs point you in the direction of libXML2 for parsing XML, yet the headers for NSXML are included in the SDK headers, and i was able to build and run fine using NSXMLDoc/element/node. (in fact i just dropped my existing NSXML based code into a iPhone project and it worked fine, so it never even occurred to me that this might not be supported). Would this constitute using a private API ? (is there a good definition of private API somewhere?) Tx Simon ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: localizable strings file with macro
1) Might not be a good idea to use CFStringGetCStringPtr as it might return
NULL2) kCFStringEncodingMacRoman is likely not a good choice as the
encoding, especially for localized strings
If I were you, I'd do something like (warning: Mail compiled code):
const char* MyGetLocalizedCString(CFStringRef string)
{
// if you need to be thread safe, add proper locks around that
static CFMutableDictionaryRef mapping = NULL;
if(mapping == NULL) {
mapping = CFDictionaryCreateMutable(NULL, 0,
kCFTypeDictionaryKeyCallBacks, NULL);
}
const char* result = (const char *)CFDictionaryGetValue(mapping,
string);
if(result == NULL) {
CFStringRef localizedString = CFCopyLocalizedStringFromTable(string,
CFSTR(LibraryLocalizable), NULL);
result = (const char
*)malloc(CFStringGetMaximumSizeForEncoding(CFStringGetLength(localizedString),
kCFStringEncodingUTF8);
CFStringGetCString(localizedString, (char *)result, bufferSize,
kCFStringEncodingUTF8);
CFDictionarySetValue(mapping, string, result);
}
return result;
}
And use that as you localizing function:
#define _(a) MyGetLocalizedCString(CFSTR(a))
Even better, if you can change that, add an extra parameter to your
localizing macro and give it a better name:
#define _loc(a, comment) MyGetLocalizedCString(CFSTR(a))
This keeps your code portable and you should be able to replace the
occurrences of _ in all your code fairly easily with a search and replace.
Added value is:
A) it makes your code more readable (IMHO)
B) it adds a proper way to explain the context of the localization
C) you can generate your strings file directly using genstrings:
genstrings -s _loc *.c -o /tmp ; mv /tmp/Localized.strings
path/LibraryLocalizable.strings
If you really can't do that, preprocess your files before launching
genstrings and call use -s MyGetLocalizedCString
--
Julien
On Tue, Mar 4, 2008 at 3:53 AM, Mitchell Livingston [EMAIL PROTECTED]
wrote:
Hello,
I have C code that I want to generate a strings file for use on Mac. I
use:
#if defined(SYS_DARWIN)
#include CoreFoundation/CFBundle.h
#define _(a) CFStringGetCStringPtr(CFCopyLocalizedStringFromTable( \
CFSTR( a ), CFSTR(LibraryLocalizable), comment ),
kCFStringEncodingMacRoman)
#else
...
Unfortunately genstrings does not seem to compile the code before
attempting to create the strings file. Is there some way to generate a
string file while still considering that the code will also be run on
other operating systems that generate localizable strings in different
ways?
Thanks,
Mitchell Livingston
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/jjalon%40gmail.com
This email sent to [EMAIL PROTECTED]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
