Re: adding something to a setter

[Jim Correia]

Do not use self as your observation context.

Your observation context has to be unique for all observers of the object.  
“self” is a bad choice.

Also, you should only do your own work in the case that the observer is your 
observer. In the code below, you are also doing your work in the case where you 
should only be calling through to super.

—Jim

On Oct 6, 2011, at 5:47 AM, Stephan Michels wrote:

> I use KVO to execute custom code if a property has changed.
> 
> [self addObserver:self forKeyPath:@"graphics" 
> options:(NSKeyValueObservingOptionNew|NSKeyValueObservingOptionOld) 
> context:self];
> 
> - (void)observeValueForKeyPath:(NSString *)keyPath ofObject:(id)object 
> change:(NSDictionary *)change context:(void *)context {
>if (context != self) {
>[super observeValueForKeyPath:keyPath ofObject:object change:change 
> context:context];
>}
> 
>if ([keyPath isEqualToString:@"graphics"]) {
>[self setNeedsDisplay:YES];
>}
> }
> 
> AFAIK the observe method will not be executed instantaneously, but on the 
> other side setNeedsDisplay has also a
> deferred execution.
> 
> Stephan.
___

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

-[NSURLCredentialStorage removeCredential:forProtectionSpace:] broken?

[Jim Correia]

Is there a known problem with -[NSURLCredentialStorage 
removeCredential:forProtectionSpace:], or are my expectations wrong?
 
I iterate all the credials in the storage looking for the one I wish to remove, 
then send -removeCredential:forProtectionSpace: to the instance, and it doesn't 
actually remove the item from the keychain.
 
Is this supposed to work?
 
(Yes, I know I can, and probably will have to, drop down to the keychain APIs 
directly.)
 
—Jim


___

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: Core Data : Undo Delete : Cannot fulfill a fault

[Jim Correia]

On Sep 30, 2011, at 5:41 AM, Jerry Krinock wrote:

> I'm still working on this issue, trying to find the trigger which moves my 
> corner case into the corner, hoping for a workaround.  It's tedious because 
> of different operations and states.  

But you have a pretty reliable way to reproduce the bug, right?

Have you taken my previous advice?

Have you broken at the point of objc_exception_throw for the 
NSObjectInaccessibleException and examined what, specifically, is firing the 
fault? 

Start here and work backwards and you may be able to infer why someone has a 
reference to a stale object. At the very least you’ll know who. (You’ve already 
got the type and ID of the stale object.)

> The implication of the last two points is that Core Data implements Undo 
> quite on its own; the undo manager acts as little more than a counter.

I'm not sure how you jumped to this conclusion. Core Data is recording 
invocations in NSUndoManager in much the same way you'd do if you were using 
NSUndoManager directly.

—Jim

___

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: Core Data : Undo Delete : Cannot fulfill a fault

[Jim Correia]

On Sep 30, 2011, at 12:06 AM, Quincey Morris wrote:

> The actual failure scenario I described has 2 necessary conditions. One is a 
> save boundary. The other is the flushing of cached property information for 
> the deleted object. The latter is difficult to cause, especially in a trivial 
> example, since there's no API for affecting the cache directly, except for 
> resetting the MOC, which would probably break undo for different reasons.
> 
> I guess the question here is this:
> 
> If it's true that deleting a managed object causes it to be turned into a 
> fault at some point (at the next save boundary at the latest), and saving a 
> managed context with a deleted object actually causes the object's persistent 
> representation to be deleted from the persistent store, and post-save the 
> object is still referenced by an undo action, what possible mechanism can 
> re-populate the object's properties if the undo action is undone? By 
> definition, the necessary information no longer exists.

You are assuming that the only place this information can exist is in the row 
cache itself. The information can exist anywhere. At the risk of 
over-trivializing the problem—It’s just code™.

All the information necessary to resurrect the object (or restore properties) 
can be stored on the undo stack itself.

—Jim___

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: Core Data : Undo Delete : Cannot fulfill a fault

[Jim Correia]

On Sep 29, 2011, at 11:20 AM, Quincey Morris wrote:

> I think the answer to your original question is that You're Doing It Wrong™. 
> :)
> 
> I'm pretty sure (though I never really thought about it before today) that 
> Core Data undo *doesn't* work across 'save:' boundaries. The documentation 
> for [NSManagedObjectContext undo:] says this:
> 
>> "Sends an undo message to the receiver’s undo manager, asking it to reverse 
>> the latest uncommitted changes applied to objects in the object graph."
> 
> 
> Note that word "uncommitted". After 'save:', all changes are committed, so 
> you're not permitted to undo past that point. At least, that's how I read it.

That documentation is unfortunately confusing. Someone should file a bug to 
have it clarified.

Core Data’s undo stack does allow you to undo past the last save point. And if 
it didn’t, it would remove actions from the undo manager, resulting in a 
disabled undo menu, not leave a bunch of actions on the undo stack that would 
generate exceptions if the user had the nerve to actually keep undoing past the 
last save point.

> If I'm right, you should be clearing the undo stack at a save, at least if 
> there are deleted objects in the picture.

If you in fact had to do this, this would be a pretty grievous bug. If the MOC 
registers actions with its undo manager that it can no longer deal with, it 
should be the one who removes them or clears the stack.

A trivial NSPersistentDocument sample application handles undo correctly across 
save boundaries for both property changes and deletes. 


Jerry Krinock wrote:

> It's always the same "exceptional" object which whose deletion is unable to 
> be undone because Core Data could not fulfill a fault.  Its parent is an 
> object that was not deleted and therefore does not need to be restored.  It 
> has one child which has already been successfully restored at the time the 
> exception occurs.

Have you broken at the point of objc_exception_throw for the 
NSObjectInaccessibleException and examined what, specifically, is firing the 
fault? 

Start here and work backwards and you may be able to infer why someone has a 
reference to a stale object. At the very least you’ll know who. (You’ve already 
got the type and ID of the stale object.)

If you can isolate this into a standalone sample, (and in the process haven’t 
found a bug in your code causing it) you can file a radar and post a link to 
the sample here and maybe we can offer further advice. 

—Jim___

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: confused by what seems to be a recursive declaration/definition

[Jim Correia]

On Aug 11, 2011, at 7:27 PM, William Squires wrote:

> On Aug 11, 2011, at 6:51 PM, Kyle Sluder wrote:
> 
>> On Thu, Aug 11, 2011 at 4:26 PM, Jens Alfke  wrote:
>>> 
>>> On Aug 11, 2011, at 4:10 PM, Luther Baker wrote:
>>> 
 static void *AVPlayerDemoPlaybackViewControllerRateObservationContext = &
 AVPlayerDemoPlaybackViewControllerRateObservationContext;
>>> 
>>> It’s initializing the variable to point to itself. Which seems sort of 
>>> pointless, but I don’t know the context.
>> 
>> It's a quick way to initialize a unique pointer when you don't care
>> about the value.
> 
> Of course, if you don't care about the value, why would you need a unique 
> one? :)

For KVO contexts, you don't care about the value, but you do care that it is 
unique.

—Jim

___

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 Multi-Gesture events with Carbon

[Jim Correia]

On Jun 29, 2011, at 2:24 PM, Abdul Sowayan wrote:

> I have a carbon based application that we are converting overtime to Cocoa. 
> What I would like to do, if possible, is to incorporate multi-gesture events 
> (such as magnify event) into my carbon application. I can see that 
> NSResponder has magnifyWithEvent: selector. I am wondering if there is anyway 
> to do so. Any feedback on how to do so would be greatly appreciated.

Magnify, swipe, and rotate are available as Carbon Events.

—Jim

___

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: TransformProcessType() still doesn't show menu (Bug ID# 5905139)

[Jim Correia]

On May 26, 2011, at 7:21 PM, Jerry Krinock wrote:

> I just thought I'd lift my head up long to say thank you, Anders.  
> Unfortunately, your code doesn't work in my app.  Tried several mutations; 
> still no good.
> 
> Another thing I learned is that programmatically switching apps, simulating 
> the user typing cmd-tab, which works when the user does it, does not work 
> when the program does it.  I hard-coded pids of several running programs, 
> activated each for 1 second, then activated my app.  Looked very impressive 
> until it got to my app, and the menu bar continued to show the 
> previously-activated app.

This is an unfortunate bug, but you should be able to work around it with this 
sequence:

[[NSApplication sharedApplication] hide: nil];

ProcessSerialNumber psn = {0, kCurrentProcess};
verify_noerr(TransformProcessType(&psn, 
kProcessTransformToForegroundApplication));

[[NSApplication sharedApplication] 
activateIgnoringOtherApps:YES];

—Jim

___

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: -[NSSet containsObject:] returns NO when it should return YES

[Jim Correia]

On Feb 19, 2011, at 2:17 PM, Michael Crawford wrote:

> I'm trying to compare instances of NSNumber using NSSet.  I have a set of 
> numbers, which represent 64-bit persistent IDs for iTunes media-items.  I 
> access these values using the -[NSNumber longLongValue] method.  I'm adding 
> them to an instance of NSSet and then use that instance to determine whether 
> or not I've seen the given item before.


> Anyone have previous experience with this problem?  Can you shed some light 
> on where my thinking and assumptions are wrong.

The documentation for -[NSSet containsObject:] says:

Return Value — YES if anObject is present in the set, otherwise NO.

The documentation for -[NSSet member:] says:

Return Value — If the set contains an object equal to object (as determined by 
isEqual:) then that object (typically this will be object), otherwise nil.

Jim


___

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: Return key should enter return in textfield and not press default button

[Jim Correia]

On Oct 11, 2010, at 2:21 AM, Thorsten Lemke wrote:

> I have a dialog in my app wich should let the user enter a carriage return.
> 
> How can I avoid pressing the default button with return? Only enter should
> press it in this case.

As you probably know, implementing this will be hugely popular with people who 
understand how it works, and confusing to everyone else.

That said…

For your edit field implement

- (BOOL)control:(NSControl *)control textView:(NSTextView *)textView 
doCommandBySelector:(SEL)commandSelector;

When the command selector is insertNewline:, act accordingly based on the 
current event.

—Jim


___

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: Sending a Cmd-V keystroke

[Jim Correia]

On Mar 25, 2010, at 8:23 PM, Jim Graham wrote:

> This is a problem that occurred a few months back in a project. I Never 
> actually used it in the project but it has bugged me ever since. I have 
> simplified the problem into a small example
> 
> There is a window with three NSTextFields and a button. The button is 
> connected to the doPaste action and the fields are outlets

Before explaining why your code, as written doesn’t work, and how to fix it, 
can you give some background as to why you are doing it this way?

Unless you are trying to write a tool similar to Keyboard Maestro, you probably 
don’t want to be doing it this way.

- In general, you probably want to modify the model objects programmatically, 
and have the views react and update their display values.

- In the case that you want to update the views directly, you really ought to 
be reading the value from the pasteboard and setting the view’s display value 
programatically.

If this really is an academic exercise in pulling puppet strings on the UI, and 
you have legitimate reasons for doing so, including them in the original 
question would have avoided the speculation about why you are doing something 
which seems superficially crazy.

>   [pb setString:@"Field 1" forType:NSStringPboardType];
>   [textField1 becomeFirstResponder];

Problem #1: 

You change the first responder to sending -makeFirstResponder: to the window. 
It may fail if editing could not be committed for he current editor.

