Capture video from core animation (QC)
Hi, I've made a funny cuartz composition with a couple of video filters, webcam, and so on, and now i'd like to record it to a video file (like quicktime). Question is, do i need to use OpenGL and render each frame or whatever, or is there an easier way to accomplish this? Thanks ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
non scrolling table view header...
Hi, I'm trying to obtain a table header that doesn't disappear when scrolling rows. tableHeader/Footer properties support custom views that scroll with rows and so disappear. Please, anyone knows how to do this? Thanks in advance -- Oscar A. Alvarado ___ 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
KVC error from bound SortDescriptor
Hi folks,
I have a customer WindowController class and within it I have a method
that returns a NSSortDescriptor. Like this;
-(NSArray *)nameSortDescriptors
{
NSSortDescriptor *sorter;
sorter = [[[NSSortDescriptor alloc]
initWithKey: @name
ascending: YES] autorelease];
return ([NSArray arrayWithObject: sorter]);
}
In my NIB I have an NSArrayController serving a Core Data entity
called Account. I've bound this entity to the SortDescriptor above
and the data is displayed in a bound NSTableview.
I'm getting the following error whenever I click the column header to
re-order the tableView and I can't understand why the array controller
is trying to set the sortdescriptor method or what to do to surpress it.
Error setting value for key path nameSortDescriptors of object
MLAccountController: 0x58ae390 (from bound object
NSArrayController: 0x589d1a0[entity: Account, number of selected
objects: 1](null)): [MLAccountController 0x58ae390
setValue:forUndefinedKey:]: this class is not key value coding-
compliant for the key nameSortDescriptors.
___
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
Service request fails if provider has not yet launched
My app provides a System Service, which works well if it is running when the request arrives; however, if the request rises and my app has not yet launched, the Service is failed to be served, even though my app can be launched normally by the Service request. Here is the corresponding Console log message: PM Safari[368] Application XXX never opened its Services port before the timeout. But I'm sure that my app has been ready before the timeout which is the default 30s, since I register the Service provider in my applicationDidFinishLaunching:(NSNotification *) after all the preparation work is done, and the time needed to get to that point is no more than 15s. I can't figure out what is missed after perusing the System Service section of Dev Doc. Any hint will be appreciated, and any further info will be supplemented if needed. Thanks in advance. ___ 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: Service request fails if provider has not yet launched
On Jan 14, 2009, at 8:56 AM, an0...@gmail.com wrote: My app provides a System Service, which works well if it is running when the request arrives; however, if the request rises and my app has not yet launched, the Service is failed to be served, even though my app can be launched normally by the Service request. Here is the corresponding Console log message: PM Safari[368] Application XXX never opened its Services port before the timeout. But I'm sure that my app has been ready before the timeout which is the default 30s, since I register the Service provider in my applicationDidFinishLaunching: (NSNotification *) after all the preparation work is done, and the time needed to get to that point is no more than 15s. I can't figure out what is missed after perusing the System Service section of Dev Doc. Any hint will be appreciated, and any further info will be supplemented if needed. Without seeing some code, I can only offer a couple general ideas: 1. Are you sure that applicationDidFinishLaunching:(NSNotification *) is being called? If you've got your app delegate wired up wrong, then you'd never receive this notification. 2. Are you passing the right object to -[NSApplication setServicesProvider:]? 3. Do you implement the proper service method? A typo could be preventing the service from being found. HTH -- Kevin Kevin Gessner http://kevingessner.com ke...@kevingessner.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: File association
I saved the project using NSArchiver with an extension say .myExtension. And also I opened the saved project by unachieved it from menu. Now I need to open the saved project when double clicking on it. How can I do this file association? See the Runtime Configuration Guidelines: http://developer.apple.com/documentation/MacOSX/Conceptual/BPRuntimeConfig/Articles/PListKeys.html#//apple_ref/doc/uid/20001431-101685 Where I need to write code for unarchiving its content ? You should read about the Cocoa document architecture: http://developer.apple.com/DOCUMENTATION/Cocoa/Conceptual/Documents/Documents.html -Ben ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: How to center a custom NSView in a NSScrollView
I've already tried with a custom NSClipView, as explained here: http://www.bergdesign.com/missing_cocoa_docs/nsclipview.html This is the standard way to do this. The question is, how can I center a custom NSView in a NSScrollView, if the visible width and/or height is greater than the fixed size of the NSScrollView? What does centering mean if the view’s size is larger than that of the scroll view? -Ben ___ 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: Constructing class names dynamically
The output is the same, but there is also: #import objc/runtime.h objc_getClass([myString UTF8String]); I am going to guess that NSClassFromString(myString) is probably using objc_getClass(), maybe not. I didn't know NSClassFromString existed though, so I am switching to that instead of using the runtime.h function. On Tue, Jan 13, 2009 at 8:45 PM, Mohan Parthasarathy surut...@gmail.com wrote: On Tue, Jan 13, 2009 at 8:06 PM, Ken Thomases k...@codeweavers.com wrote: On Jan 13, 2009, at 3:33 PM, Jean-Daniel Dupas wrote: he is just talking about class name, not class. NSClassFromString() is probably what you're looking for. NSMutableString *clsName = derive class name from the entry. Class cls = NSClassFromString(clsName); idYouProtocol instance = [[cls alloc] init]; I'll just add the following: If you don't need such complete flexibility -- for example, if you're selecting from a fixed set of classes by some tag -- then you don't need to compute a class name and look up the class that way. Classes are objects and so they can be stored in collections. For example, you could have a lookup dictionary that mapped from keys to class objects. You would construct the dictionary like this: [NSDictionary dictionaryWithObjectsAndKeys: [SomeClass class], key1, [OtherClass class], key2, [ThirdClass class], key3, // ... etc. nil]; This adds a small amount of safety in the same way that statically-specified stuff generally does. For example, the compiler will catch typos in class names. Thanks for this suggestion. This looks fairly clean except that the space is allocated at the beginning. This may not be a big deal in some cases. In the other way, you allocate space and insert in the dictionary only when needed. Also, eventually i need the real object instances inserted in the dictionary. -mohan -mohan Cheers, Ken ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/aventurella%40gmail.com This email sent to aventure...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: How font size managed without affecting the font? Please help
I use this code to change the fontsize: NSFont *font1 = [NSFont fontWithName:[pop1 titleOfSelectedItem] size:[[pop2 titleOfSelectedItem] doubleValue]]; In the second popup’s action method, you should be invoking -[NSFontManager convertFont:toSize:] to convert the fonts to the right size. (And, in the first popup, -[NSFontManager convertFont:toFace:].) http://developer.apple.com/DOCUMENTATION/Cocoa/Reference/ApplicationKit/Classes/NSFontManager_Class/Reference/Reference.html#//apple_ref/occ/instm/NSFontManager/convertFont:toSize: -Ben ___ 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: Seamlessly converting any type of document to PDF?
If you can direct the application via AppleScript to print the document to the default printer, then you can set up a printer that prints to a PDF file, and switch the printer programmatically. (Combination of lpadmin defaults write commands.) -- Scott Ribe scott_r...@killerbytes.com http://www.killerbytes.com/ (303) 722-0567 voice ___ 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: Capture video from core animation (QC)
Okay, so if I understood correctly, i need to set up an OpenGL context (in order to capture frames) which I use with a QCRenderer, and then render each frame manually? Am I on the right track? 14 jan 2009 kl. 16.25 skrev douglas welton: If you just want to record your composition, try something like QuartzCrystal by Kineme. If you want to add recording functionality to your application take a look at the QCTV sample code. On Jan 14, 2009, at 5:38 AM, Jonathan Selander wrote: I've made a funny cuartz composition with a couple of video filters, webcam, and so on, and now i'd like to record it to a video file (like quicktime). Question is, do i need to use OpenGL and render each frame or whatever, or is there an easier way to accomplish this? ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Relaunch App (Quit, Relaunch If Quit)?
thanks! your way seems much easier :)
On Wed, Jan 14, 2009 at 10:36 AM, Jean-Daniel Dupas
devli...@shadowlab.org wrote:
The proper way to do it is probably to listen for Finder death using a
timeout, not polling the process list.
You can either use the NSWorkspaceDidTerminateApplicationNotification or use
the low level kqueue function which provide such facility.
Le 14 janv. 09 à 16:25, Chunk 1978 a écrit :
my goal is to quit an app, check to see if it has quit, and then
relaunch the app if it has quit successfully. i'm sending
quit/relaunch commands with applescripts (tell application Finder |
quit | end tell, tell application Finder | launch | end tell), and
i'm attempting to make this code work without using any type of delay
mechanism but i'm lost.
the following code works very well on quitting and relaunching the
Finder, except if the Finder is busy (copying a file, or something),
the program will continuously try to quit the finder while the finder
prompts that it's busy - it's a continuous loop... a bad scene...
essentially i need to somehow figure out how to have this loop quit
the finder, but stop the loop if it's busy...
-=-=-=-
- (IBAction)restartFinder:(id)sender
{
//Quit Finder
NSDictionary* errorDict = nil;
NSAppleEventDescriptor *returnDescriptor = nil;
returnDescriptor = [quitFinderScriptObject
executeAndReturnError:errorDict];
//Check To See If Finder Has Quit
NSWorkspace *ws = [NSWorkspace sharedWorkspace];
NSArray *runningAppDictionaries = [ws launchedApplications];
NSDictionary *aDictionary;
for (aDictionary in runningAppDictionaries)
{
//If Finder Hasn't Quit Then Return
if ([[aDictionary
valueForKey:@NSApplicationBundleIdentifier]
isEqualToString:@com.apple.finder])
{
[self restartFinder:nil];
return;
break;
}
}
//If Finder Has Quit Then Continue To Relaunch Finder
NSDictionary* errorDict = nil;
NSAppleEventDescriptor *returnDescriptor = nil;
returnDescriptor = [startFinderScriptObject
executeAndReturnError:errorDict];
}
-=-=-=-
___
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/devlists%40shadowlab.org
This email sent to devli...@shadowlab.org
___
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: KVC error from bound SortDescriptor
On Jan 14, 2009, at 4:54 AM, Steven Hamilton wrote:
I have a customer WindowController class and within it I have a
method that returns a NSSortDescriptor. Like this;
-(NSArray *)nameSortDescriptors
{
NSSortDescriptor *sorter;
sorter = [[[NSSortDescriptor alloc]
initWithKey: @name
ascending: YES] autorelease];
return ([NSArray arrayWithObject: sorter]);
}
In my NIB I have an NSArrayController serving a Core Data entity
called Account. I've bound this entity to the SortDescriptor above
and the data is displayed in a bound NSTableview.