> -(void)pasteIt

Problem #2a:

This code assumes that your window is the key window and the application is 
frontmost.

Problem #2b:

You have posted an event, but until AppKit dequeues the event and processes it, 
the paste doesn’t actually happen. This is why you eventually get 3 pastes in 
the last edit field.

- Jim___

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: Sending a Cmd-V keystroke

[Jim Correia]

On Mar 25, 2010, at 8:23 PM, Jim Graham wrote:

> This is a problem that occurred a few months back in a project. I Never 
> actually used it in the project but it has bugged me ever since. I have 
> simplified the problem into a small example
> 
> There is a window with three NSTextFields and a button. The button is 
> connected to the doPaste action and the fields are outlets

Before explaining why your code, as written doesn’t work, and how to fix it, 
can you give some background as to why you are doing it this way?

Unless you are trying to write a tool similar to Keyboard Maestro, you probably 
don’t want to be doing it this way.

- In general, you probably want to modify the model objects programmatically, 
and have the views react and update their display values.

- In the case that you want to update the views directly, you really ought to 
be reading the value from the pasteboard and setting the view’s display value 
programatically.

If this really is an academic exercise in pulling puppet strings on the UI, and 
you have legitimate reasons for doing so, including them in the original 
question would have avoided the speculation about why you are doing something 
which seems superficially crazy.

Problem #1: 

> 
> Here is the code
> 
> 
> -(IBAction)doPaste:(id)sender{
>   
>   NSPasteboard *pb = [NSPasteboard generalPasteboard];
>   [pb declareTypes:[NSArray arrayWithObject:NSStringPboardType] 
> owner:self];
>   
>   [pb setString:@"Field 1" forType:NSStringPboardType];
>   [textField1 becomeFirstResponder];
>   [self pasteIt];
>   
>   [pb setString:@"Field 2" forType:NSStringPboardType];
>   [textField2 becomeFirstResponder];
>   [self pasteIt]; 
>
>   [pb setString:@"Field 3" forType:NSStringPboardType];
>   [textField3 becomeFirstResponder];
>   [self pasteIt];
> 
> }
> 
> -(void)pasteIt{
> 
>   CGEventSourceRef sourceRef = 
> CGEventSourceCreate(kCGEventSourceStateCombinedSessionState);
> if (!sourceRef)
> { 
> NSLog(@"No event source");
> return;
> }
>   //9 = "v"
>   CGEventRef eventDown = CGEventCreateKeyboardEvent(sourceRef, 
> (CGKeyCode)9, true);
>   CGEventSetFlags(eventDown, kCGEventFlagMaskCommand);
>   CGEventRef eventUp = CGEventCreateKeyboardEvent(sourceRef, 
> (CGKeyCode)9, false);
>   CGEventPost(kCGHIDEventTap, eventDown);
>   CGEventPost(kCGHIDEventTap, eventUp);
>   CFRelease(eventDown);
>   CFRelease(eventUp);
>   CFRelease(sourceRef);
> 
> }
> 
> I expected that each of the fields would have their respective text pasted 
> into them. What actually happens is that the last field gets three copies of 
> it's text, Field 3, pasted into it. The other two fields are blank.
> 
> Can anybody explain why that should happen.
> 
> Jim 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/jim.correia%40pobox.com
> 
> This email sent to jim.corr...@pobox.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: Core Data: fetching a subset of attributes/columns

[Jim Correia]

On Mar 20, 2010, at 1:52 PM, Gabriel Fernandez wrote:

> I wanted to get a subset of an entity's attributes because one of the 
> ManagedObject's attributes is a large image. 

[…]

> 2) using the -setPropertiesToFetch: method of NSFetchRequest, with the "id" 
> NSAttributeDescription I'm looking for as the only object in the properties 
> array. But when I use the -(NSArray*) properties method of the fetched 
> objects, the "image" attribute is there with a non-nil value.

Read the API documentation, or the header comment for this method. 

/* Specifies a collection NSPropertyDescriptions that should be returned by the 
result. This value is only used if resultType is set to NSDictionaryResultType. 
Property descriptions added to valuesToFetch may represent attributes, 
expressions, or to one relationships; for attribute or relationship 
descriptions, the name must match the name of a description on the fetch 
entity. Results of the fetch will be returned as dictionaries containing 
key/value pairs where the key is the name of the specified property 
description. */

— Jim___

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: NSComboBox problem in 10.6.2

[Jim Correia]

On Mar 19, 2010, at 3:21 PM, H. Miersch wrote:

> just had another surprise: i inserted NSLogs into the method that initializes 
> the array and into -numberOfItemsInComboBox, trying to see the contents of my 
> array and find out why -numberOfItemsInComboBox returns zero. the surprise 
> was that -numberOfItemsInComboBox is called BEFORE the method that 
> initializes the array. then i commented the line that sends 
> -selectItemAtIndex to the combobox, and the combobox worked normally. 
> 
> now the question is, why is -numberOfItemsInComboBox called first when the 
> first call in awakeFromNib is to the initialisation method? weird...
> 
> -awakeFromNib {
>   init array;
>   [combobox selectItemAtIndex:0];   // if this line is turned into a 
> comment, it works normally.
>   [combobox2 selectItemAtIndex:2];
>   [combobox3 selectItemAtIndex:0];
>   register for notifications from combobox;
> }

You are setting up your array in your -awakeFromNib.

You cannot control (or know) the order in which -awakeFromNib will be sent to 
objects in your nib. The combo box may be awoken long before you.

But more to the point, your contract with the combo box as its data source is 
that you will send it -noteNumberOfItemsChanged or -reloadData whenever you 
change the data in the data source.

You became the data source for the combo box during nib unarchiving, changed 
the number of items in the model after nib unarchiving, and didn’t inform the 
combo box about that change. Sending the combo box the appropriate method after 
initializing your data source should fix the problem.

Jim

___

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: NSComboBox problem in 10.6.2

[Jim Correia]

On Mar 18, 2010, at 3:38 PM, H. Miersch wrote:

> when i try to launch the app from within xcode, the console shows the 
> following:
> 
> 2010-03-18 19:26:28.515 StoX[63999:a0f] An uncaught exception was raised
> 2010-03-18 19:26:28.534 StoX[63999:a0f] *** -[NSCFArray objectAtIndex:]: 
> index (-1 (or possibly larger)) beyond bounds (5)
> 2010-03-18 19:26:28.596 StoX[63999:a0f] *** Terminating app due to uncaught 
> exception 'NSRangeException', reason: '*** -[NSCFArray objectAtIndex:]: index 
> (-1 (or possibly larger)) beyond bounds (5)'
> *** Call stack at first throw:
> (
>   0   CoreFoundation  0x7fff87ddb444 
> __exceptionPreprocess + 180
>   1   libobjc.A.dylib 0x7fff800470f3 
> objc_exception_throw + 45
>   2   CoreFoundation  0x7fff87ddb267 
> +[NSException raise:format:arguments:] + 103
>   3   CoreFoundation  0x7fff87ddb1f4 
> +[NSException raise:format:] + 148
>   4   Foundation  0x7fff8798e080 
> _NSArrayRaiseBoundException + 122
>   5   Foundation  0x7fff878f0b81 -[NSCFArray 
> objectAtIndex:] + 75
>   6   AppKit  0x7fff80ba2012 
> -[NSComboBoxCell selectItemAtIndex:] + 191
>   7   StoX0x00011693 
> -[AppController awakeFromNib] + 50

This is the interesting frame

>   8   CoreFoundation  0x7fff87d8a82d -[NSSet 
> makeObjectsPerformSelector:] + 205
>   9   AppKit  0x7fff8086c913 
> -[NSIBObjectData nibInstantiateWithOwner:topLevelObjects:] + 1445
>   10  AppKit  0x7fff8086ab49 loadNib + 226
>   11  AppKit  0x7fff8086a059 
> +[NSBundle(NSNibLoading) _loadNibFile:nameTable:withZone:ownerBundle:] + 248
>   12  AppKit  0x7fff80869e91 
> +[NSBundle(NSNibLoading) loadNibNamed:owner:] + 326
>   13  AppKit  0x7fff80867413 
> NSApplicationMain + 279
>   14  StoX0x000115a8 start + 52
> )
> terminate called after throwing an instance of 'NSException'
> 
> i've checked that i'm using the types that the docs say i should be using, 
> but i don't understand why index comes back as -1. 
> 
> am i doing something wrong? should I be doing something i'm not doing? is the 
> documentation wrong? is there a bug somewhere?

Look at your code in awakeFromNib. It appears you are ending 
-selectItemAtIndex: to the combo box, and the index is out of bounds.

- Jim

___

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: File Backup Date

[Jim Correia]

On Mar 16, 2010, at 12:09 PM, gMail.com wrote:

> Anyway, I have noticed that if in the Finder (MacOS X 10.6.2) I open a file
> which contains a backup date, then I modify the content and I save it, the
> backup date disappears.

What are you using to modify the content of the file? Consider the possible 
ways this could be implemented:

a) application updates the file in place
b) application writes an entirely new file, then swaps it in place of the 
original file.

In case (b) the application may or may not preserve metadata on the original 
file.

Jim


___

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: NSServices - Validation bug

[Jim Correia]

On Mar 8, 2010, at 5:02 AM, tr...@atheeva.com wrote:

> I have some queries regarding Services menu validation . I would like to 
> enable different services provided by my app based on whether a file or 
> folder is selected in the Finder. I have set NSFilenamesPboardType  as the 
> send type for the services.I have gone through the 

> 
> - (id) validRequestorForSendType:(NSString*)sendType returnType:(NSString 
> *)returnType
> 
> method but my issue is that the validation there seems to be done based on 
> the sendType and return type. In my case , the selected file and folder 
> pasteboard type is the same and I cannot determine whether the selected item 
> in the Finder is a file or folder during the validation process(This is 
> before the actual service gets invoked i.e when the services menu is being 
> shown to the user )?

> 
> So my question is that is there any way I can get some info about the 
> selected item in the Finder and validate the different service menus offered 
> by my application based on some info regarding the item rather than the basic 
> validation of the send and return types ?

You can only validate by the type in your service validation method, not the 
contents of the pasteboard.

You can switch to using more specific UTIs for our service (if you only, for 
example, want to accept a file, folder, or package). You can also use 
NSRequiredContext in your service specification to restrict service 
availability based on pasteboard content.

- Jim___

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: Core Data: inverse relationship not updated until next Run Loop

[Jim Correia]

On Mar 11, 2010, at 8:11 PM, Eric Lin wrote:

> In using Core Data, I was under the impression that if I do this:
> 
> [department addEmployeesObject:employee]  // department now has 1 employee
> [moc deleteObject:employee]
> 
> Then department will end up with no employees, assuming that the inverse
> relationship is set correctly.
> 
> But it turns out within the same run loop, if I inspect
> [department.employees count], then the value is 1. But in the next run loop,
> the value becomes 0, which is what it should be.
> 
> Is my observation accurate or is something wrong?

Deletes are propagated at the end of he current event, or at save time, 
depending how the MOC is configured.

Jim

___

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: Making a beta version expire

[Jim Correia]

On Mar 11, 2010, at 11:03 PM, Ulai Beekam wrote:

> Do you think this would work? I mean, it seems that it should, but one can 
> never be sure

Why can’t you be sure? 

After reading your algorithm and convincing yourself that the code is correct, 
step through it in the debugger with expirations dates both in the past and the 
future.

- Jim___

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 Core Data in Command Line utility

[Jim Correia]

On Mar 8, 2010, at 8:44 PM, Paul Johnson wrote:

> The problem I'm having is the line that tries to store the integer:
> [object setValue:i forKey:@"fileIndex"];
> 
> I get the compiler error "Passing argument 1 of 'setValue"forKey'
> makes pointer from integer without a cast”.

-setValue:forKey: takes an object for the value parameter. You have to pass an 
object - in this case our integer boxed in an NSNumber object.

> Can anyone tell me how to fix this? I'm also wondering if I'm using
> the correct methods to accomplish the task.

You should also take a look Core Data’s dynamically generated accessor methods. 
These are typically what you should be using to set and access properties on 
your managed objects.

Jim

___

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: NSWindow - how to resize it keeping aspect ratio?

[Jim Correia]

On Feb 18, 2010, at 1:55 PM, Alexander Bokovikov wrote:

> I'm sorry if this is a stupid question, but I don't see an evident answer... 
> I need the subj, but don't see appropriate NSWindow method... Am I missing 
> something?

-[NSWindow setAspectRatio:]
-[NSWindow setContentAspectRatio:]

- Jim

___

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: “EXC_BAD_ACCESS” on Custom TableV iew with custom NSCell when clicking.

[Jim Correia]

On Feb 17, 2010, at 4:41 PM, Gustavo Pizano wrote:

> Maybe I misunderstood I must conform NSCopying protocol also even when using 
> core data? 

If you have an NSCell subclass which adds iVars, you must implement 
-copyWithZone: to do correct memory management.

Your -copyWithZone: needs to work correctly within the bounds of the object 
ownership rules.

If you have an object iVar which shouldn’t be copied, but is still an owned 
reference, then you can (and must) retain it instead in the copy.

(You should do this for all cells with iVars, not just ones with object iVars, 
so that your subclass will behave correctly in the future even if NSCell 
doesn’t implement its behavior using NSCopyObject.)

- Jim___

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: Typo in notification name

[Jim Correia]

On Mon, Jan 18, 2010 at 4:20 AM, Mark Munz  wrote:
> Apple usually addresses binary compatibility by changing behavior if
> you link against a newer major SDK. So a behavior changes if you link
> against 10.6 vs. 10.5. Apple has done this numerous times.
>
> With notifications, they could also send out both notifications. I'm
> thinking com.apple.loginItemsListDidChange doesn't happen often enough
> to cause a performance hit if you sent out both each time.

The other thing to consider is that if you are observing a public
notification, you should use the symbolic constant for that
notification when it is available. In that case, the underlying value
doesn't matter.

The public API for interacting with, and observing, the login items
list is LSSharedFileList.

- Jim
___

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: Managed Object with Getter

[Jim Correia]

On Jan 13, 2010, at 7:29 PM, Richard Somers wrote:

> On Jan 13, 2010, at 7:14 AM, Jim Correia wrote:
> 
>> This style is typically used for non-object BOOL values. If you use it for 
>> an NSNumber attribute, you run the risk of someone familiar with the pattern 
>> assuming it is a BOOL property and writing
>> 
>>  if ([managedObject isSelected]) {
>> 
>>  }
>> 
>> which will of course be be true whenever the property is non-nil, no 
>> regardless of the actual value.
> 
> Thanks for the reply and the information. The attribute is a bool in the 
> Xcode data modeler. When you copy the Obj-C 2.0 method declarations for the 
> bool attribute to the clipboard this is what you get.
> 
> @property (nonatomic, retain) NSNumber *selected;
> 
> So I assume that Core Data frameworks default is to use a NSNumber for a bool 
> value.

That is correct. Core Data always expresses its attribute as object values.

If you want to have a scalar BOOL property on your object, you can, but you 
must write your own accessors. (And be aware of and deal with 
-setNilValueForKey: as appropriate.)

For example:

- (BOOL)isSelected 
{
NSNumber *selected = nil;

[self willAccessValueForKey: @"selected"];
selected = [self primitiveSelected];
[self didAccessValueForKey: @"selected"];

return (selected != nil) ? [selected boolValue] : NO;
}

- (void)setSelected:(BOOL)selected 
{
[self willChangeValueForKey: @"selected"];
[self setPrimitiveSelected: [NSNumber numberWithBool: selected]];
[self didChangeValueForKey: @"selected"];
}

- Jim



___

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: Managed Object with Getter

[Jim Correia]

On Jan 13, 2010, at 8:39 AM, Richard Somers wrote:

> On Jan 13, 2010, at 1:17 AM, Kyle Sluder wrote:
> 
>> This code cannot generate that error because it doesn't send any
>> messages. You need to post the code to which the error actually
>> refers.
> 
> This works.
> 
> @property (nonatomic, retain) NSNumber *selected;
> 
> if ([[self selected] boolValue]) { ... }
> 
> This produces error "unrecognized selector sent to instance".
> 
> @property (nonatomic, retain, getter=isSelected) NSNumber *selected;
> 
> if ([[self isSelected] boolValue]) { … }

Assuming you are using @dynamic and relying on Core Data to generate the 
accessors for you, that isn’t going to work. Core Data doesn’t support the 
isPropertyName getter style.

This style is typically used for non-object BOOL values. If you use it for an 
NSNumber attribute, you run the risk of someone familiar with the pattern 
assuming it is a BOOL property and writing

if ([managedObject isSelected]) {

}

which will of course be be true whenever the property is non-nil, no regardless 
of the actual value.

If you want or need to use custom getters/setters, you’ll have to provide your 
own accessors rather than relying on the @dynamic generated ones.

- Jim___

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: Enabling NSZombieEnabled programatically

[Jim Correia]

On Jan 9, 2010, at 7:54 PM, Graham Cox wrote:

> I'm getting something odd though. If I set NSZombieEnabled during app 
> delegate -applicationDidFinishLaunching: method from my preference, a later 
> deliberate over-release still causes a EXC_BAD_ACCESS. If I set it as an 
> environment variable (Xcode launch arguments for my app) I get the zombie 
> logged as expected. In either case I can see that NSZombieEnabled is YES, but 
> directly setting it doesn't appear to work.
> 
> Any idea what I'm doing wrong?
> 
> Code:
> 
> #include 
> 
> 
> - (void)  applicationDidFinishLaunching:(NSNotification*) 
> aNotification

The comment in NSDebug.h reads:

"This is mostly a benefit if you need to initialize these variables to some 
non-default value before your program's main() routine gets control”

My guess is that -applicationDidFinishLauching: time is too late.

- Jim___

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: Enabling NSZombieEnabled programatically

[Jim Correia]

On Jan 9, 2010, at 6:22 PM, Sean McBride wrote:

> I've never tried, but NSDebug.h seems to say that there is an
> NSZombieEnabled global var that you can set.

Read the warning at the top of the header. While those variables have been 
there as long as I can remember, as soon as you ship software which relies on 
them, Murphy will invoke his law.

(For debug and pre-release builds, go wild. For production builds, take 
appropriate precautions.)

- Jim___

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: NSFilesPromisePboardType

[Jim Correia]

On Jan 8, 2010, at 11:33 PM, Jim Correia wrote:

> Ultimately, it appears that iTunes is putting bad promise data in the drag. 
> At this point it is probably best to file a bug against iTunes.

Looks like this might simply be a byte order bug with the value in the 
promisedFlavor field in the PromiseHFSFlavor data on the pasteboard (though 
that doesn’t really help when accessing the data via NSPasteboard.)

On Jan 8, 2010, at 11:37 PM, Nick Paulson wrote:

> I am trying to get the path of an item dropped onto a view by iTunes.
> 
> iTunes seems to only do this with NSFilesPromisePboardType.  Maybe I am 
> wrong, though.

The promised data isn’t really what you wanted anyway. The idea behind the 
promise is that the drag provider *promises* to create a new file in location 
specified in the drag receiver. It sounds like you are interested in the file 
that is being dragged, not a copy of it.

iTunes does provide all the data in the drag that you need (and more), but does 
so through the ‘itun’ drag flavor. I haven’t seen this documented anywhere as a 
public, supported pasteboard flavor. It has been available for some time, but 
unless it is documented as a such, it could disappear (or change) at any time. 
Proceed accordingly.

- Jim___

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: NSFilesPromisePboardType

[Jim Correia]

On Jan 8, 2010, at 10:24 PM, Nick Paulson wrote:

> I am doing the following code:
> 
> NSArray *filenames = [sender namesOfPromisedFilesDroppedAtDestination:[NSURL 
> fileURLWithPath:@"/"]];
> 
> However, it is getting the following lines in console:
> Couldn't get a copy of an HFS Promise from the pasteboard
> Looked for HFSPromises on the pasteboard, but found none.
> 
> I am dragging from iTunes in list view to my own view.
> 
> Am I making a mistake somewhere?

A couple of things to keep in mind here…

* See the comment in the documentation about having to assume the files haven’t 
been created yet? You have to assume that, and it can complicate things. (I’ve 
logged a lengthy bug requesting ways to simplify file promise dragging, but the 
problem is complex because it requires system framework support *and* the 
cooperation of applications doing the dragging.)

* You specified the drop location as /. Not all users have write permissions to 
the root of the disk, and it is unlikely that you really wanted iTunes to 
create the files here anyway.

Ultimately, it appears that iTunes is putting bad promise data in the drag. At 
this point it is probably best to file a bug against iTunes.

What is it you are actually trying to accomplish? Perhaps there is another 
solution.

Jim___

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: NSFilesPromisePboardType

[Jim Correia]

On Jan 8, 2010, at 7:52 PM, Nick Paulson wrote:

> Can someone please explain to me how I handle NSFilesPromisePboardType?  I 
> register for the dragged types, but I don't understand exactly how to get the 
> data.

Did you read the documentation?



What part of the inline sample code are you having trouble understanding?

NSURL *dropLocation; // Assume this exists
 
- (BOOL)performDragOperation:(id )sender
{
NSPasteboard *pboard = [sender draggingPasteboard];
 
if ( [[pboard types] containsObject:NSFilesPromisePboardType] ) 
{
NSArray *filenames = [sender

namesOfPromisedFilesDroppedAtDestination:dropLocation];
// Perform operation using the files’ names, but 
without the
// files actually existing yet
}
return YES;
}


- Jim___

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: duplicate method found, but can't locate

[Jim Correia]

On Jan 6, 2010, at 7:33 PM, Shane wrote:

> I'm getting this error from my main app controller which is trying to
> send a message to a thread on a queue (using YAMessageQueue).
> 
> [[networkQueue proxyForTarget:networkServer] save:[self saveFile]];
> Cannot convert 'NSString*' to 'NSError**' in argument passing
> 
> I have no idea where it thinks it needs to convert to NSError**. [self
> saveFile] returns an NSString*. In my server file I have the
> following:

The problem is likely that -proxyForTarget: returns an `id`.

> I can see that there seems to be some confusion with the save: method
> in the @protocol because the warning seems to imply that it found
> 'another' save: by saying 'Also found …'.
> 
> But I have no idea where it's finding the save: method that needs to
> use NSError**, if that is the case.

NSManagedObjectContext defines -(BOOL)save:(NSError **)error;

If -proxyForTarget: returns an id, then the compiler doesn’t know which of the 
available signatures you really meant to use.