I'm getting the following error whenever I click the column header
to re-order the tableView and I can't understand why the array
controller is trying to set the sortdescriptor method or what to do
to surpress it.
Error setting value for key path nameSortDescriptors of object
MLAccountController: 0x58ae390 (from bound object
NSArrayController: 0x589d1a0[entity: Account, number of selected
objects: 1](null)): [MLAccountController 0x58ae390
setValue:forUndefinedKey:]: this class is not key value coding-
compliant for the key nameSortDescriptors.
The sortDescriptors binding is just like any other binding. Just like
when binding to a text field, when you edit the text field, the
binding causes your model's bound value to change as well. So, when
you try to reorder the tableview, you are altering the sort
descriptors, so the binding tries to change your model's bound
property. You have two choice: 1) don't allow the table to be re-
sorted; or 2) make nameSortDescriptors key value coding compliant.
HTH,
Keary Suska
Esoteritech, Inc.
Demystifying technology for your home or business
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: UTIs and type codes and extensions
On Jan 14, 2009, at 7:48 AM, Sean McBride wrote: On 1/14/09 2:31 AM, Julien Jalon said: I'm just making sure I'm not missing something... There's no system-defined way to go from a UTI to an old-style file type code (which the UTI docs seem to refer to as a tag) and back again, am I correct? extern CFStringRef UTTypeCopyPreferredTagWithClass( CFStringRef inUTI, CFStringRef inTagClass) AVAILABLE_MAC_OS_X_VERSION_10_3_AND_LATER; Note that none of the 'UT...' functions are documented in Xcode's built- in viewer, but browsing through the .h's reveals good header documentation. Whew. So I'm not losing it. (Well, not in this particular case, anyway...) Thanks! randy ___ 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: Strange Webview Problem
On Tue, Jan 13, 2009 at 6:11 PM, Diez B. Roggisch de...@web.de wrote: Nick Zitzmann schrieb: On Jan 13, 2009, at 3:45 PM, Diez B. Roggisch wrote: Just for the record - the timoutInterval of 0 is the culprit. I don't have the slightest idea why it stopped working, but I don't care either. It might be worth mentioning in the docs though that passing 0 actually terminates the request immediatly, instead of setting the timeout to unlimited. That makes sense, actually. If you want the timeout interval to be unlimited, then you should pass in DBL_MAX, which will cause it to time out ~300 million years from now, by which point I assume your computer will no longer be around (or at least you won't be using it)... 0 usually means do it now when it comes to time intervals. But it doesn't mean do it now, it means don't do it at all... Well no, it means do it unless it takes longer than the time I specify, in which case give up. Of course if you specify 0 time, it has to give up instantaneously, unless it can somehow do the work instantaneously. Practically speaking, a 0 timeout might (might!) mean to return cached data but not attempt to get fresh data. Mike ___ 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: runModalForWindow, best solution to modal session
On Jan 13, 2009, at 12:21 PM, Alexander Reichstadt wrote: If so, do you override -initWithWindowNibName:? The implementation may matter. No, I only override initWithWindow. Should I override it, and if so should I call on NSBundle to load my nib there then? I recommend that you allow NSWindoController to manage the nib, which it will if you override -initWithWindowNibName: instead. That alone may very well solve your problem. Note that you will *not* manually load the nib in this override. NSWindoController will do that automatically, although lazily (that's why you call -window, which forces the nib to load). HTH, Keary Suska Esoteritech, Inc. Demystifying technology for your home or business ___ 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
NSTableView
I thought I had this solved but I need help. Looking at iTunes with 130 songs the table scrolls quick and resizing the window is snappy. When I resize a window or split view containing a table of 30 rows or scroll the table it is slow. I have read Cocoa Performance Guidelines. I cannot get a call to preservesContentDuringLiveResize for my table view. So, how can I make my users happy with a snappy table view during resize operations? David Blanton ___ 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: NSTableView
On Jan 14, 2009, at 9:14 AM, David Blanton wrote: I thought I had this solved but I need help. Looking at iTunes with 130 songs the table scrolls quick and resizing the window is snappy. When I resize a window or split view containing a table of 30 rows or scroll the table it is slow. I have read Cocoa Performance Guidelines. I cannot get a call to preservesContentDuringLiveResize for my table view. So, how can I make my users happy with a snappy table view during resize operations? For any performance problem, I highly recommend finding out what is slow. Use Shark to sample your program and find out what it is. Then post the relevant information to the list saying why is X slow? There isn't much that people can recommend until then. corbin ___ 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: Enable Scripting Failing - Missing something? Based on SimpleScripingVerbs sample
On Tue, 13 Jan 2009 13:17:11 -0800, Adam Venturella aventure...@gmail.com said: keyOSAScriptingDefinition/key stringMyScriptingDef.def/string (1) Why do you call it .def when the example so clearly calls it .sdef? (2) There are lots of tutorials (including mine online, plus my book) on getting started with scriptability. m. -- matt neuburg, phd = m...@tidbits.com, http://www.tidbits.com/matt/ A fool + a tool + an autorelease pool = cool! One of the 2007 MacTech Top 25: http://tinyurl.com/2rh4pf AppleScript: the Definitive Guide - Second Edition! http://www.amazon.com/gp/product/0596102119 ___ 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: Enable Scripting Failing - Missing something? Based on SimpleScripingVerbs sample
the .def was a typo on my part when writing the e-mail. Anyway I managed to get it working, I must have been doing something wrong, as I built 2 test apps for the sole purpose of testing scripting and everything was fine. So the problem was me =) On Wed, Jan 14, 2009 at 9:24 AM, Matt Neuburg m...@tidbits.com wrote: On Tue, 13 Jan 2009 13:17:11 -0800, Adam Venturella aventure...@gmail.com said: keyOSAScriptingDefinition/key stringMyScriptingDef.def/string (1) Why do you call it .def when the example so clearly calls it .sdef? (2) There are lots of tutorials (including mine online, plus my book) on getting started with scriptability. m. -- matt neuburg, phd = m...@tidbits.com, http://www.tidbits.com/matt/ A fool + a tool + an autorelease pool = cool! One of the 2007 MacTech Top 25: http://tinyurl.com/2rh4pf AppleScript: the Definitive Guide - Second Edition! http://www.amazon.com/gp/product/0596102119 ___ 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: NSTableView
I understand what you are saying but it seems to me that if I could preserve content during live resize all would be well. My table view subclass implements preservesContentDuringLiveResize but it is not called. Is preservesContentDuringLiveResize a valid method for a table view subclass? On Jan 14, 2009, at 10:21 AM, Corbin Dunn wrote: On Jan 14, 2009, at 9:14 AM, David Blanton wrote: I thought I had this solved but I need help. Looking at iTunes with 130 songs the table scrolls quick and resizing the window is snappy. When I resize a window or split view containing a table of 30 rows or scroll the table it is slow. I have read Cocoa Performance Guidelines. I cannot get a call to preservesContentDuringLiveResize for my table view. So, how can I make my users happy with a snappy table view during resize operations? For any performance problem, I highly recommend finding out what is slow. Use Shark to sample your program and find out what it is. Then post the relevant information to the list saying why is X slow? There isn't much that people can recommend until then. corbin David Blanton ___ 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: NSTableView
On Jan 14, 2009, at 9:41 AM, David Blanton wrote: I understand what you are saying but it seems to me that if I could preserve content during live resize all would be well. My table view subclass implements preservesContentDuringLiveResize but it is not called. Is preservesContentDuringLiveResize a valid method for a table view subclass? Yes, it is a valid method, but there are many caveats to having it work with NSTableView, due to the way things autoresize. As I said before, you should really Shark/Sample your program first. corbin On Jan 14, 2009, at 10:21 AM, Corbin Dunn wrote: On Jan 14, 2009, at 9:14 AM, David Blanton wrote: I thought I had this solved but I need help. Looking at iTunes with 130 songs the table scrolls quick and resizing the window is snappy. When I resize a window or split view containing a table of 30 rows or scroll the table it is slow. I have read Cocoa Performance Guidelines. I cannot get a call to preservesContentDuringLiveResize for my table view. So, how can I make my users happy with a snappy table view during resize operations? For any performance problem, I highly recommend finding out what is slow. Use Shark to sample your program and find out what it is. Then post the relevant information to the list saying why is X slow? There isn't much that people can recommend until then. ___ 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: NSTableView
On Jan 14, 2009, at 12:14 PM, David Blanton wrote: I thought I had this solved but I need help. Looking at iTunes with 130 songs the table scrolls quick and resizing the window is snappy. When I resize a window or split view containing a table of 30 rows or scroll the table it is slow. I have read Cocoa Performance Guidelines. I cannot get a call to preservesContentDuringLiveResize for my table view. What do you mean cannot get a call? In any case, if you get the problem while scrolling then I would guess resizing is not the problem. And it's adding complexity prematurely if you don't know where the actual time is being spent. I forget if you said whether you're using bindings or a data source. If you're using a data source, how about posting your code for numberOfRowsInTableView: and tableView:objectValueForTableColumn:row:? The docs for these two methods say they need to be very efficient. If you're using bindings, maybe you have a model class with a slow getter method. If nothing leaps out from a look at these areas, then you really should profile as Corbin suggested rather than assume you know what the problem is. --Andy So, how can I make my users happy with a snappy table view during resize operations? David Blanton ___ 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/aglee%40mac.com This email sent to ag...@mac.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: KVC error from bound SortDescriptor
On Jan 14, 2009, at 08:35, Keary Suska wrote:
On Jan 14, 2009, at 4:54 AM, Steven Hamilton wrote:
I have a customer WindowController class and within it I have a
method that returns a NSSortDescriptor. Like this;
-(NSArray *)nameSortDescriptors
{
NSSortDescriptor *sorter;
sorter = [[[NSSortDescriptor alloc]
initWithKey: @name
ascending: YES] autorelease];
return ([NSArray arrayWithObject: sorter]);
}
...