You can either rename your -save: method, or cast the result of -proxyForTarget 
to the static type you are sending the message to.

- Jim___

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 do bindings make my image disabled?

[Jim Correia]

On Jan 1, 2010, at 5:39 PM, Michael Bishop wrote:

> What I have been able to figure out is that using bindings is somehow making 
> my NSImageCell instances draw an image semi-transparently. I haven't been 
> able to explain why this is but I'm hoping someone else has seen this.

[…]

> In the project you'll find an app delegate which has a status. There are some 
> Image cells along the left that are bound to the status. They have different 
> border types but are otherwise identical (copy and pasted).
> 
> On the right, you'll see an image who is manually set by the delegate when 
> its status is flipped.
> 
> The only one that is faded out is the image that has no border and is bound.
> 
> I can't for the life of me explain the faded image. Can anyone else?

On 10.6, NSImageView will draw its content as dimmed when the control is 
disabled.

Your binding has “Conditionally Sets Enabled” turned on.

- Jim ___

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: Reference NSSearchFieldCell within NSSearchField?

[Jim Correia]

On Dec 29, 2009, at 3:02 PM, Mike Chambers wrote:

> The NSSearchField docs mention that it wraps an NSSearchFieldCell,
> which contains such APIs as setSendsWholeSearchString (which i need to
> dynamically set).
> 
> Is there any programatic way to get a reference to the
> NSSearchFieldCell wrapped in a particular NSSearchField? I know I can
> link an IBOutlet to it directly in IB, but I would prefer just to use
> my existing outlet to the search field.

NSSearchField inherits from NSControl.

NSControl has a -cell method.

- Jim
___

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: Creating CoreData classes

[Jim Correia]

On Dec 27, 2009, at 11:52 PM, Brian Bruinewoud wrote:

> If I've generated core data classes from my model, how do I use them to 
> create new values?
> Currently I'm doing this:
> 
>CDClass *cdObject = (CDClass *)[ NSEntityDescription 
> insertNewObjectForEntityForName: @"CDClass" inManagedObjectContext: context ];

This is in fact the correct idiom for doing this.

> But I'm looking for something more like this:
> 
>CDClass *cdObject = [ CDClass insertNewObjectInManagedObjectContext: 
> context ];

You can use 

CDClass *object = [[CDClass alloc] initWithEntity: entity 
insertIntoManagedObjectContext: context];

instead, but it is more verbose since you have to separately look up the entity 
by name.

- Jim

___

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: NSViewAnimationFadeInEffect broken in Snow Leopard?

[Jim Correia]

On Tue, Dec 22, 2009 at 7:58 PM, Kevin Wojniak  wrote:
> NSViewAnimationFadeInEffect doesn't seem to be working like it was in Leopard.

I filed this as when SL was being seeded. If

Regression: NSViewAnimation doesn't fade in view if initially hidden
rdar://problem/7020352

If this affects you, you should also file a bug.

The workaround is to show the view before running the animation.

- Jim
___

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: Are views active or inactive?

[Jim Correia]

On Dec 22, 2009, at 10:45 PM, Rick Mann wrote:

> I'm listening for that notification. Sure is a clunky way to do things. I've 
> never used a view framework that didn't tell views when they became 
> active/inactive.

AppKit views don’t have a native active/inactive status like carbon HIToolbox 
controls.

For views which alter their appearance based on whether or not they are first 
responder, or in a key/main window, you must track this state yourself.

> I'm not sure viewDidMoveToWindow works here. I'm using it already for 
> something else, and it gets called 3 times when my window is created, but 
> never when activated.

You can use viewWill/DidMoveToWindow to stop observing notifications on your 
preview window (if any) and start observing notifications on your new container 
window (if any), not for the tracking of the “active” state itself.

- Jim

___

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: Core Data registered editors?

[Jim Correia]

On Tue, Dec 22, 2009 at 9:33 PM, Rick Mann  wrote:
> I noticed that NSPersistentDocument's docs for isDocumentEdited talks about 
> "editors registered with
> the managed object context."
>
> Is there something that explains what a "registered editor" is?

See the docs about NSEditorRegistration protocol:

http://developer.apple.com/mac/library/documentation/Cocoa/Reference/ApplicationKit/Protocols/NSEditorRegistration_Protocol/Reference/Reference.html

- Jim
___

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: Detaching a thread to save a file?

[Jim Correia]

On Fri, Dec 11, 2009 at 8:43 PM, Keith Blount  wrote:

> Oops, sorry, forgot to ask: what happens if the user hits save again while 
> the separate
> thread is already saving the search indexes to disk?

Also consider what happens when the user hits save, then immediately
quits before your file writing thread has finished writing the file to
disk.

(Since users generally get upset when you lose their data, you'll have
to handle this case as well.)

- Jim
___

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 Text Colors Right for Derived NSTextFieldCell under Various Conditions

[Jim Correia]

On Dec 4, 2009, at 10:54 PM, Grant Erickson wrote:

> Thanks for the prompt reply. I forgot to mention one key element. The
> subclass must support 10.4- and later so I cannot rely on 10.5- and later
> APIs such as this.

In that situation I’d probably write (and test!) 

if ([self respondsToSelector: @selector(interiorBackgroundStyle)]) {
… compute text color from background style ...
} else {
… best effort text color calculation ...
}

Hopefully sometime soon the 10.4 branch can just go away.

(If you have an existing product which already targets 10.4, changing the 
system requirements for something like this seems drastic. However, if you are 
developing a new product, you may want to seriously consider whether you really 
need to support Tiger, and the opportunity cost of doing so.)

Jim___

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 Text Colors Right for Derived NSTextFieldCell under Various Conditions

[Jim Correia]

On Dec 4, 2009, at 1:52 PM, Grant Erickson wrote:

> I have implemented a custom class that derives from NSTextFieldCell to draw
> status for an associated device (not unlike the network source list in the
> Network System Preferences) overriding only:

[…]

> The issue I am having is determining the correct set of conditionals to use
> to determine when to invert the color of my text strings from
> backgroundColor (when the cell is selected and the associated table has
> focus) to textColor (when the cell is not selected or is selected and does
> not have focus).

In general, the way to do this is to send -interiorBackgroundStyle to yourself, 
and pick a foreground text color suitable for the background.

- Jim

___

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

Behavior of CFStringUpper/Lowercase for NULL Locale?

[Jim Correia]

The documentation for CFStringUppercase says:

locale

A CFLocale object that specifies a particular language or
region. Prior to Mac OS X v10.3, this parameter was an untyped
pointer and not used.

The locale argument affects the case mapping algorithm. For
example, for the Turkish locale, case-insensitive compare
matches “I” to “ı” (Unicode code point U+0131, Latin Small
Dotless I), not the normal “i” character.


Prior to 10.3, the parameter was unused and NULL was always passed.

The documentation doesn’t specify what the case mapping behavior is if NULL is 
passed for a locale. What is the behavior in that case? (Current locale? System 
locale?)

- Jim___

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: Dealing with faults in CoreData / AppleScript

[Jim Correia]

On Nov 19, 2009, at 10:19 PM, Greg Hoover wrote:

> I've been working through adding scriptability to my CoreData application but 
> ran into a snag: accessing relationships may return a fault instead of the 
> actual set of objects.  Is there an "accepted" way of solving this issue?  I 
> don't want the accessor to "always" fault the objects in, just when they're 
> being accessed from AppleScript.  Is creating another accessor the proper way 
> to get around this?

Can you be more specific about the problem you are having and what specifically 
you are trying to do?

In particular, I’m not sure what you mean by:

I don't want the accessor to "always" fault the objects in, ust when 
they’re being accessed from AppleScript.

Unless you prefetch relationships, they will be a fault. (If you need to know, 
you can test this with -hasFaultForRelationshipNamed:. If you need to access 
the related objects though, you need to fire the fault.

- Jim


___

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: Question about Style wrt "private" methods

[Jim Correia]

On Nov 18, 2009, at 2:32 PM, Jens Alfke wrote:

> On Nov 18, 2009, at 11:15 AM, Jim Correia wrote:
> 
>> This problem is just not restricted to private methods, or additions through 
>> categories. You can also run afoul of a namespace conflict with a public 
>> method in your subclass.
> 
> Yes; but this is less likely because Apple engineers add public methods less 
> often (almost never in minor updates) and more carefully.

Agreed. Subtle breakage is still possible across a major release when both you 
and Apple choose the obvious name for new functionality.

-isHidden/-setHidden: on NSView is the sort of thing (which was added recently 
enough that many people likely remember it.)

If you had subclassed NSView to add that functionality, its pretty likely you 
would have chose those names, or -isVisible/-setVisibible: because they are the 
obvious solution to the problem.

You can avoid namespace clash altogether by writing ugly code, with names no 
one would ever want to use for your methods and properties. But who wants to do 
that? :-)

Jim___

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: Question about Style wrt "private" methods

[Jim Correia]

On Nov 18, 2009, at 1:49 PM, Jens Alfke wrote:

This is unfortunately true. The danger is that if you add an _- 
prefixed method to your class, it might conflict with a private  
method declared in a superclass. If this happens your method will  
override the internal one, and Really Bad Things will probably  
happen. It's a rare problem but it has happened before. (And even if  
everything works well now, any future OS update might add a new  
private method in a superclass that breaks your code. Again, this  
has actually happened.)


This problem is just not restricted to private methods, or additions  
through categories. You can also run afoul of a namespace conflict  
with a public method in your subclass.


Suppose you have a subclass of NSView which adds - 
reallyNiceMethodThatOughtToBeInAppKit. If Apple adds such a method in  
the next OS release, the same kinds of problem can result. (Similar  
problems when you extend delegate protocols for your subclasses too.)


This problem is much less insidious than the category problem, but the  
potential does exist. (For both private and public methods.)


- Jim

___

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

Dependent/injected tests vs. linker visibility

[Jim Correia]

[Moved from Xcode-Users since it really isn’t an Xcode issue.]

I have some dependent/injected unit tests to test application-specific  
functionality. I've added a test that calls a C function which is  
defined in the application. This causes a link-time error because, by  
default, applications are compiled with -fvisibility=hidden.


What practical solution are you all using for this problem?

- Globally make symbols visible?
- Only for Debug (or a Test specific) configuration?
- Mark specific symbols, as required for the unit tests, to be visible?
- Avoid using dependent/injected tests?

Jim___

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: removeObserver:forKeyPath: when already removed

[Jim Correia]

On Nov 11, 2009, at 7:36 PM, Jerry Krinock wrote:

>> Trying to remove an observer for a key path which has already been removed 
>> (or never registered) falls into the programming error class...
> 
> Almost always true.  But what if, and yes I know this is pretty flakey but 
> just assume it could happen, that I needed to do some very expensive 
> operation to determine whether or not an observation had been removed.  Then, 
> it would be efficient and smart to send the removeObserver:: message 
> unconditionally.

I think that’s a bit of a stretch. It may be less tedious to always send 
-removeObserver:forKeyPath:, but it usually is not expensive (in either time or 
space) to keep track of the objects you are observing. (And since it is 
currently necessary, it is a mostly moot point.)

(For an extra bit of trivia, see the Snow Leopard release notes about automatic 
removal of observers on 10.6 and later when running under GC.)