Error setting value for key path nameSortDescriptors of object
MLAccountController: 0x58ae390 (from bound object
NSArrayController: 0x589d1a0[entity: Account, number of selected
objects: 1](null)): [MLAccountController 0x58ae390
setValue:forUndefinedKey:]: this class is not key value coding-
compliant for the key nameSortDescriptors.
... 2) make nameSortDescriptors key value coding compliant.
Jumping in with a quibble ...
The OP's window controller class *is* KVC-compliant for the
'nameSortDescriptors' property, but it's an immutable property. The
sortable column needs it to be a mutable property. The error message
is slightly at fault here.
So the solution is to put the array of sort descriptors in an instance
variable and write a setter for the property, which is, I think, what
you were implying the OP should do.
___
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: NSTableView
On Jan 14, 2009, at 1:06 PM, Andy Lee wrote: I forget if you said whether you're using bindings or a data source. If you're using a data source, how about posting your code for numberOfRowsInTableView: and tableView:objectValueForTableColumn:row:? The docs for these two methods say they need to be very efficient. I found your earlier post. It sounds like you already know tableView:objectValueForTableColumn:row: is slow. *Why* is it slow? Is it making some kind of network or database call? If your table only has 30 rows or so, why not cache the objects in an NSArray? Again, posting code might help. --Andy ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Debugging objc_msg errors
I occasionally get difficult-to-debug crashes where that last thing in the crash log is a message being sent, which ends up being a message sent to an object that has been released (I think it is objc_msg, but I am not looking at a crash log now). These happen much more often in Leopard than in Tiger. Since I have many messages going to many objects, it can take me a long time to track these problems down (in one case several weeks). I am fairly certain it is not a notification message because I also remove objects as observers before they are deallocated. Is there a debugging tool that can provide more information about what message was sent to what object at the time of the crash? --- John Nairn GEDitCOM - Genealogy Software for the Macintosh http://www.geditcom.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: Debugging objc_msg errors
John, See the following for a good tutorial: http://www.sealiesoftware.com/blog/archive/2008/09/22/objc_explain_So_you_crashed_in_objc_msgSend.html -Jeff On Jan 14, 2009, at 1:04 PM, John Nairn wrote: I occasionally get difficult-to-debug crashes where that last thing in the crash log is a message being sent, which ends up being a message sent to an object that has been released (I think it is objc_msg, but I am not looking at a crash log now). These happen much more often in Leopard than in Tiger. Since I have many messages going to many objects, it can take me a long time to track these problems down (in one case several weeks). I am fairly certain it is not a notification message because I also remove objects as observers before they are deallocated. Is there a debugging tool that can provide more information about what message was sent to what object at the time of the crash? --- John Nairn GEDitCOM - Genealogy Software for the Macintosh http://www.geditcom.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: NSTableView
Here is the code. Some notes,
_browserView is an NSMatrix of NSImageCells
Str is form a portable string library
there are a few C++ classes used
Once a cell value is computed it is stored.
When moving a split view divider and covering the table that action
is jerky. I thought storing the cell values would improve that but no.
Any and all comments please.
Oh, Shark said 16% objc-msgsend.
Thanks in advance!
/* numberOfRowsInTableView */
- (int)numberOfRowsInTableView:(NSTableView *)tableView {
if(_browserView != nil)
{
int count = _browserView-fileList.GetCount();
[_browserView selectAll:self];
[_browserCells release];
_browserCells = [_browserView selectedCells];
[_browserCells retain];
return count;
}
else return 0;
}
/* tableViewobjectValueForTableColumn:row */
- (id)tableView:(NSTableView *)aTableView objectValueForTableColumn:
(NSTableColumn *)aTableColumn row:(int)rowIndex {
DTVCell* cell = [aTableColumn dataCell];
NSString* colID = [aTableColumn identifier];
[self setupCell:cell forRow:rowIndex column:colID];
return cell;
}
/* setupCell */
- (void)setupCell:(DTVCell*)cell forRow:(int)rowIndex column:
(NSString*)colID {
BMatrixCell *xcell = [_browserCells objectAtIndex:rowIndex];
BEUtil beu;
if(xcell-_haveDesign == NO)
{
BEDesign design;
xcell-m_Design.SetFilename(xcell-mBFilename);
beu.GetFile(xcell-m_Design);
xcell-m_StitchCount = xcell-m_Design.StitchCount();
BEColorList becl;
xcell-m_Design.GetColorList(becl);
xcell-m_ColorCount = becl.GetCount();
if(xcell-m_ColorCount0) xcell-m_StitchCount -= (xcell-
m_ColorCount-1);
xcell-_haveDesign = YES;
}
if([colID compare:@NAME] == NSOrderedSame)
{
if(xcell-_name == nil)
{
Str s((const
STRCHAR*)xcell-mBFilename.StripExtension());
xcell-_name = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_name retain];
}
cell-_value = xcell-_name;
return;
}
if([colID compare:@EXTENSION] == NSOrderedSame)
{
if(xcell-_extension == nil)
{
Str s((const STRCHAR*)xcell-mBFilename.GetExtension());
xcell-_extension = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_extension retain];
}
cell-_value = xcell-_extension;
return;
}
if([colID compare:@COLORS] == NSOrderedSame)
{
if(xcell-_colors == nil)
{
xcell-_colors = [NSString stringWithFormat:@%d,xcell-
m_ColorCount];
[xcell-_colors retain];
}
cell-_value = xcell-_colors;
return;
}
if([colID compare:@STITCHES] == NSOrderedSame)
{
if(xcell-_stitches == nil)
{
xcell-_stitches = [NSString stringWithFormat:@%d,xcell-
m_StitchCount];
[xcell-_stitches retain];
}
cell-_value = xcell-_stitches;
return;
}
if([colID compare:@JUMPS] == NSOrderedSame)
{
if(xcell-m_jumps == 0)
{
BEObj *obj;
POSITION pos = xcell-m_Design.GetHeadPosition();
while(pos)
{
obj = xcell-m_Design.GetNext(pos);
obj-SetStitchCounts();
int jmp;
obj-m_data.GetData(CIMJUMPS,jmp);
xcell-m_jumps += jmp;
}
xcell-_jumps = [NSString
stringWithFormat:@%d,xcell-m_jumps];
[xcell-_jumps retain];
}
cell-_value = xcell-_jumps;
return;
}
if([colID compare:@SIZE] == NSOrderedSame)
{
if(xcell-_size == nil)
{
Str measure((const STRCHAR *)beu.GetMeasureString(1,xcell-
m_Design));
NSString *nsMeasure = [NSString stringWithCString:(const STRCHAR*)
measure encoding:NSASCIIStringEncoding];
NSRange range = [nsMeasure rangeOfString:@:];
NSString *sz = [nsMeasure
auto garbage collection race condition?
I'm running into a scenario where it appears the application is allocating memory faster than it can be reclaimed by garbage collection. I believe this is how it happens. There are a pool of threads that are used to collect some data as it is collected (by using NSXML* objects, XQuery). In the beginning the generational collections seem to keep up well since almost all of the NSXML* objects become garbage very quickly. After a small bit of time though no garbage collection takes place and memory is consumed to the point of exhaustion. (I have AUTO_LOG_COLLECTIONS=YES set to see when collections take place). Some analysis during the period that no collections are taking place actually reveals that the collector thread is at this point: #0 0x001d3a37 in Auto::Bitmap::bit at AutoBitmap.h:131 #1 0x001dca84 in Auto::Region::is_pending at AutoRegion.h:209 #2 0x001db587 in Auto::Admin::is_pending at AutoAdmin.cpp:415 #3 0x001ee3cd in Auto::Subzone::is_pending at AutoSubzone.h:472 #4 0x001eec1e in Auto::scan_pending_blocks_visitor::visit at AutoMemoryScanner.cpp:631 #5 0x001eed55 in Auto::visitAllocatedBlocksAuto::scan_pending_blocks_visitor at AutoBlockIterator.h:53 #6 0x001eddb4 in Auto::MemoryScanner::scan_pending_blocks at AutoMemoryScanner.cpp:658 #7 0x001ede0e in Auto::MemoryScanner::scan_pending_until_done at AutoMemoryScanner.cpp:680 #8 0x001edfb5 in Auto::MemoryScanner::scan at AutoMemoryScanner.cpp:731 #9 0x001e8c31 in Auto::Collector::collect at AutoCollector.cpp:64 #10 0x001e0f89 in Auto::Zone::collect at AutoZone.cpp:1324 #11 0x001d20c7 in auto_collect_internal at auto_zone.cpp:258 #12 0x001d282c in auto_collect_with_mode at auto_zone.cpp:375 #13 0x001d28c2 in auto_collection_thread at auto_zone.cpp:397 #14 0x91dc3095 in _pthread_start #15 0x91dc2f52 in thread_start Whenever I break into gdb to see where it's at during this point Auto::visitAllocatedBlocksAuto::scan_pending_blocks_visitor has always been in the stack trace. My guess would be that pending blocks are being added faster than they are scanned... To get around this problem I've tried making each thread call objc_collect(OBJC_EXHAUSTIVE_COLLECTION|OBJC_WAIT_UNTIL_DONE); after it's done with its task, but it doesn't actually seem to wait at all. Since there already seems to be a collection taking place the call returns and the thread continues. Is there anyway to truly block execution until collection has finished? Are there any callbacks that could notify when collection has started and ended? I wouldn't mind moving the temporary objects allocation to an unscanned zone, but there doesn't seem to be an equivalent CoreFoundation way to use XQuery as there is in NSXMLNode. -- Michael ___ 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: NSTableView
I still reiterate that you need a Shark sample to really find the
source of the problem. A complete backtrace during the slowdown is
really what we would need to see, not that it is slow in objc_msgSend.