> But how is the programmer to know this is an error if it's not documented as 
> a "don't do"?  If indeed this is not documented, then in my opinion it's a 
> documentation error.

Without being overly pedantic, the documentation says:

observer
The object to remove as an observer.

keyPath
A key-path, relative to the receiver, for which observer is
registered to receive KVO change notifications. This value must 
not
be nil.

Combining the two, ‘oberver’ must be registered as an observer of ‘keyPath’ or 
else it is an error.

Could the documentation be more explicit? Sure.

Since this method has always raised an exception for this programming error, as 
soon as you commit this error you’ll get the exception and fix it. 

The real problem for this type of exception comes when the exception is added 
for a programming error that used to be silently allowed (even if that error 
resulted in undefined behavior.) In this case, Apple cannot unconditionally add 
the exception, because they may break binary compatibility (such that it is) 
with applications relying on the silent, but undefined error behavior. 

In these situations, an exception is usually added with a linked-on-or-after 
test, and documented in the release notes.

- Jim___

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: removeObserver:forKeyPath: when already removed

[Jim Correia]

On Nov 11, 2009, at 5:35 PM, James Walker wrote:

> I had a situation where removeObserver:forKeyPath: was called twice for the 
> same receiver (an NSUserDefaultsController), the same observer, and the same 
> key path.  It threw an NSRangeException.  But the KVO docs don't say anything 
> about removeObserver:forKeyPath: throwing an exception.  Should this be 
> considered a documentation bug, or is it a bug that 
> -[NSUserDefaultsController removeObserver:forKeyPath:] threw an exception, or 
> should I have just expected this due to some general principle?

Cocoa reserves the use of exceptions for programming errors an unexpected 
runtime errors.

Trying to remove an observer for a key path which has already been removed (or 
never registered) falls into the programming error class, so I wouldn’t expect 
this to be explicitly documented.

Jim

___

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: Programmatic Binding KVC KVO

[Jim Correia]

On Nov 4, 2009, at 3:17 PM, Richard Somers wrote:


@interface MyView : NSView
{
double num;
}
@end

@implementation MyView

- (double)num
{
return num;
}

- (void)setnum:(double)newNum
{
[self willChangeValueForKey:@"num"];
num = newNum;
[self didChangeValueForKey:@"num"];
}

@end


When the view changes num, it need to push the change to the bound  
model objects.


See mmalc’s Graphics Bindings sample:

http://homepage.mac.com/mmalc/CocoaExamples/controllers.html

- Jim

___

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: nonsense from cellFrameAtRow:column: on Leopard

[Jim Correia]

On Oct 31, 2009, at 2:34 PM, Matt Neuburg wrote:

Solved my own problem, which had nothng to do with what I thought it  
had to
do with. The source of the difficulty is actually this: On Snow  
Leopard,
[self controlView] works - it points from the NSCell to the NSMatrix  
that

contains it. On Leopard, it doesn't work.


See the section titled "NSCell controlView property changes” in the  
Snow Leopard AppKit release notes.


Jim

___

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: Core Data Managed Object with Custom Data Caches

[Jim Correia]

On Oct 31, 2009, at 12:25 PM, Richard Somers wrote:


On Oct 31, 2009, at 10:12 AM, Jim Correia wrote:


Typically you set up the baseline value in -awakeFromFetch.

Is there a specific reason you need this to be undoable?


When the managed object is removed, didTurnIntoFault is called and  
the custom data caches are deleted. When the user does an undo on  
this action the managed object is fetched but awakeFromFetch is  
never called so the custom data caches are now missing.


On 10.5 and earlier, you’ll have to re-create your data caches in an  
on-demand accessor.


On 10.6 and later you can use -awakeFromSnapshotEvents: to re- 
establish them.


Ji



smime.p7s
Description: S/MIME cryptographic signature
___

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: Core Data Managed Object with Custom Data Caches

[Jim Correia]

On Sat, Oct 31, 2009 at 12:05 PM, Richard Somers
 wrote:

> The documentation states didTurnIntoFault "method may be used to clear out
> custom data caches". So far so good.
>
> I was creating custom data caches in awakeFromFetch (or awakeFromInsert)
> until I realized that awakeFromFetch does not participate in undo. The
> documentation confirms that "change processing is explicitly disabled
> around this method".
>
> So Apple provides a way to remove custom data caches associated with a
> managed object but no way to create them.

Typically you set up the baseline value in -awakeFromFetch.

Is there a specific reason you need this to be undoable?

Jim
___

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: Programmatically invoking double-click bindings on NSTableView

[Jim Correia]

On Oct 28, 2009, at 1:51 PM, Gerd Knops wrote:


On Oct 28, 2009, at 11:41 AM, Jim Correia wrote:


If you are looking for a more general solution, you can retrieve the
binding data from the table view using -infoForBinding:, valuate the
target & arguments, and invoke the method that way.



I could, but given the flexible number of arguments that is a bit of  
work.


The cocoa frameworks provide a lot of functionality. But sometimes you  
have to write some code...


And since the tableview already knows how to do that, I was  
wondering if that code was exposed in some way. Apparently not.


For now I just added a transparent button with similar bindings and  
trigger that one.


Creating an invisible button to trigger a programmatic action is  
generally the wrong way to design your code flow. (If you are already  
doing work programmatically, don’t call back into the view layer to do  
additional work by side effect, do it directly.)


- Jim___

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: Programmatically invoking double-click bindings on NSTableView

[Jim Correia]

On Wed, Oct 28, 2009 at 12:25 PM, Gerd Knops  wrote:
> Seems this should be easy:
>
> I have an NSTableView's "Double Click Target", "Double Click Argument",
>  "Double Click Argument2" etc. binding set, and they work fine.
>
> Now I would like to programmatically get the tableview to issue a
> double-click response. Obviously "target" and "doubleAction" are not useful
> in that context.
>
> Any ideas how to get this done?

If this is just a one-off case, the simplest solution is to invoke the
method with the appropriate arguments on the target directly.

If you are looking for a more general solution, you can retrieve the
binding data from the table view using -infoForBinding:, valuate the
target & arguments, and invoke the method that way.

- Jim
___

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: keyPathsForValuesAffectingValueFor: not updating immediately

[Jim Correia]

On Tue, Oct 27, 2009 at 11:39 AM, Brad Gibbs  wrote:

> I don't think there's anything wrong with the method -- I've used it
> successfully in other apps and have seen it work in Apple sample apps, it
> just isn't working in this particular case.  I'm sure that there's something
> wrong on my end.  Just looking for a little help in figuring out what that
> could be...

We are at the point where we can't offer useful advice without seeing
your code. Since +keyPathsForValuesAffecting is not fundamentally
broken, the most likely cause is a problem with your code.

> - (NSSet *)keyPathsForValuesAffectingValueForFullAddress {
>return [NSSet setWithObjects:self.streetAddress, self.city,
> self.state, self.zipCode, nil];
> }

This is wrong for 2 reasons.

Reason #1 - see below.

Reason #2 - you should be passing the literal key paths which affect
the value, not self.streetAddress which is the current value of that
key path.

> I've also tried:
>
> - (NSSet *)keyPathsForValuesAffectingValueForFullAddress {
>return [NSSet setWithObjects:@"streetAddress", @"city", @"state",
> @"zipCode", nil];
> }

+keyPathsForValuesAffecting should be implemented as a class
method. (That's what the + prefix means.) You've implemented it as an
instance method. Change it to a class method and you should find that
it works correctly.

(If you set a breakpoint on your code as written, you'll find that it
is never called, which should be a warning sign to you that something
is wrong.)

- Jim
___

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: keyPathsForValuesAffectingValueFor: not updating immediately

[Jim Correia]

On Mon, Oct 26, 2009 at 10:03 PM, Brad Gibbs  wrote:

> I did read the documentation, which is why I used
> +keyPathsForValuesAffectingFullAddress.  I also tried
> +keyPathsForValuesAffectingValueForFullAddress:
>
> Both methods work, but only after changing the view and then coming back to
> it.

+keyPathsForValuesAffecting is not fundamentally broken - I use
it on my NSManagedObject subclasses as needs dictate.

At this point, I think we need more information to help. At the very
least, let's see your implementation of +
keyPathsForValuesAffectingFullAddress.

- Jim
___

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: Trying to understand -awakeFromNib problem

[Jim Correia]

On Oct 25, 2009, at 12:22 PM, Roland King wrote:

Where's that documented??? I looked for that when I saw the original  
question. For iPhone it's documented under NSObject UIKit Additions  
Reference, for 10.6 it's under NSNibAwakingProtocol which doesn't  
show up anywhere as an implemented protocol of NSObject.


I'm sure you're right but I've tried every search I can reasonably  
think of in the documentation and I can't find it. 	


See the Snow Leopard AppKit release notes.

- Jim



smime.p7s
Description: S/MIME cryptographic signature
___

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: Trying to understand -awakeFromNib problem

[Jim Correia]

On Oct 25, 2009, at 7:53 AM, Graham Cox wrote:

Then, running on 10.5.(8) I stared to get an error that the object  
 did not respond to - 
awakeFromNib. I'd not previously seen this error until I refactored  
my code. Sure enough my new base class and its new fork don't  
implement awakeFromNib because they don't need to. On 10.6 there is  
no error. On 10.5.8 I get the error, suggesting that 10.6 first  
checks whether -awakeFromNib is implemented and 10.5 does not.


Do your controller subclasses send [super awakeFromNib]?

NSObject implements -awakeFromNib on 10.6 and later (but not  
previously.)


- Jim

smime.p7s
Description: S/MIME cryptographic signature
___

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: Preserving camelCase when prefixing a method name?

[Jim Correia]

On Oct 13, 2009, at 10:24 AM, Graham Cox wrote:

I think just making the first character uppercase would be  
sufficient, but I'm not sure how to do that reliably with the  
unichar data type, so that's my first question.


Well you don't have to consider all of unicode, just those characters  
which are valid language identifiers.


Nonetheless, applying -uppercaseString to the first character should  
solve the problem.


The second is that when generating selectors programmatically like  
this, are there other pitfalls to consider so that the resulting  
selector is always predictable? The scheme needs to work with  
arbitrary properties, just as various parts of KVC and KVO do.


Don't forget to account for the pattern often used for BOOL  
'properties'. (I put properties in air quotes because actual  
properties allow you to specify the getter and setter name.)


-isFlagged
-setFlagged:

- Jim
___

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: [UTI] Anyway to retrieve the UTI of a file without having to use a FSRef?

[Jim Correia]

On Mon, Oct 12, 2009 at 6:36 PM, Iceberg-Dev  wrote:

> From what I've found in the documentation, the UTI type of a file can be 
> retrieved using the
> LaunchServices APIs. This requires to provide a FSRef.
>
> Wouldn't there be an API I didn't see in Foundation that lets you obtain the 
> type without having to
> convert, at least, a NSURL to a FSRef?

See, on NSWorkspace:

- (NSString *)typeOfFile:(NSString *)absoluteFilePath error:(NSError
**)outError;

- Jim
___

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

NSAttributedString always returns NO for -isEqualToAttributedString with attachments present?

[Jim Correia]

The documentation for -[NSAttributedString isEqualToAttributedString:]  
says:


	Attributed strings must match in both characters and attributes to be  
equal.


It doesn't mention attachments at all.