Overall, the code can/should be improved. Specifically:
/* tableViewobjectValueForTableColumn:row */
- (id)tableView:(NSTableView *)aTableView objectValueForTableColumn:
(NSTableColumn *)aTableColumn row:(int)rowIndex {
DTVCell* cell = [aTableColumn dataCell];
NSString* colID = [aTableColumn identifier];
[self setupCell:cell forRow:rowIndex column:colID];
return cell;
}
You should not return a cell from this method; you should return the
object value for the cell. This method shouldn't do anything with a
cell; instead, you should setup the cell's properties in -
willDisplayCell. That alone may help you with performance...but, as I
said..a Shark trace would be needed to see the real issue. If you need
steps for getting a good trace, let us know..
corbin
/* setupCell */
- (void)setupCell:(DTVCell*)cell forRow:(int)rowIndex column:
(NSString*)colID {
BMatrixCell *xcell = [_browserCells objectAtIndex:rowIndex];
BEUtil beu;
if(xcell-_haveDesign == NO)
{
BEDesign design;
xcell-m_Design.SetFilename(xcell-mBFilename);
beu.GetFile(xcell-m_Design);
xcell-m_StitchCount = xcell-m_Design.StitchCount();
BEColorList becl;
xcell-m_Design.GetColorList(becl);
xcell-m_ColorCount = becl.GetCount();
if(xcell-m_ColorCount0) xcell-m_StitchCount -= (xcell-
m_ColorCount-1);
xcell-_haveDesign = YES;
}
if([colID compare:@NAME] == NSOrderedSame)
{
if(xcell-_name == nil)
{
Str s((const
STRCHAR*)xcell-mBFilename.StripExtension());
xcell-_name = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_name retain];
}
cell-_value = xcell-_name;
return;
}
if([colID compare:@EXTENSION] == NSOrderedSame)
{
if(xcell-_extension == nil)
{
Str s((const STRCHAR*)xcell-mBFilename.GetExtension());
xcell-_extension = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_extension retain];
}
cell-_value = xcell-_extension;
return;
}
if([colID compare:@COLORS] == NSOrderedSame)
{
if(xcell-_colors == nil)
{
xcell-_colors = [NSString stringWithFormat:@%d,xcell-
m_ColorCount];
[xcell-_colors retain];
}
cell-_value = xcell-_colors;
return;
}
if([colID compare:@STITCHES] == NSOrderedSame)
{
if(xcell-_stitches == nil)
{
xcell-_stitches = [NSString stringWithFormat:@%d,xcell-
m_StitchCount];
[xcell-_stitches retain];
}
cell-_value = xcell-_stitches;
return;
}
if([colID compare:@JUMPS] == NSOrderedSame)
{
if(xcell-m_jumps == 0)
{
BEObj *obj;
POSITION pos = xcell-m_Design.GetHeadPosition();
while(pos)
{
obj = xcell-m_Design.GetNext(pos);
obj-SetStitchCounts();
int jmp;
obj-m_data.GetData(CIMJUMPS,jmp);
xcell-m_jumps += jmp;
}
xcell-_jumps = [NSString
stringWithFormat:@%d,xcell-m_jumps];
[xcell-_jumps retain];
}
cell-_value = xcell-_jumps;
return;
}
if([colID compare:@SIZE] == NSOrderedSame)
{
if(xcell-_size == nil)
{
Str measure((const STRCHAR *)beu.GetMeasureString(1,xcell-
m_Design));
NSString *nsMeasure = [NSString stringWithCString:(const STRCHAR*)
measure encoding:NSASCIIStringEncoding];
NSRange range = [nsMeasure rangeOfString:@:];
NSString *sz = [nsMeasure
substringToIndex:range.location];
range = [sz rangeOfString:@mm];
NSString *w = [sz substringToIndex:range.location];
NSString *h = [sz substringFromIndex:range.location +
Re: NSTableView
On Jan 14, 2009, at 2:01 PM, David Blanton wrote:
- (id)tableView:(NSTableView *)aTableView objectValueForTableColumn:
(NSTableColumn *)aTableColumn row:(int)rowIndex {
DTVCell* cell = [aTableColumn dataCell];
NSString* colID = [aTableColumn identifier];
[self setupCell:cell forRow:rowIndex column:colID];
return cell;
}
Typically this method does not return a cell -- it returns a simple
object like an NSString or NSNumber. I glanced (only very quickly) at
your Shark profile and noticed a lot of calls to -[NSCell
copyWithZone:], which I'm *guessing* is because the cell you're
returning from this method is getting copied unnecessarily. If you
returned a string, it would get copied too, but that would probably be
a lot cheaper than copying a cell.
- (void)setupCell:(DTVCell*)cell forRow:(int)rowIndex column:
(NSString*)colID {
I haven't looked closely at the code for this method, but it seems to
be doing a lot of work. I don't know if that work involves loading
large files or hitting a network connection, but I assume *you* do.
If it's doing something you ALREADY KNOW is expensive, you shouldn't
be calling it every time tableView:objectValueForTableColumn:row: is
called. Again, I refer to the doc for
tableView:objectValueForTableColumn:row:
Note: tableView:objectValueForTableColumn:row: is called each time
the table cell needs to be redisplayed, so it must be efficient.
(Technically, being efficient is not the same as running in a very
short amount of time, but that's obviously what they meant.)
A common pattern for a table view's data source is for it to have an
array of objects that you want to display, one object for each row of
the table. numberOfRowsInTableView: returns the size of the array.
For a single-column table, tableView:objectValueForTableColumn:row:
returns the nth element of the array. For a multi-column table, it
gets the nth element and asks it for the value to display in the
specified column. I suspect you will get better results if you take
this approach.
If you *don't* know why tableView:objectValueForTableColumn:row: is
slow (if you expect it to be almost instantaneous), then you're back
to analyzing the Shark profile, and your problem has nothing to do
with your table view per se.
As a side note, you shouldn't use direct ivar access as in cell-
_value = xcell-_date. This violates not only encapsulation but
memory management rules.
--Andy
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: NSTableView
Holy Joker Batman!
Changing
return cell;
to
return cell-_value
gives the performance I am looking for!
Corbin, u 2 cool!
Thanks.
On Jan 14, 2009, at 12:38 PM, Corbin Dunn wrote:
I still reiterate that you need a Shark sample to really find the
source of the problem. A complete backtrace during the slowdown is
really what we would need to see, not that it is slow in objc_msgSend.
Overall, the code can/should be improved. Specifically:
/* tableViewobjectValueForTableColumn:row */
- (id)tableView:(NSTableView *)aTableView
objectValueForTableColumn:(NSTableColumn *)aTableColumn row:(int)
rowIndex {
DTVCell* cell = [aTableColumn dataCell];
NSString* colID = [aTableColumn identifier];
[self setupCell:cell forRow:rowIndex column:colID];
return cell;
}
You should not return a cell from this method; you should return
the object value for the cell. This method shouldn't do anything
with a cell; instead, you should setup the cell's properties in -
willDisplayCell. That alone may help you with performance...but, as
I said..a Shark trace would be needed to see the real issue. If you
need steps for getting a good trace, let us know..
corbin
/* setupCell */
- (void)setupCell:(DTVCell*)cell forRow:(int)rowIndex column:
(NSString*)colID {
BMatrixCell *xcell = [_browserCells objectAtIndex:rowIndex];
BEUtil beu;
if(xcell-_haveDesign == NO)
{
BEDesign design;
xcell-m_Design.SetFilename(xcell-mBFilename);
beu.GetFile(xcell-m_Design);
xcell-m_StitchCount = xcell-m_Design.StitchCount();
BEColorList becl;
xcell-m_Design.GetColorList(becl);
xcell-m_ColorCount = becl.GetCount();
if(xcell-m_ColorCount0) xcell-m_StitchCount -= (xcell-
m_ColorCount-1);
xcell-_haveDesign = YES;
}
if([colID compare:@NAME] == NSOrderedSame)
{
if(xcell-_name == nil)
{
Str s((const
STRCHAR*)xcell-mBFilename.StripExtension());
xcell-_name = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_name retain];
}
cell-_value = xcell-_name;
return;
}
if([colID compare:@EXTENSION] == NSOrderedSame)
{
if(xcell-_extension == nil)
{
Str s((const STRCHAR*)xcell-mBFilename.GetExtension());
xcell-_extension = [NSString stringWithCString:(const char *)s
encoding:NSASCIIStringEncoding];
[xcell-_extension retain];
}
cell-_value = xcell-_extension;
return;
}
if([colID compare:@COLORS] == NSOrderedSame)
{
if(xcell-_colors == nil)
{
xcell-_colors = [NSString stringWithFormat:@%d,xcell-
m_ColorCount];
[xcell-_colors retain];
}
cell-_value = xcell-_colors;
return;
}
if([colID compare:@STITCHES] == NSOrderedSame)
{
if(xcell-_stitches == nil)
{
xcell-_stitches = [NSString stringWithFormat:@%d,xcell-
m_StitchCount];
[xcell-_stitches retain];
}
cell-_value = xcell-_stitches;
return;
}
if([colID compare:@JUMPS] == NSOrderedSame)
{
if(xcell-m_jumps == 0)
{
BEObj *obj;
POSITION pos = xcell-m_Design.GetHeadPosition();
while(pos)
{
obj = xcell-m_Design.GetNext(pos);
obj-SetStitchCounts();
int jmp;
obj-m_data.GetData(CIMJUMPS,jmp);
xcell-m_jumps += jmp;
}
xcell-_jumps = [NSString
stringWithFormat:@%d,xcell-m_jumps];
[xcell-_jumps retain];
}
cell-_value = xcell-_jumps;
return;
}
if([colID compare:@SIZE] == NSOrderedSame)
{
if(xcell-_size == nil)
{
Str measure((const STRCHAR *)beu.GetMeasureString(1,xcell-
m_Design));
NSString *nsMeasure = [NSString stringWithCString:(const
STRCHAR*)measure encoding:NSASCIIStringEncoding];
NSRange range = [nsMeasure rangeOfString:@:];
NSString *sz = [nsMeasure
substringToIndex:range.location];
Report writer for Cocoa?
Namaste! I could be missing the obvious, however, my question is whether a report writer tool (like Crystal Reports or MS Access, by way of example) exists for Cocoa? I Googled around for a bit tyring to find something (other than ReportMill). I've looked at (and read) the printing-relevant Apple dox. I've read the relevant chapter in Hillegass' book too. I can see how much work writing pretty reports is going to be. Unless... I missed a point concerning an NSView - it knows about PDF, etc. The dox don't really go into this much, so here's a follow-on question to the first: If I take the time to draw out a view in IB with all my fields, etc., will that print as-is? This, then, would become the report writer, right? I'm looking for good, solid advice on creating reports. I do have some graphics I'd like to include in my reports, but most are lists, and some are lists with child details too. Many thanks in advance (and I hope this isn't a Help Vampire question)!!! Peace, Love, and Light, /s/ Jon C. Munson II ___ 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: Report writer for Cocoa?