Consider the following code:

	attributedString is the input string, and it may contain image  
attachments


	NSData *d = [NSKeyedArchiver archivedDataWithRootObject:  
attributedString];
	NSAttributedString *s1 = [NSKeyedUnarchiver unarchiveObjectWithData:  
d];
	NSAttributedString *s2 = [NSKeyedUnarchiver unarchiveObjectWithData:  
d];


NSLog(@"isEqual = %d", [s1 isEqualToAttributedString: s2]);

When 'attriburedString' does not contain image attachments, 1 is logged.

When 'attributedString' does contains image attachments, 0 is logged.

Is this the expected behavior? The correct behavior?

- Jim
___

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: Can an app query the values in its own Info.plist?

[Jim Correia]

On Sep 28, 2009, at 1:34 AM, Kyle Sluder wrote:


Done.  rdar://problem://7257097

Also notes that the same collision issue exists for user defaults,
where AppKit stores window and splitter autosave information.


It is unlikely that any apple framework is ever going to introduce a  
key of the form


com.your-domain.KeyName

If you define your custom Info.plist keys using reverse DNS style  
keys, the collision issue should become a non-issue.


Jim
___

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: [UNSOLVED] How to observe every child entity

[Jim Correia]

First, a bit about terminology.

What you are talking about is not observing child entities. You wish  
to observe related instances (NSManagedObject or subclass thereof).  
The entity describes the instance, but is not the instance.


(I'm trying to avoid being overly pedantic, but talking about things  
using a common language that everyone agrees upon is important.)


On Sep 16, 2009, at 9:25 PM, Matthew Lindfield Seager wrote:


When a user undoes a deletion it calls the primitive accessor which
does not emit KVC/KVO notifications.


It is correct that when you undo/redo, Core Data does not typically  
call the public accessors. It does, however, emit the required KVO  
change notifications when updating the values.


(And as such, you can observer the relationship if you need to know  
when it changes. Self-observing, unfortunately, opens up another can  
of worms.)


If you have a specific situation where NSManagedObject is no emitting  
KVO change notifications on undo or redo, that would be a bug.


- Jim
___

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: restoring NSSplitView divider's position

[Jim Correia]

On Sep 3, 2009, at 6:00 PM, kvic...@pobox.com wrote:

ps. i do see the method for setting the divider's position... i just  
don't see how to get the current position for saving.


The divider’s current position can be computed by looking at the  
frames of the subviews (account for both orientation and width of the  
divider too.)


Jim

___

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 a StringPtr from Gestalt on 64-bit

[Jim Correia]

On Sep 2, 2009, at 9:02 PM, Scott Lahteine wrote:

My preference pane uses Gestalt(gestaltUserVisibleMachineName,  
&mySInt32), coercing the SInt32 into a StringPtr to get the Machine  
Name. On 64-bit I get a warning because StringPtr is a 64-bit  
pointer on that architecture. Does Gestalt() ensure that the  
returned pointer is in the low 4 gigs of RAM, or is there a more  
appropriate way to get the Machine Name that is 64-bit compatible?


extern CFStringRef CSCopyMachineName(void)  
AVAILABLE_MAC_OS_X_VERSION_10_0_AND_LATER;


- Jim


___

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: Replace NSScroller instances in NSScrollView?

[Jim Correia]

On Aug 21, 2009, at 12:22 PM, Jim Correia wrote:

I have a subclass of NSScrollView in which I'm also using a custom  
NSScroller subclass.


If I am unarchiving from a NIB, I can set the object classes for the  
NSScroller to my subclass.


However, there doesn't appear to be an API to specify the class of  
the scroller used when I alloc+init an NSScrollView, nor does there  
seem to be an API to replace the default scrollers with my custom  
scrollers.


Have I overlooked the solution?


Yes. There are public accessors to replace the scroller.

- Jim

___

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

Replace NSScroller instances in NSScrollView?

[Jim Correia]

I have a subclass of NSScrollView in which I'm also using a custom  
NSScroller subclass.


If I am unarchiving from a NIB, I can set the object classes for the  
NSScroller to my subclass.


However, there doesn't appear to be an API to specify the class of the  
scroller used when I alloc+init an NSScrollView, nor does there seem  
to be an API to replace the default scrollers with my custom scrollers.


Have I overlooked the solution?

Thanks,
Jim
___

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: NSArrayController subclass computed property not being updated

[Jim Correia]

On Jul 29, 2009, at 1:09 AM, Alexander Heinz wrote:


I've got an NSArrayController subclass with a computed property:

@property (readonly) BOOL canRemoveSelection;

(This basically adds some application-specific checks to the  
canRemove: property, so that I can bind a "remove" button in  
Interface Builder.)


I want the property to be recomputed every time the "selectionIndex"  
or "selectionIndexes" of the array controller change, so, in my ".m"  
file, I have:


+ (NSSet*)keyPathsForValuesAffectingCanRemoveSelection
{
	return [NSSet setWithObjects:@"selectionIndex",  
@"selectionIndexes", nil];

}


[...]


Does anyone see anything immediately wrong here?


My previous post explains why this doesn't work:

http://lists.apple.com/archives/Cocoa-dev/2009/Apr/msg01683.html

Jim

___

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?

[Jim Correia]

On Jul 22, 2009, at 3:42 PM, Squ Aire wrote:

Yes I was aware of this method. However, according to the docs it  
only guarantees uniqueness for a given process, and doesn't  
guarantee it will be "globally" unique, if you know what I mean.  
Thanks for the suggestion though.


The documentation for -[NSProcessInfo globallyUniqueString] does  
guarantee that the string will be globally unique. (The word global is  
right in the name.)


What it it doesn't guarantee, though, is that it will give you back a  
UUID string (which is generally meant to be a specific kind of  
globally unique string.)


-globallyUniqueString returns UUID strings today, but this wasn't  
always true. (And since it isn't guaranteed by the docs, may not be  
true in the future.)


If you specifically need a UUID string, you should probably use CFUUID.

- Jim
___

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: AppleScript Record and NSDictionary

[Jim Correia]

On Jul 21, 2009, at 6:51 PM, Steve Cronin wrote:

From my Cooca app I want to call an AppleScript with several  
parameters one of which is an AppleScript record.


[...]

When I look at the docs for -recordDescriptor and I see the 4  
character code business - I get the ugly sinking feeling but maybe  
that's the only way...


I've tried to move the coercion into AppleScript but there is  
nothing that will do that except the magic 'user record fields'.

But there are warnings about resouce use for the 'magic' values
(And anyway while seated in front of this machine, I don;t really  
like magic!!)


- (NSAppleEventDescriptor) coerceDictionary(NSDictioanary  
*)dictionary {
	NSAppleEventDescriptor *result = [NSAppleEventDescriptor  
recordDescriptor];

NSArray *keysArray  = [dictionary allKeys];
NSArray *valuesArray = [dictionary allValues];
unsigned int keyCoount = [keysArray count];
for ( i = 0; i < keyCount; i++) {

//the good stuff happens here

}
return result;
}

So  is there an exprienced soul out there who might share a snippet  
of code or a an illuminating link?


Since you omitted "// the good stuff happens here" it isn't clear to  
me what, if anything, you've already tried.


An AE record is keyed by four character codes. There is a special key  
- keyASUserRecordFields - which is an AEList of sequential keys and  
values. When this is present in the record, these appear as string  
keys to AppleScript.


So, create an AE record, and add an appropriate list to the record  
with the key of keyASUserRecordFields. Whether you write this code  
using the C API or the Foundation NSAppleEventDescriptor API is now  
just an implementation detail :-)


Jim

___

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: NSTextView help

[Jim Correia]

On Jul 20, 2009, at 11:49 PM, Dave DeLong wrote:

The problem I'm facing is that in order to display the RTF data  
properly, I need an NSTextView, but NSTextView has its own enclosing  
scroll view that's conflicting with my overall scrollview.


NSTextView doesn't intrinsically have its own scroll view.

When you drag a text view (or a table view) out of the Library pallet  
in IB, you are actually dragging an NSScrollView with an NSTextView as  
the document view since this is a typically useful configuration.


If some other configuration is more useful to you, feel free to  
configure the view hierarchy that way.


Look at the TextSizingExample in the examples folder for examples of  
other common configurations.


- Jim
___

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 approach for selective spell-checking...

[Jim Correia]

On Jul 20, 2009, at 9:48 PM, Daniel Jalkut wrote:

I'm trying to figure out how to prevent spell checking in my  
NSTextView from being applied to certain ranges within the text  
document.


[...]

It would be nice if NSSpellChecker offered a delegate relationship,  
something like "willMarkMispelledWordInRange:" ... where I could  
choose to have it keep quiet instead.


On 10.5 and later, there is public API which you can override in your  
NSTextView subclass:


- (void)setSpellingState:(NSInteger)value range:(NSRange)charRange;
// May be called or overridden to control setting of spelling and  
grammar indicators.  Values are those listed for  
NSSpellingStateAttributeName.  Calls the delegate method  
textView:shouldSetSpellingState:range:.


or the delegate variant:

- (NSInteger)textView:(NSTextView *)textView shouldSetSpellingState: 
(NSInteger)value range:(NSRange)affectedCharRange;
// Delegate only.  Allows delegate to control the setting of  
spelling and grammar indicators.  Values are those listed for  
NSSpellingStateAttributeName.


If you need to support 10.4, perhaps you can use a custom layout  
manager which avoids setting the temporary spelling attributes for  
ranges you don't wish to display misspellings for.


Jim

___

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: Clarification on accessors? (was: Yet another memory management question)

[Jim Correia]

On Jul 9, 2009, at 5:52 PM, Charles Srstka wrote:


On Jul 9, 2009, at 4:29 PM, Ben Trumbull wrote:

It may be that under certain circumstances (dictionary storage,  
synthesized properties on 10.5, etc) that you must use an accessor  
method even in init & dealloc. That requires more care, is more  
error prone, and less flexible.  In particular, additional behavior  
should not be adorned to the accessor, and the accessor must never  
leak knowledge of the receiver to a third party during init or  
dealloc.  Basically, these accessors must be very very boring.


But if it's a synthesized accessor, wouldn't it be pretty much a  
given that that would be the case?


That will be the case for your class, with its synthesized accessors.  
If a subclass overrides those accessors, though, you'll be invoking  
them during setup and teardown too.


The overridden accessors, by definition, add side effects or behavior.

- Jim




___

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: NSLinkAttributeName: can I control the visual presentation?

[Jim Correia]

On Jul 4, 2009, at 9:18 AM, Stuart Malin wrote:

I have a mutable attributed string that is displayed in an  
NSTextView. For a certain range, I set several attributes, including  
NSLinkAttributeName. However, the presentation shows as blue  
underlined text even though I have set other attributes for the font  
color and also have set the underline style to none. Is there  
something else I must do to control the visual presentation?


On NSTextView:

- (void)setLinkTextAttributes:(NSDictionary *)attributeDictionary;
- (NSDictionary *)linkTextAttributes;


Jim

___

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: Determining OS at Runtime

[Jim Correia]

On Jul 3, 2009, at 1:06 PM, Steve Christensen wrote:

For the case of a function, if you're deploying on 10.4 but using a  
10.5 function, that function will be weak-linked so doing a runtime  
check is both faster and more accurate in determining whether a  
function exists or not.


Beware that there have been edge cases in the past where a non-NULL  
function test will give you a false positive with disastrous results.