On Jan 14, 2009, at 3:01 PM, Jon C. Munson II wrote: I could be missing the obvious, however, my question is whether a report writer tool (like Crystal Reports or MS Access, by way of example) exists for Cocoa? I think what a lot of folks are doing is just generating HTML and printing it with WebView. sherm-- ___ 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: Report writer for Cocoa?
A lot of people tend to link in webkit and layout a printed report using html and css. It's somewhat eqasier than standard web dev since you know that only webkit will be displaying your layout. No need for cross browser compatability. I use that aqpproach for one of my apps and it works well enough. Don't get me wrong - it still feels like a major hack, but thre results are decent and it was pretty quick and easy to implement. -jim.. sent from my G1 On Jan 14, 2009 2:13 PM, Jon C. Munson II jmun...@his.com wrote: Namaste! I could be missing the obvious, however, my question is whether a report writer tool (like Crystal Reports or MS Access, by way of example) exists for Cocoa? I Googled around for a bit tyring to find something (other than ReportMill). I've looked at (and read) the printing-relevant Apple dox. I've read the relevant chapter in Hillegass' book too. I can see how much work writing pretty reports is going to be. Unless... I missed a point concerning an NSView - it knows about PDF, etc. The dox don't really go into this much, so here's a follow-on question to the first: If I take the time to draw out a view in IB with all my fields, etc., will that print as-is? This, then, would become the report writer, right? I'm looking for good, solid advice on creating reports. I do have some graphics I'd like to include in my reports, but most are lists, and some are lists with child details too. Many thanks in advance (and I hope this isn't a Help Vampire question)!!! Peace, Love, and Light, /s/ Jon C. Munson II ___ 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%40jimandkoka.com This email sent to j...@jimandkoka.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
UITableViewController with dynamic data
Hi, I am using a UITableViewController and the data for the table view is generated dynamically (fetched from the network). The data is normally fetched based on some search. But when the Controller is loaded and i don't have the data ready, i want to return something back sane and later reload the data when it becomes ready. What is the best way to do this ? -numberOfSectionsinRow : Is it safe to assume that this function is called always *first* before constructing the table ? At least i want to be able to initiate the fetch (if it is not already done before) here. -As i can't return 0 for numberOfSectionsInRow, i still have to return 1 there and in numberOfRowsInSection, return zero till the data becomes available The problem is that i need to initiate the fetch somewhere during the table construction and i don't want to put hooks into all of the tableView delegate methods. As the data for the tableview is dynamic, i need to tell the framework not to call the other delegate methods till i return something valid in numberOfSectionsInRow/numberOfRowsInSection. Is this a safe assumption ? thanks mohan ___ 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: NSTableView
The code provided looks very unusual and clearly does not follow the
Model-View-Controller design. It is almost always a bad idea to store
information in the user interface.
The following line is extremely suspect:
int count = _browserView-fileList.GetCount();
Why doesn't your tableView's data source have its own direct reference to
fileList ?
I suggest the following:
- (int)numberOfRowsInTableView:(NSTableView *)tableView {
int count = 0;
if(_fileList != NULL)
{
count = _fileList.GetCount();
}
return count;
}
What is going on with the following?
[_browserView selectAll:self];
[_browserCells release];
_browserCells = [_browserView selectedCells];
[_browserCells retain];
Don't select stuff in - (int)numberOfRowsInTableView:
Even if you wanted the cells from the matrix for some reason, why not
[_browserCells release];
_browserCells = [[_browserView cells] copy];
But using the cells of a matrix outside the matrix is extremely suspect. The
cells should not contain any information that you can't get from some other
source like _fileList.
The rest of teh code that copies information out of cells in a matrix in order
to set information in cells used by a tableView is so horrific that I won't
comment any more. Please read about Model-View-Controller.
___
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: UITableViewController with dynamic data
I assume you meant numberOfSectionsInTableView:anyway, a good strategy might be to just go ahead and return 0 initially, and then whenever you have successfully pulled your data from the network, do [tableView reloadData] and then you can return the appropriate numbers at that time, since they will be ready. Luke On Jan 14, 2009, at 12:24 PM, Mohan Parthasarathy wrote: Hi, I am using a UITableViewController and the data for the table view is generated dynamically (fetched from the network). The data is normally fetched based on some search. But when the Controller is loaded and i don't have the data ready, i want to return something back sane and later reload the data when it becomes ready. What is the best way to do this ? -numberOfSectionsinRow : Is it safe to assume that this function is called always *first* before constructing the table ? At least i want to be able to initiate the fetch (if it is not already done before) here. -As i can't return 0 for numberOfSectionsInRow, i still have to return 1 there and in numberOfRowsInSection, return zero till the data becomes available The problem is that i need to initiate the fetch somewhere during the table construction and i don't want to put hooks into all of the tableView delegate methods. As the data for the tableview is dynamic, i need to tell the framework not to call the other delegate methods till i return something valid in numberOfSectionsInRow/numberOfRowsInSection. Is this a safe assumption ? thanks mohan ___ 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/luketheh%40apple.com This email sent to luket...@apple.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: UITableViewController with dynamic data
Yes, that's what i meant. Just to confirm.. It is safe to assume that numberOfSectionsInTableView will be first function called ? I am thinking of this sequence.. 1) The code receives low memory warning and frees up the data that the tableView is currently displaying 2) CellForRowAtIndexPath is called now for Nth Row. (2) in my code would reference the data that has been freed in step (1). Is it possible to invalidate the tableView Data also while freeing memory at step (1) ? thanks mohan On Wed, Jan 14, 2009 at 12:32 PM, Luke the Hiesterman luket...@apple.comwrote: I assume you meant numberOfSectionsInTableView:anyway, a good strategy might be to just go ahead and return 0 initially, and then whenever you have successfully pulled your data from the network, do [tableView reloadData] and then you can return the appropriate numbers at that time, since they will be ready. Luke On Jan 14, 2009, at 12:24 PM, Mohan Parthasarathy wrote: Hi, I am using a UITableViewController and the data for the table view is generated dynamically (fetched from the network). The data is normally fetched based on some search. But when the Controller is loaded and i don't have the data ready, i want to return something back sane and later reload the data when it becomes ready. What is the best way to do this ? -numberOfSectionsinRow : Is it safe to assume that this function is called always *first* before constructing the table ? At least i want to be able to initiate the fetch (if it is not already done before) here. -As i can't return 0 for numberOfSectionsInRow, i still have to return 1 there and in numberOfRowsInSection, return zero till the data becomes available The problem is that i need to initiate the fetch somewhere during the table construction and i don't want to put hooks into all of the tableView delegate methods. As the data for the tableview is dynamic, i need to tell the framework not to call the other delegate methods till i return something valid in numberOfSectionsInRow/numberOfRowsInSection. Is this a safe assumption ? thanks mohan ___ 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/luketheh%40apple.com This email sent to luket...@apple.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: UITableViewController with dynamic data
The way to invalidate data in the tableView when you free it would be to reloadData. So, get rid of whatever you're getting rid of, build your new, truncated list of items, and then call [tableView reloadData] so you put only the data you've kept around in the table. Also, if you're worried about a race condition, you could set a flag before you begin freeing things, so that if any data is requested while you're in the process of freeing but you haven't yet called reloadData, you will do the right thing. Luke On Jan 14, 2009, at 12:57 PM, Mohan Parthasarathy wrote: Yes, that's what i meant. Just to confirm.. It is safe to assume that numberOfSectionsInTableView will be first function called ? I am thinking of this sequence.. 1) The code receives low memory warning and frees up the data that the tableView is currently displaying 2) CellForRowAtIndexPath is called now for Nth Row. (2) in my code would reference the data that has been freed in step (1). Is it possible to invalidate the tableView Data also while freeing memory at step (1) ? thanks mohan On Wed, Jan 14, 2009 at 12:32 PM, Luke the Hiesterman luket...@apple.com wrote: I assume you meant numberOfSectionsInTableView:anyway, a good strategy might be to just go ahead and return 0 initially, and then whenever you have successfully pulled your data from the network, do [tableView reloadData] and then you can return the appropriate numbers at that time, since they will be ready. Luke On Jan 14, 2009, at 12:24 PM, Mohan Parthasarathy wrote: Hi, I am using a UITableViewController and the data for the table view is generated dynamically (fetched from the network). The data is normally fetched based on some search. But when the Controller is loaded and i don't have the data ready, i want to return something back sane and later reload the data when it becomes ready. What is the best way to do this ? -numberOfSectionsinRow : Is it safe to assume that this function is called always *first* before constructing the table ? At least i want to be able to initiate the fetch (if it is not already done before) here. -As i can't return 0 for numberOfSectionsInRow, i still have to return 1 there and in numberOfRowsInSection, return zero till the data becomes available The problem is that i need to initiate the fetch somewhere during the table construction and i don't want to put hooks into all of the tableView delegate methods. As the data for the tableview is dynamic, i need to tell the framework not to call the other delegate methods till i return something valid in numberOfSectionsInRow/numberOfRowsInSection. Is this a safe assumption ? thanks mohan ___ 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/luketheh%40apple.com This email sent to luket...@apple.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
Custom drawn menus?
Is there any good way to custom draw a menu? Or is the old carbon 'mdef', like the star menu sample, the only way? And will it break in 10.6? Scott ___ 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: Custom drawn menus?
On Wed, Jan 14, 2009 at 4:14 PM, Scott Andrew scottand...@roadrunner.com wrote: Is there any good way to custom draw a menu? Or is the old carbon 'mdef', like the star menu sample, the only way? And will it break in 10.6? You could use a custom view and set it as the menu item's view ... -- I.S. ___ 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: Custom drawn menus?
Not up for the menu item route. I am looking at just changing the background and shape slightly or the whole menu to match my UI. Scott On Jan 14, 2009, at 1:16 PM, I. Savant wrote: On Wed, Jan 14, 2009 at 4:14 PM, Scott Andrew scottand...@roadrunner.com wrote: Is there any good way to custom draw a menu? Or is the old carbon 'mdef', like the star menu sample, the only way? And will it break in 10.6? You could use a custom view and set it as the menu item's view ... -- I.S. ___ 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: Custom drawn menus?
Not up for the menu item route. I am looking at just changing the background and shape slightly or the whole menu to match my UI. Unfortunately there's not much of a good route to go here. Using a custom view requires you to do handle *all* drawing, mouse handling, keyboard events, etc. It is possible to do using an undocumented method, but that's not on-topic for this list. -- Joe Ranieri ___ 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: NIB's owner doesn't get released because of NSArrayController
I just found a solution, but I don't like it. I added method -beforeRelease to my view controller where I send -unbind: to my array controller. I call this method from my window controller just before -release call. This helps. But looks ugly... Does somebody know better way? On Thu, Jan 15, 2009 at 12:51 AM, Vitaly Ovchinnikov vitaly.ovchinni...@gmail.com wrote: Hello all, I have a custom-loaded NIB file with NSView and NSArrayController. Array controller is bound to file's owner (NSObject's derived class) and use some keypath for contentArray. The problem is when I close a window and main window controller tries to release this file's owner - array controller still holds a reference to it and doesn't allow to do that. So I have two questions: 1. What is the correct way to catch closing of the view to unbound array controller? 2. Do I really need to catch something? Maybe there is a simpler solution? Any ideas? Thank you. ___ 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: NIB's owner doesn't get released because of NSArrayController
On Wed, Jan 14, 2009 at 3:51 PM, Vitaly Ovchinnikov vitaly.ovchinni...@gmail.com wrote: Hello all, I have a custom-loaded NIB file with NSView and NSArrayController. Array controller is bound to file's owner (NSObject's derived class) and use some keypath for contentArray. The problem is when I close a window and main window controller tries to release this file's owner - array controller still holds a reference to it and doesn't allow to do that. So I have two questions: 1. What is the correct way to catch closing of the view to unbound array controller? 2. Do I really need to catch something? Maybe there is a simpler solution? Any ideas? Top-level objects in a nib need to be released. You can do this in the File's Owner object when the window is closed, for example. See this page for some additional info on nib object memory management: http://developer.apple.com/documentation/Cocoa/Conceptual/MemoryMgmt/Articles/MemMgmtNibObjects.html Jason ___ 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: KVC error from bound SortDescriptor
On Jan 14, 2009, at 11:10 AM, Quincey Morris wrote:
On Jan 14, 2009, at 08:35, Keary Suska wrote:
On Jan 14, 2009, at 4:54 AM, Steven Hamilton wrote:
I have a customer WindowController class and within it I have a
method that returns a NSSortDescriptor. Like this;
-(NSArray *)nameSortDescriptors
{
NSSortDescriptor *sorter;
sorter = [[[NSSortDescriptor alloc]
initWithKey: @name
ascending: YES] autorelease];
return ([NSArray arrayWithObject: sorter]);
}
...
Error setting value for key path nameSortDescriptors of object
MLAccountController: 0x58ae390 (from bound object
NSArrayController: 0x589d1a0[entity: Account, number of selected
objects: 1](null)): [MLAccountController 0x58ae390
setValue:forUndefinedKey:]: this class is not key value coding-
compliant for the key nameSortDescriptors.
... 2) make nameSortDescriptors key value coding compliant.
Jumping in with a quibble ...
The OP's window controller class *is* KVC-compliant for the
'nameSortDescriptors' property, but it's an immutable property. The
sortable column needs it to be a mutable property. The error message
is slightly at fault here.
So the solution is to put the array of sort descriptors in an
instance variable and write a setter for the property, which is, I
think, what you were implying the OP should do.
I stand corrected. To be more specific, I would say a setter or
indexed accessors.
Best,
Keary Suska
Esoteritech, Inc.
Demystifying technology for your home or business
___
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
addObserver problem with protocols
Hi, I have protocol X that is implemented by a few classes of my own. These classes are not known directly to my main code but only through the protocol (idprotocol). When my main code tries to addObserver:self to the object idprotocol, i get a warning saying that my protocol does not implement it. This would have worked if i used the real object behind. What should i do ? Just add a method like addObserver in the protocol and in the real object make the method just invoke the [super addobserver]. Or is there a better way to solve it ? thanks mohan ___ 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: addObserver problem with protocols
On Jan 14, 2009, at 3:18 PM, Mohan Parthasarathy wrote: I have protocol X that is implemented by a few classes of my own. These classes are not known directly to my main code but only through the protocol (idprotocol). When my main code tries to addObserver:self to the object idprotocol, i get a warning saying that my protocol does not implement it. This would have worked if i used the real object behind. What should i do ? Just add a method like addObserver in the protocol and in the real object make the method just invoke the [super addobserver]. Or is there a better way to solve it ? If you search the archives you will find a few discussions on this subject. I don't remember the reason exactly, but you just can't use that syntax. You can use an informal protocol instead, but you won't get the compile-time checks that protocol gives you. HTH, Keary Suska Esoteritech, Inc. Demystifying technology for your home or business ___ 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: Debugging objc_msg errors
On Jan 14, 2009, at 11:18 AM, Jeff Johnson wrote: On Jan 14, 2009, at 1:04 PM, John Nairn wrote: I occasionally get difficult-to-debug crashes where that last thing in the crash log is a message being sent, which ends up being a message sent to an object that has been released (I think it is objc_msg, but I am not looking at a crash log now). These happen much more often in Leopard than in Tiger. Since I have many messages going to many objects, it can take me a long time to track these problems down (in one case several weeks). I am fairly certain it is not a notification message because I also remove objects as observers before they are deallocated. Is there a debugging tool that can provide more information about what message was sent to what object at the time of the crash? See the following for a good tutorial: http://www.sealiesoftware.com/blog/archive/2008/09/22/objc_explain_So_you_crashed_in_objc_msgSend.html That will help you extract info from the crash log itself. There are also other debugging modes to run your program that can either catch the bug before it crashes, or provide more details about the problem when it does crash. Mac OS X Debugging Magic http://developer.apple.com/technotes/tn2004/tn2124.html Look in particular for libgmalloc, MallocStackLogging, and NSZombie. Clang Static Analyzer http://clang.llvm.org/StaticAnalysis.html A compile-time checker that can identify some retain/release errors. -- Greg Parker gpar...@apple.com Runtime Wrangler ___ 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: addObserver problem with protocols
On Jan 14, 2009, at 2:18 PM, Mohan Parthasarathy wrote: I have protocol X that is implemented by a few classes of my own. These classes are not known directly to my main code but only through the protocol (idprotocol). When my main code tries to addObserver:self to the object idprotocol, i get a warning saying that my protocol does not implement it. This would have worked if i used the real object behind. What should i do ? Just add a method like addObserver in the protocol and in the real object make the method just invoke the [super addobserver]. Or is there a better way to solve it ? Some options: 1. Use `NSObjectMyProtocol*` instead of `idMyProtocol`. Then any method from class NSObject is allowed along with the protocol's own methods. Since addObserver: is declared as a class on NSObject, this should pacify the compiler. 2. Add -addObserver: to the protocol. 3. Create another protocol that includes -addObserver:. Then either use `idMyProtocol,MyObserverProtocol`, or make MyProtocol itself adopt MyObserverProtocol. #1 is probably the best, assuming all of your objects are in fact NSObjects. -- Greg Parker gpar...@apple.com Runtime Wrangler ___ 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: Debugging objc_msg errors
On 1/14/09 2:33 PM, Greg Parker said: On Jan 14, 2009, at 11:18 AM, Jeff Johnson wrote: On Jan 14, 2009, at 1:04 PM, John Nairn wrote: I occasionally get difficult-to-debug crashes where that last thing in the crash log is a message being sent, which ends up being a message sent to an object that has been released (I think it is objc_msg, but I am not looking at a crash log now). These happen much more often in Leopard than in Tiger. Since I have many messages going to many objects, it can take me a long time to track these problems down (in one case several weeks). I am fairly certain it is not a notification message because I also remove objects as observers before they are deallocated. Is there a debugging tool that can provide more information about what message was sent to what object at the time of the crash? See the following for a good tutorial: http://www.sealiesoftware.com/blog/archive/2008/09/22/ objc_explain_So_you_crashed_in_objc_msgSend.html That will help you extract info from the crash log itself. Not to hijack this thread, but... anyone know what objc_msgSend_fixup is? Recently I've seen a lot of these types of crashes in my (64bit, GC) app. Thread 0 Crashed: 0 libobjc.A.dylib 0x7fff809bbcf2 _objc_fixupMessageRef + 39 1 libobjc.A.dylib 0x7fff809bdc7f objc_msgSend_fixup + 119 2 com.apple.CoreFoundation0x7fff82b798b7 __CFMachPortPerform + 119 3 com.apple.CoreFoundation0x7fff82b9b3ac CFRunLoopRunSpecific + 3516 4 com.apple.HIToolbox 0x7fff80548da2 RunCurrentEventLoopInMode + 278 5 com.apple.HIToolbox 0x7fff80548bd8 ReceiveNextEventCommon + 322 6 com.apple.HIToolbox 0x7fff80548a83 BlockUntilNextEventMatchingListInMode + 79 7 com.apple.AppKit0x7fff80da93b0 _DPSNextEvent + 603 8 com.apple.AppKit0x7fff80da8cf1 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 136 9 com.apple.AppKit0x7fff80da2a84 -[NSApplication run] + 455 10 com.apple.AppKit0x7fff80d6f83c NSApplicationMain + 373 -- Sean McBride, B. Eng s...@rogue-research.com Rogue Researchwww.rogue-research.com Mac Software Developer Montréal, Québec, Canada ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
/tmp folder ok?
so an app doesn't need administrative rights to writeToFile in the /tmp folder? ___ 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: /tmp folder ok?
Every app can write to the tmp folder but take a look at: NSString * NSTemporaryDirectory (void); On 15 Jan 2009, at 00:31, Chunk 1978 wrote: so an app doesn't need administrative rights to writeToFile in the / tmp folder? ___ 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/maillist%40codingmammoth.com This email sent to maill...@codingmammoth.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: /tmp folder ok?
On 15 Jan 2009, at 10:31 am, Chunk 1978 wrote:
so an app doesn't need administrative rights to writeToFile in the /
tmp folder?
___
Don't do that.
Instead use the FindFolder with the 'kTemporaryFolderType' constant.