It goes like this:

if (FunctionName != NULL) {
FunctionName(x, y, z);
}

FunctionName was SPI on 10.x but (inadvertently) exported, but API on  
10.y.


When running on 10.x, your NULL function pointer test will pass, but  
this is SPI and isn't supported for use. The signature may have  
changed, or the range of valid input values may be a subset of the  
supported ones on 10.y.


This is not a theoretical problem, but one which has happened in the  
past.


What is the take home lesson here? Test all branches of the code you  
write.


Jim
___

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: Replace -[NSKeyedUnarchiver unarchiveObjectWithData:] so it doesn't crash on corrupt archive

[Jim Correia]

On Thu, Jul 2, 2009 at 11:17 PM, Jerry Krinock  wrote:

If -[NSKeyedUnarchiver unarchiveObjectWithData:] is handed a corrupt
> archive, it raises an exception AND crashes the program.


Do you have a sample app & data combo which demonstrates the crash?

+[NSKeyedUnarchiver unarchiveObjectWithData:] is documented to raise an
exception when feed an incomprehensible archive.

If you have no exception handler in place, the program will terminate. But
that isn't due to a bug in NSKeyedUnarchiver - it is just working as
advertised - whether or not you like the advertised behavior.

- Jim
___

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: Problem with NSValueTransformer, NSTableView and NSAttributedString

[Jim Correia]

On Jun 23, 2009, at 11:06 AM, Kyle Sluder wrote:

On Tue, Jun 23, 2009 at 6:53 AM, Jim Correia  
wrote:

On Jun 23, 2009, at 1:35 AM, Kyle Sluder wrote:

I do not see where you're finding this.  The bindings documentation
says: "The object should be suitable for the passing to
setObjectValue."  Maybe it's in the conceptual docs?


In the link I included in my last email (which was wrapped when  
sent to the

list):


Okay... but you are aware that's the documentation for
NSTextFieldCell, not NSTableColumn?  Bindings in a table view are done
on the table column, not on the cells themselves.


Correct. I should have read the original question more carefully.

Binding an NSTableColumn to a model object property which returns an  
NSAttributedString works correctly. As does binding to a property  
through a value transformer which returns an NSAttributedString.


Jim
___

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: Problem with NSValueTransformer, NSTableView and NSAttributedString

[Jim Correia]

On Jun 23, 2009, at 1:35 AM, Kyle Sluder wrote:

On Mon, Jun 22, 2009 at 9:12 PM, Jim Correia  
wrote:
The documentation for the value binding of NSTableView says that  
NSString

and NSNumber are the supported value types.


I do not see where you're finding this.  The bindings documentation
says: "The object should be suitable for the passing to
setObjectValue."  Maybe it's in the conceptual docs?


In the link I included in my last email (which was wrapped when sent  
to the list):


<http://developer.apple.com/documentation/Cocoa/Reference/CocoaBindingsRef/BindingsText/NSTextFieldCell.html#//apple_ref/doc/uid/NSTextFieldCell-DontLinkElementID_657 
>


Value Bindings

value

An NSString or NSNumber that is displayed as the content of the
NSTextFieldCell. Availability:

- Jim
___

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: Problem with NSValueTransformer, NSTableView and NSAttributedString

[Jim Correia]

On Jun 21, 2009, at 10:37 AM, Donnie Lee wrote:


I created NSTableView and bind table column with NSTextFieldCell with
custom NSValueTransformer subclass to NSArrayController. My value
transformer should return NSAttributedString transformed from
NSString. Everything works fine except that the table shows NSString
instead of NSAttributedString. It looks like: "My text{NSFont =
"LucidaGrande}". How to make the right behavior for my value
transformer?


The documentation for the value binding of NSTableView says that  
NSString and NSNumber are the supported value types.


http://developer.apple.com/documentation/Cocoa/Reference/CocoaBindingsRef/BindingsText/NSTextFieldCell.html#/ 
/apple_ref/doc/uid/NSTextFieldCell-DontLinkElementID_657


Jim

___

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: Possible to set tab stops in an NSTextView?

[Jim Correia]

On Sun, Jun 21, 2009 at 11:44 AM, Ken Tozier  wrote:

>
> I'm working with an NSText view and want to display tab delimited strings
> in a table format. I don't need a full NSTable, just the ability to break up
> the text into columns using tab stops. Is there any built-in way to do this?
> Something like setTabStops?
>
> I looked in NSTextView, NSText, NSTextContainer and NSAttributedString but
> didn't see anything.


Tab stops are controlled via NSParagraphStyle.

- Jim
___

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: Add same object/key to NSMapTable

[Jim Correia]

On Wed, Jun 17, 2009 at 9:51 PM, Bill Bumgarner  wrote:

> You can't assume that the container didn't make a copy or do something else.
>
> If you need the object, you should hold a retain.

In general, I agree. However, the following code is also problematic:

[o release];
[testTable setObject:[testTable objectForKey:k] forKey:k];

NSDictionary does not have the same failure mode. Bug in NSMapTable
(for the edge case of replacing the current value)?

- Jim
___

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: orderFrontStandardAboutPanel problem on Tiger - [NSCFDictionary setObject:forKey:]: attempt to insert nil value

[Jim Correia]

On Mon, Jun 1, 2009 at 7:29 AM, Chris Jones  wrote:

> The error is: [NSCFDictionary setObject:forKey:]: attempt to insert nil
> value. And the StandardAboutPanel does not appear.

[...]

> I don't think it could get more simple, but I can't see where I'm passing a
> nil value. I thought this was related to a missing piece of info in
> info.plist, but all required fields seem to be present.

Set a breakpoint on -[NSCFDictionary setObject:forKey:] and a
condition for when the object parameter is nil. (The specific
condition is going to depend upon what architecture you are debugging
in.)

When you stop there, examine the value for the key. Hopefully at that
point the problem is obvious.

- Jim
___

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: [ NSSplitview ] -setAutosaveName

[Jim Correia]

On Sat, May 30, 2009 at 11:22 PM, Erg Consultant
 wrote:

> What does setAutosaveName actually do? I assume it saves the position of the 
> splitview to user defaults?

The behavior is documented, succinctly, here:

http://developer.apple.com/documentation/Cocoa/Reference/ApplicationKit/Classes/nssplitview_Class/Reference/Reference.html#//apple_ref/occ/instm/NSSplitView/setAutosaveName:

> If so, how does one retrieve that value?

You typically don't, since the actual key and format of the data is a
private implementation detail. NSSplitView will restore the geometry
for you automatically.

Jim
___

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: [ NS KeyedArchiver ] archiveRootObject:toFile: hoses my directory's permissions

[Jim Correia]

On Sat, May 30, 2009 at 1:39 AM, Erg Consultant 
wrote:

> Why does archiveRootObject:toFile: change the permissions on the parent
> directory? I didn't ask it to. This dir has to be writable for all users -
> the way I installed it - not by only the current user.
>
> How can I avoid having the permissions changed?

Show your code. In a trivial sample app here, this is not the behavior
I observe.

- JIm

#import 

int main (int argc, const char * argv[])
{
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];

NSString *outputFilePath =
[@"~/Desktop/output-parent/outputfile.keyedArchive"
stringByExpandingTildeInPath];
NSString *outputDir = [outputFilePath 
stringByDeletingLastPathComponent];

NSDictionary *attributes = [NSDictionary dictionaryWithObject:
[NSNumber numberWithUnsignedInt: 0777] forKey:
NSFilePosixPermissions];
BOOL success = NO;

success = [[NSFileManager defaultManager] createDirectoryAtPath:
outputDir attributes: attributes];

success = [NSKeyedArchiver archiveRootObject: @"my object" toFile:
outputFilePath];

attributes = [[NSFileManager defaultManager] attributesOfItemAtPath:
outputDir error: NULL];
NSLog(@"permissions = %o", [[attributes objectForKey:
NSFilePosixPermissions] unsignedIntValue]);

[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/archive%40mail-archive.com

This email sent to arch...@mail-archive.com

Re: NSEvent virtual keycodes

[Jim Correia]

On Mon, May 18, 2009 at 10:44 PM, Christopher Hansen
 wrote:

> The NSEvent Reference documentation for the keyCode method states that "the
> value returned is the same value returned in the kEventParamKeyCode when using
> Carbon Events."
>
> Are the virtual key codes available in a Cocoa-based header?  The only virtual
> keycode definitions I've been able to find are in Carbon/HIToolbox/Events.h.

That is the only place they are currently defined.

Including that header will not give your code cooties. That's
currently the correct place to pick up the constants from.

Jim
___

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: Which field is the NSTextFieldCell drawing?

[Jim Correia]

On Sun, May 17, 2009 at 12:08 PM, K. Darcy Otto  wrote:
> Thank you.  The thing is, the cell object returned by
> -tableView:willDisplayCell:forTableColumn:row: is not from my
> NSTextFieldCell subclass but rather the superclass – even though it is the
> subclass doing the drawing.  Any idea how to address the subclass object
> directly?

The instance passed to -tableView:willDisplayCell:forTableColumn:row:
is the same instance that is used to draw the cell.

If you are seeing an instance of NSTextFieldCell instead of an
instance of MyFancyTextFieldCell, you must have a configuration issue.

How are you setting up the table column for your cell subclass?

Jim
___

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: Which field is the NSTextFieldCell drawing?

[Jim Correia]

On Sat, May 16, 2009 at 9:05 PM, K. Darcy Otto  wrote:

> I have an NSTextFieldCell subclass that needs to draw itself differently
> depending on which NSTableView row it is drawing.  My current solution is to
> store the row that is being drawn in my document class in an instance
> variable (the assignment is done in
> -tableView:willDisplayCell:forTableColumn:row:), and have my NSTextFieldCell
> read out that variable prior to drawing itself.  Is there a more direct way
> to do this?  Is it possible to either:
>
> (1) Have the NSTextFieldCell subclass look up the row that is currently
> being drawn, or
> (2) Store the row to be drawn in an instance variable of the
> NSTextFieldCell.
>
> I think (1) would be the easiest.  What is preventing (1) is that I'm not
> sure how to query the NSTableView to ask it what object is currently being
> drawn.  What is preventing (2) is that I'm not sure how to get a pointer to
> the NSTextFieldCell subclass object that is about to be drawn, and insert
> the appropriate information just before it is drawn.

-tableView:willDisplayCell:forTableColumn:row: *is* your hook to do this.

In that delegate method, set any additional properties on your cell
that you need at display time.

- Jim
___

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: Core Data Predicate Bug?

[Jim Correia]

On Sat, May 16, 2009 at 8:13 AM, ALEXander  wrote:

> I am currently developing a Core Data Application. I have an entity (set)
> with a boolean attribute 'isActive' (optional, non-transient, non-indexed,
> default NO). For some of my objects, this attribute is set to YES. There is
> another entity (control) with a to-one relationship to a set.
>
> The problem occurs when saving and reopening a document: If I use a
> predicate in a fetch request for the entity control which contains a
> predicate format string like @"set.isActive == YES", no objects are
> returned. Before saving and reopening, the correct objects are returned.
>  How can that happen? I know (by looking at the data with the pre-built core
> data entity interfaces from IB) that there are objects where that predicate
> is true.
>
> To me, this is a bug. But of course, I hope to be wrong.

Do you have a standalone sample which demonstrates this problem? From
the description above, it is hard to say what might be going on.

Writing a standalone sample can be tedious and time consuming. But in
the process of doing so, you may discover that this is actually a bug
in your code/logic. If not, then you'll have a shareable sample that
you can show us and attach to a bug report.

Jim
___

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: -[NSScriptCommand suspendExecution]

[Jim Correia]

On Fri, May 15, 2009 at 12:26 PM, kvic...@pobox.com  wrote:

> if i call -[NSScriptCommand suspendExecution] whilst execution a script
> command, i find that the executing script will still time out after 2
> minutes.
>
> is this a bug in cocoa? or a misunderstanding on my part about
> suspendExecution?

A misunderstanding on your part about how things work.

Even if you -suspendExecution and later return a result, the timeout
that the sender specified when they sent the event - how long they
were willing to wait for either an answer or a timeout error - is
still in force.

- Jim
___

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: NSTask temrinationStatus always == 1?

[Jim Correia]

On Sun, May 10, 2009 at 2:18 PM, marc hoffman  wrote:

> i have an odd issue with NSTask. for some reason, no matter what result code
> my executable returns (im running xcodebuild, if that matters), NSTask's
> terminationStatus always reports back as 1. if instead i use system() to run
> the exact same command, i get bak proper error codes (0 for success., other
> non-zero, non-one values for real failures.

You'll have to show your code. NSTask correctly returns the error code
(see below).

NSTask *task = nil;

task = [NSTask launchedTaskWithLaunchPath: @"/usr/bin/true"
arguments: [NSArray array]];
[task waitUntilExit];
NSLog(@"terminationStatus = %d", [task terminationStatus]);

task = [NSTask launchedTaskWithLaunchPath: @"/usr/bin/false"
arguments: [NSArray array]];
[task waitUntilExit];
NSLog(@"terminationStatus = %d", [task terminationStatus]);

- Jim
___

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: Make an attributed string all uppercase without losing attributes?

[Jim Correia]

On Fri, May 8, 2009 at 10:56 PM, Graham Cox  wrote:
> I have an attributed string. I want to preserve the attributes within the
> string but set all the actual characters to uppercase, lowercase, etc.
>
> According to the docs for NSMutableAttributedString, the -mutableString
> method returns a mutable string that the receiver tracks changes to in order
> to preserve attribute runs. However, when I call -setString: or
> -replaceCharactersInRange:withString: on this, all my attributes are
> removed.
>
> How can I do this without my attributes going AWOL?

The documentation for -[NSMutableString
replaceCharactersInRange:withString:] is perhaps more enlightening.

"""The new characters inherit the attributes of the first replaced
character from aRange"""

Processing the string by attribute runs sounds like it should have the
effect you are looking for.

Jim
___

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: NSTask temrinationStatus always == 1?

[Jim Correia]

On Sun, May 10, 2009 at 2:18 PM, marc hoffman  wrote:

> i have an odd issue with NSTask. for some reason, no matter what result code
> my executable returns (im running xcodebuild, if that matters), NSTask's
> terminationStatus always reports back as 1. if instead i use system() to run
> the exact same command, i get bak proper error codes (0 for success., other
> non-zero, non-one values for real failures.
>
> anyone have any ideas what could be going wrong to cause this?

You'll have to show your code. NSTask correctly returns the exit code
(see below).

   NSTask *task = nil;

   task = [NSTask launchedTaskWithLaunchPath: @"/usr/bin/true"
arguments: [NSArray array]];
   [task waitUntilExit];
   NSLog(@"terminationStatus = %d", [task terminationStatus]);

   task = [NSTask launchedTaskWithLaunchPath: @"/usr/bin/false"
arguments: [NSArray array]];
   [task waitUntilExit];
   NSLog(@"terminationStatus = %d", [task terminationStatus]);

- Jim
___

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: NSManagedObjectContext -insertObject: Cancels Prior Deletion -- BUT!

[Jim Correia]

On Wed, May 6, 2009 at 1:05 PM, Jerry Krinock  wrote:

>> In any case using multiple MOCs for different jobs is a good idea.
>
> Accessing the same store?  I always thought that was a bad idea, but I'll
> give it some thought.

Using multiple MOCs with a single persistent store coordinator is
supported. It is pattern #1 in the Core Data threading documentation.

http://developer.apple.com/documentation/Cocoa/Conceptual/CoreData/Articles/cdMultiThreading.html#//apple_ref/doc/uid/TP40003385-SW2

Jim
___

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: WTF? Is a file corrupted?

[Jim Correia]

On May 5, 2009, at 9:40 AM, Scott Ribe wrote:

Application that's been running a long time, now one user out of  
dozens gets
this crash first thing after logging in (and, obviously, going to  
select a

menu item to get started):

Thread 0 Crashed:
0   com.apple.QD   0x91757360 SetOrigin + 136
1   com.apple.HIToolbox0x932d04a8 HLTBSetPortOrigin  
+ 48

2   com.apple.HIToolbox0x9335c180
GetMenuWindow(OpaqueMenuRef*, unsigned short, Rect const*,  
OpaqueRgnHandle*)

+ 896


I see that crash with some frequency in our automated crash reporter.  
However, all of machines appear to be running 10.4.x.


What version of the OS are you on?

Jim

___

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: -[CALayer hitTest:] incorrect when embedding layer-backed NSView subclass inside NSScrollView

[Jim Correia]

On Mon, May 4, 2009 at 6:19 PM, Steven Degutis  wrote:

> Specifically, you'll notice that the -hitTest: returns the correct
> layer *only* if you click the view while it is fully visible in the
> window, and before you have made it not-fully-visible for the first
> time. After you resize the scroll view to hide any portion of this
> NSView subclass, -hitTest: will consistently return the wrong layer
> every time.

[...]

> - (void) mouseDown:(NSEvent*)event {
>        NSPoint point = [self convertPoint:[event locationInWindow] 
> fromView:nil];
>        CALayer *layer = [self.layer hitTest:NSPointToCGPoint(point)];
>        NSLog(@"%@", [layer name]);
> }
>
> I've tried to replace `self` with the enclosing scroll view, or its
> content view or document view, in the -mouseDown: method, and many
> combinations of this. Nothing seems to do the trick.

-hitTest on CALayer, like NSView, takes a point in the superlayer's coordinate
system.

Jim
___

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: Modifying NSTableView cell data just prior to invoking field editor

[Jim Correia]

On Thu, Apr 30, 2009 at 11:38 PM, K. Darcy Otto  wrote:

> Option 2: Moving the text displayed by the NSTableView to the right by some
> way other than inserting spaces.  This might be the best way, alleviating
> the need for a custom field editor and editing the field editor text prior
> to display.  I'm not really sure how to do this though.

So you are inserted spaces into the value to achieve an fixed width indent?

Subclass NSTextFieldCell, and override -titleRectForBounds: to add
your left padding.

- Jim
___

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 NSString->FSRef so hard?

[Jim Correia]

On Thu, Apr 30, 2009 at 12:02 AM, Charles Srstka
 wrote:
> With that said, I don't really see what is harmful about recommending the
> use of NSString's path concatenation methods, which could save quite a few
> headaches in the case that this *does* somehow change for some bizarre
> reason, however unlikely that may be.

Even if the path separator never changes, they are still beneficial to
use because they automatically handle both the trailing slash present
and trailing slash missing cases transparently

[@"/foo" stringByAppendingPathComponent: @"bar"] -> /foo/bar
[@"/foo/" stringByAppendingPathComponent: @"bar"] -> /foo/bar

- Jim
___

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: Key Value Coding is not case-sensitive?

[Jim Correia]

On Tue, Apr 28, 2009 at 12:09 AM, DairyKnight  wrote:

>   And have the following set methods:
>
>   -(void) setfido ...
>   -(void) setFido ...
>
>   Both by calling [self setValue: ... forKey:@"fido"] or [self setValue:...
> forKey:@"Fido"], the runtime would call the same 'setFido' method. How could
> this be?

-setfido is not a KVC compliant method name.

Fido is not a KVC compliant iVar name. (Though generally you should be
using KVC via accessor methods and not via direct iVar access anyway.)

http://developer.apple.com/documentation/Cocoa/Conceptual/KeyValueCoding/Concepts/Compliant.html

- Jim
___

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: figuring out which TableView I am?

[Jim Correia]

On Sun, Apr 26, 2009 at 8:04 PM, Graham Cox  wrote:

> This is true, but your example of non-scalability does reflect a poor
> approach that you wouldn't be likely to follow in practice.
>
> A tag is 32 bits, so can represent 2^32 different states, if used wisely.
> Dividing this into bitfields would allow you to partition it into several
> properties, for example four different colour properties, each with 256
> possible colours each.

Even when used as a bitfield, I'd still argue it is not good design. (It might
be an expedient hack to get something done, and if you are happy with that
as a solution, great.)

It overloads the tag to imply control behavior/appearance, while at the same
time preventing users of your class from using the tag for its original purpose.
(Yes, I realize that the the user of the custom table view is the same person
as the provider of the custom table view.)

- Jim
___

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: figuring out which TableView I am?

[Jim Correia]

On Sun, Apr 26, 2009 at 7:05 PM, WT  wrote:

> More importantly, perhaps, is the fact that adding property/ivars is not a
> scalable solution in that it requires adding more code in several places if
> more tables are added. The tags solution only requires some fiddling with IB
> and updating a switch statement. That sounds to me a much simpler
> maintenance problem than having to update code in several places.

I guess I'm not following your logic.

I want table 1 to have a blue background, so I set its tag to 3.
I want table 2 to also have a blue background, so I set its tag to 3.

I want table 1 to have red text. Switch off tag, 3 == red text.
I want table 2 to have green text. Oops - tag 3 already means blue
background + red text.

(Substitute any 2 actual properties for my background color + text
color - they are meant for illustration purposes only.)

Overloading tags for other properties is not scalable, because you
only have one property (the tag) which you are overloading to mean N
other properties and/or behaviors.

The only scalable solution to this problem is explicity
iVars/properties for each property/behavior you want to control in
your subclass. (Yes, this means configuring things by hand after
unarchiving your nib, or writing an IB plug-in so you can specify them
in IB.)

- Jim
___

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: NSPopUpButtonCell Keeps on Trackin'! Demo, Movie

[Jim Correia]

On Sun, Apr 26, 2009 at 10:41 AM, Jerry Krinock  wrote:

> // Table View delegate method.
> // A little more complicated since we have to dig into the
> // parts of the table, and assign it as a header cell.
> - (void)               tableView:(NSTableView*)tableView
>  mouseDownInHeaderOfTableColumn:(NSTableColumn *)tableColumn {
>    NSTableHeaderView *headerView = [tableView headerView] ;
>
>    NSPopUpButtonCell* popUpCell = [[NSPopUpButtonCell alloc] init] ;
>    [tableColumn setHeaderCell:popUpCell] ;
>
>    [popUpCell setMenu:[self makeMenu]] ;
>    [popUpCell performClickWithFrame:NSZeroRect
>                          inView:headerView] ;
>    [popUpCell release] ;
> }

What you've done is replaced the table column's header cell, and
started a menu tracking session while the table view was already in
the middle of a mouse tracking session.

What are you really trying to accomplish here?

It is hard to offer advice without knowing what you want to build. But
you are probably going to want to take the approach of using a custom
header view (along with custom header cells as necessary.)

- Jim
___

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