Here's a Cocoa wrapper (a category on NSFileManager) to help:
@implementation NSFileManager (FindFolder)
- (NSString*) pathToFolderOfType:(const OSType) folderType
shouldCreateFolder:(BOOL) create
{
OSErr err;
FSRef ref;
NSString* path = nil;
err = FSFindFolder( kUserDomain, folderType, create, ref);
if ( err == noErr )
{
// convert to CFURL and thence to path
CFURLRef url = CFURLCreateFromFSRef( kCFAllocatorSystemDefault,
ref );
path = (NSString*) CFURLCopyFileSystemPath( url,
kCFURLPOSIXPathStyle );
CFRelease( url );
}
return [path autorelease];
}
@end
But yes, once you have the path you should have write permission for it.
Note also though, that for the case of safe saving (i.e. write to a
temporary file then swap with the 'real' file), you don't need to do
that yourself. Just pass yes in the 'atomically' parameter in methods
such as [NSData writeToFile:atomically:];
hth,
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/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: Debugging objc_msg errors
On Jan 14, 2009, at 4:21 PM, Sean McBride wrote: Not to hijack this thread, but... anyone know what objc_msgSend_fixup is? Recently I've seen a lot of these types of crashes in my (64bit, GC) app. I could have sworn I've seen this before, and the problem ended up being a subtle memory corruption bug. Have you tried turning on Guard Malloc lately? Memory corruption can still happen in GC-enabled apps... Nick Zitzmann http://www.chronosnet.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: Debugging objc_msg errors
On Jan 14, 2009, at 11:04 AM, John Nairn wrote: I occasionally get difficult-to-debug crashes where that last thing in the crash log is a message being sent, which ends up being a message sent to an object that has been released (I think it is objc_msg, but I am not looking at a crash log now). These happen much more often in Leopard than in Tiger. Since I have many messages going to many objects, it can take me a long time to track these problems down (in one case several weeks). I am fairly certain it is not a notification message because I also remove objects as observers before they are deallocated. Is there a debugging tool that can provide more information about what message was sent to what object at the time of the crash? What the others have posted is good, but IMHO the easiest way to debug overreleased objects is with Instruments. You frequently want to know where the over release was. http://www.corbinstreehouse.com/blog/index.php/2007/10/instruments-on-leopard-how-to-debug-those-random-crashes-in-your-cocoa-app/ corbin ___ 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: /tmp folder ok?
NSTemporaryDirectory() already does (effectively) that, and yields the
same resulting folder.
You should just use NSTemporaryDirectory().
.chris
On Wed, Jan 14, 2009 at 3:44 PM, Graham Cox graham@bigpond.com wrote:
On 15 Jan 2009, at 10:31 am, Chunk 1978 wrote:
so an app doesn't need administrative rights to writeToFile in the /tmp
folder?
___
Don't do that.
Instead use the FindFolder with the 'kTemporaryFolderType' constant. Here's
a Cocoa wrapper (a category on NSFileManager) to help:
@implementation NSFileManager (FindFolder)
- (NSString*) pathToFolderOfType:(const OSType) folderType
shouldCreateFolder:(BOOL) create
{
OSErr err;
FSRef ref;
NSString* path = nil;
err = FSFindFolder( kUserDomain, folderType, create, ref);
if ( err == noErr )
{
// convert to CFURL and thence to path
CFURLRef url = CFURLCreateFromFSRef(
kCFAllocatorSystemDefault, ref );
path = (NSString*) CFURLCopyFileSystemPath( url,
kCFURLPOSIXPathStyle );
CFRelease( url );
}
return [path autorelease];
}
@end
But yes, once you have the path you should have write permission for it.
Note also though, that for the case of safe saving (i.e. write to a
temporary file then swap with the 'real' file), you don't need to do that
yourself. Just pass yes in the 'atomically' parameter in methods such as
[NSData writeToFile:atomically:];
hth,
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/ctparker%40gmail.com
This email sent to ctpar...@gmail.com
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: /tmp folder ok?
On Wed, Jan 14, 2009 at 7:24 PM, Chris Parker ctpar...@gmail.com wrote: You should just use NSTemporaryDirectory(). Do not just use NSTemporaryDirectory if you need to perform an atomic save operation (write new data to location X and then overwrite old data at location Y by moving X - Y). If will not work if the destination is on a different volume from the source. In this case you must use FSFindFolder with the volume reference number of the volume on which you wish to save your data. Read FSFolder.h for more information. --Kyle Sluder ___ 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: Report writer for Cocoa?
I make an RTFD report template in TextEdit and read this template into an NSMutableAttributedString. My template contains keywords such as FIRST_NAME, LAST_NAME, etc. I then substitute the keywords in the attributed string with real data, and then initialize an NSTextView with the string. This text view can be printed, or the attributed string can be written to an RTFD or other type of file. I haven't tried substituting graphics into the report template yet. One big benefit of this simple technique is that end-users can customize the report template themselves using TextEdit. You can certainly print a view created in IB, but NSTableViews, etc. are designed for the screen and don't look good in print. Dave On Jan 14, 2009, at 3:01 PM, Jon C. Munson II wrote: Namaste! I could be missing the obvious, however, my question is whether a report writer tool (like Crystal Reports or MS Access, by way of example) exists for Cocoa? I Googled around for a bit tyring to find something (other than ReportMill). I've looked at (and read) the printing-relevant Apple dox. I've read the relevant chapter in Hillegass' book too. I can see how much work writing pretty reports is going to be. Unless... I missed a point concerning an NSView - it knows about PDF, etc. The dox don't really go into this much, so here's a follow-on question to the first: If I take the time to draw out a view in IB with all my fields, etc., will that print as-is? This, then, would become the report writer, right? I'm looking for good, solid advice on creating reports. I do have some graphics I'd like to include in my reports, but most are lists, and some are lists with child details too. Many thanks in advance (and I hope this isn't a Help Vampire question)!!! Peace, Love, and Light, /s/ Jon C. Munson II ___ 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: WebView (Caching)
On 15/01/2009, at 6:53 AM, Joseph Crawford wrote: When I drop a WebView on my XIB does it automatically cache the file(s) that it loads? I ask because I could not get a JS command to work properly even after I fixed the would be problem. This morning it started working just fine. If it does cache can anyone direct me to something I can read on how to disable caching? This has been discussed several times on the WebKitSDK-dev list and the short answer is that yes, it does caching and no, there is nothing you can do to turn it off. I and others have tried a myriad of workarounds but unfortunately the caching is down in WebCore. The WebKit developers are aware of this issue and so hopefully it will be resolved in future but please file a bug. The only solution that does work is the old web developer trick of appending a random argument to the URL, e.g.: file:///path/to/myjavascript.js?random=12345234 If you change the random argument each time the page is loaded, the JavaScript file will be reloaded each time. -- Rob Keniger ___ 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: IKImageView setImage only works once.
On 15 Jan 2009, at 12:22 pm, James Trankelson wrote: Well, judging from how often ImageKit questions appear to be answered here, my expectations aren't high, but here goes anyway... I'm using an IKImageView and calling setImage to set its contents. The problem is that setImage only works the first time. Subsequent calls don't update the image. The first always remains. Any ideas? IKImageView doesn't implement -setImage: It has -setImage:imageProperties: and -setImageWithURL: Did you mean one of these? You may need to call -setNeedsDisplay:YES also after setting the image (the docs are silent on this point but other ImageKit classes need this call for some operations - they don't necessarily flag the needed update internally). any help? --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/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Apply Preferences To Current Window
I've just worked through Chapter 13 of Hillegass's book, which deals with setting user defaults. My problem is that the preferences (e.g. colorwell which changes the background color of a tableview) only take effect when I open a new document, whereas I'd like them to be applied to the current one. I'm pretty sure this would involve bindings, and there was a brief section in the book detailing binding a checkbox, but it didn't help me with changing the color. I also looked here http://developer.apple.com/DOCUMENTATION/Cocoa/Conceptual/DrawColor/Tasks/StoringNSColorInDefaults.html#/ /apple_ref/doc/uid/20001693-SW3 Thanks for the help! - Walker Argendeli ___ 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: Report writer for Cocoa?
On Wed, 14 Jan 2009 15:01:02 -0500, Jon C. Munson II jmun...@his.com said: I've looked at (and read) the printing-relevant Apple dox. I've read the relevant chapter in Hillegass' book too. I can see how much work writing pretty reports is going to be. Unless... I missed a point concerning an NSView - it knows about PDF, etc. The dox don't really go into this much, so here's a follow-on question to the first: If I take the time to draw out a view in IB with all my fields, etc., will that print as-is? This, then, would become the report writer, right? I would suggest breaking the problem down in to the basic elements of a page, classifying the kinds of thing you'd like to have appear on the printed page. By default, a view draws itself by asking its subviews to draw themselves. Thus you can have your own special NSView subclasses that know how to draw themselves, and combine them within superviews, and now everything becomes very tight and simple and object-oriented and easily maintained. If you want to design certain aspects of a view in IB, of course you can, and you can load a nib multiple times to get multiple copies of it. m. -- matt neuburg, phd = m...@tidbits.com, http://www.tidbits.com/matt/ A fool + a tool + an autorelease pool = cool! One of the 2007 MacTech Top 25: http://tinyurl.com/2rh4pf AppleScript: the Definitive Guide - Second Edition! http://www.amazon.com/gp/product/0596102119 ___ 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: IKImageView setImage only works once.
Ha! I actually tried both of your suggestions, but without avail. It turns out that calling [setImage: imageProperties:] is happiest to be called from the main thread. Calling it by using performSelectorOnMainThread: works. jt On Thu, Jan 15, 2009 at 2:51 AM, Graham Cox graham@bigpond.com wrote: On 15 Jan 2009, at 12:22 pm, James Trankelson wrote: Well, judging from how often ImageKit questions appear to be answered here, my expectations aren't high, but here goes anyway... I'm using an IKImageView and calling setImage to set its contents. The problem is that setImage only works the first time. Subsequent calls don't update the image. The first always remains. Any ideas? IKImageView doesn't implement -setImage: It has -setImage:imageProperties: and -setImageWithURL: Did you mean one of these? You may need to call -setNeedsDisplay:YES also after setting the image (the docs are silent on this point but other ImageKit classes need this call for some operations - they don't necessarily flag the needed update internally). any help? --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/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Localized resources not loading from external frameworks
I have an external framework I built in XCode 2.5. I have an English.lproj folder and a Japanese.lproj folder. I build my framework and put it in /Library/Frameworks. All the files are in the right locations, but when I run my app which is linked against the framework, the English version of my UI from the framework always loads no matter what language I have set in System Preferences. Any ideas how to get the localized resources to load? Thanks, ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
JOB OPENING- MAC OS modem interface SW ENGINEER
JOB OPENING--- I am a recruiter of senior level hardware and software engineers for contract positions nationwide. We place engineers in the areas of semiconductors, aerospace, commercial, telecom, networking, and government defense. If you are interested in this contract position or others, please send me you current resume and manager references. Also, please feel free to call me directly if you have any questions about this position. Lindsey Glass Triple Crown Consulting, LLC Technical Recruiter Direct: 512-623-5773 Toll Free: 866-901-8880 ext 140 Fax: 512-233-5341 www.tripleco.com MAC OS position- Montreal, Canada 6-12 months new modem product development, architecture through verification and test big thing is Mac OS experience USB drivers (mac related) ACM, ECM, CCM arm 9 exp is ideal but 7 will work embedded C (95%) C++ Required Prior Experience/job Qualifications Bachelor Degree in Electrical Engineering/Computer Science/Computer Engineering or equivalent. Masters preferred. 5+ years experience in design, development, and verification of complex Embedded Real-Time Software Systems. Experience in the design and implementation of Real-Time / High Performance software, required. Wireless (GSM, GPRS, EDGE, TDMA, WCDMA), data communication, or networking experience preferred. Experience in design, development, and verification of platform related software preferred (RTOS integration, inter-processor communication, HW / SW interfacing, etc.). Serial (USB), MAC-OS network interface (ACM, ECM, CCM, …) and card bus drivers experience preferred. Experience with ARM embedded processors, preferred. Clear ability to learn and grow in Software Engineering skills, required to design, implement, integrate, and maintain complex pieces of software and software products. Commercial Product experience required. Proficient in C/C++, and assembly. Familiarity with model-based engineering (SDL, UML, or the equivalent) preferred. Excellent verbal and written communication skills. Excellent problem solving, logic, and analytical skills Summary of Job Duties Responsible for the design, implementation, integration and architecture of Wideband CDMA/GSM-GPRS-EDGE terminal products, specifically platform drivers, and protocol software. Assist in integration / verification, and maintenance (defect analysis, isolation, resolution, and verification) of modem interface software for MAC OS data card reference and customer platforms. Lindsey Glass Triple Crown Consulting, LLC Technical Recruiter Direct: 512-623-5773 Toll Free: 866-901-8880 ext 140 Fax: 512-233-5341 www.tripleco.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
Text field with buttons like the To field in Mail
Hello all, I am very new to cocoa development and just making my first steps. What I am trying to do is something like a To field in Mail.app which provides autocomplete and once matched replaces the text with the button. After hours of googling it looks like I have to put NSButtonCell inside NSTextView (or NSTextField?) however, I can't understand how do it though it should be fairly common as I see same kind of behavior in many applications! (especially for list of tags) Maybe there is a ready made component? Thank you advance for your help and sorry for such a lame question! -- Sergey Kuleshov ___ 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
What determines VSIZE allocation?
I have searched for an answer to this question, including in Amit Singh's book, but have not found the answer (at least, not in a form that I recognize it). I have a fairly simple Cocoa app that creates an NSStatusItem with a small menu. It is compiled using Garbage Collection and in 64bit mode (it identifies in the Activity Monitor as Intel(64 bit). In non-debug mode it basically just sits there. The active memory usage is quite reasonable -- about 4.5 to 7.5MB of RPRVT, 10MB RSIZE and 23-30MB RSHRD. However, the VSIZE is 8.36GB (giga, with a G). Compiling in 32bit mode drops VSIZE to 475MB, which is still an awful lot for such a small app. I know there are people who will say VSIZE doesn't matter if there is no paging, but if I ever sought to distribute my app people will not want to install a statusitem that looks like a memory hog. My question is two-fold: (1) How does the OS determine how much to allocate to VSIZE? and (2) is there something I am doing wrong in my app (likely) that is causing the out-sized allocation? I am a newbie w/r/t cocoa and objective-C, but I have read the article on garbage collection and tried to understand and apply it. ___ 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
ImageAndTextCell + NSOutlineView editing problem
Greetings Cocoa-Dev!
I am having the hardest time with a persistent and strange error with
Apple's ImageAndTextCell in my NSOutlineView.
I have the ImageAndTextCell set to be the outline's data cell via
NSTableColumn's -setDataCell: method, and I set the image of the cell
in the outlineView's delegate, using -
(void)outlineView:willDisplayCell:forTableColumn:item: And here's
where the trouble starts.
The problem I am encountering is this: When I have an image set in the
cell, and I go to edit the text of the *first* item in any particular
set of children, after the first keystroke, the text field goes
haywire, turning white, with the text appearing only as shadow lines,
and editing stops. If I click off the cell, the mess goes away, the
one keystroke's worth of edit remains, but does not get noticed by the
bindings.
Here is a picture of what happens: http://www.pileofdebris.com/editError.png
Does anyone have any thoughts about why this might be happening? I've
figured out that if my willDisplayCell: delegate method is empty, this
does not happen. Here is the code for that method, should it be of
any help.
- (void)outlineView:(NSOutlineView *)outlineView willDisplayCell:
(id)cell forTableColumn:(NSTableColumn *)tableColumn item:(id)item
{
NSManagedObject *node = [item representedObject];
NSString *type = [node valueForKey:@type];
NSString *tag = [node valueForKey:@tag];
if ([type isEqual:@collection]) {
[cell setImage:[self defaultFolderImage]];
} else if ([type isEqual:@library]) {
if ([tag isEqual:@booksLibrary]) {
[cell setImage:[self booksLibraryImage]];
} else if ([tag isEqual:@articlesLibrary]) {
[cell setImage:[self articlesLibraryImage]];
} else {
[cell setImage:[self articlesLibraryImage]];
}
} else if ([type isEqual:@section]) {
[cell setImage:nil];
[cell setEditable:NO];
[cell setSelectable:NO];
} else if ([type isEqual:@smartCollection]) {
[cell setImage:[self smartFolderImage]];
}
}
Thanks so much for any input people can offer.
pax,
Andy S.
___
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: IKImageView setImage only works once.
On 15 Jan 2009, at 2:07 pm, James Trankelson wrote: It turns out that calling [setImage: imageProperties:] is happiest to be called from the main thread. Calling it by using performSelectorOnMainThread: works. Anything concerned with views or other drawing must be called on the main thread. That goes for any method implemented by any view. --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/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: What determines VSIZE allocation?
On 14 Jan 09, at 14:31, Tobias Zimmerman wrote: I have searched for an answer to this question, including in Amit Singh's book, but have not found the answer (at least, not in a form that I recognize it). I have a fairly simple Cocoa app that creates an NSStatusItem with a small menu. It is compiled using Garbage Collection and in 64bit mode (it identifies in the Activity Monitor as Intel(64 bit). In non-debug mode it basically just sits there. The active memory usage is quite reasonable -- about 4.5 to 7.5MB of RPRVT, 10MB RSIZE and 23-30MB RSHRD. However, the VSIZE is 8.36GB (giga, with a G). Compiling in 32bit mode drops VSIZE to 475MB, which is still an awful lot for such a small app. I know there are people who will say VSIZE doesn't matter if there is no paging, but if I ever sought to distribute my app people will not want to install a statusitem that looks like a memory hog. My question is two-fold: (1) How does the OS determine how much to allocate to VSIZE? and (2) is there something I am doing wrong in my app (likely) that is causing the out-sized allocation? The short version: Ignore the number under the VSIZE column. In fact, turn that column off. It has no practical meaning whatsoever. The long version: VSIZE is the total amount of address space that's mapped for your application. This includes things like libraries and shared memory, so it doesn't really correlate directly with *real* memory usage (which is tracked much more accurately by RSIZE). Beyond that, I understand that garbage collection also uses virtual memory in various clever ways. This is most noticeable under 64-bit builds, but I think it does similar things (albeit on a more limited scale) under 32-bit builds. If you're curious what all the virtual address space is being used for within an application, play around with the vmmap and vmmap64 utilities. But don't try to minimize address space usage unless you're actually low on it (in 32-bit apps). ___ 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: What determines VSIZE allocation?
On Jan 14, 2009, at 2:31 PM, Tobias Zimmerman wrote: I have searched for an answer to this question, including in Amit Singh's book, but have not found the answer (at least, not in a form that I recognize it). I have a fairly simple Cocoa app that creates an NSStatusItem with a small menu. It is compiled using Garbage Collection and in 64bit mode (it identifies in the Activity Monitor as Intel(64 bit). In non-debug mode it basically just sits there. The active memory usage is quite reasonable -- about 4.5 to 7.5MB of RPRVT, 10MB RSIZE and 23-30MB RSHRD. However, the VSIZE is 8.36GB (giga, with a G). Compiling in 32bit mode drops VSIZE to 475MB, which is still an awful lot for such a small app. I know there are people who will say VSIZE doesn't matter if there is no paging, but if I ever sought to distribute my app people will not want to install a statusitem that looks like a memory hog. My question is two-fold: (1) How does the OS determine how much to allocate to VSIZE? and (2) is there something I am doing wrong in my app (likely) that is causing the out-sized allocation? Nope -- you aren't doing anything wrong. VSIZE does not actually indicate the amount of memory allocated, but not necessarily used or paged out, for the application. A more accurate description is that VSIZE indicates the amount of address space the application has reserved. Under 64 bit, the garbage collector reserves 8GB of address space to ensure that the Auto Zone -- the allocation zone from which the collector allocates all objects -- remains contiguous up to 8GB of GC'd memory usage. As allocation requests are fulfilled, chunks of the address space are doled out to the program and utilized. Note that reserving address space comes nearly for free (there is a bit of administrative metadata). It is touching the memory -- writing data into it -- that is costly. The absolute VSIZE of your application isn't terribly useful. The change over time is, though. If the VSIZE is increasing relatively steadily over time *and* you don't know exactly why (there are legitimate reasons, especially under 64 bit), then it likely indicates that there is a memory leak in the form of, say, mapped files. In general, the RSIZE and RPRVT of your application are more immediately interesting. b.bum ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Text field with buttons like the To field in Mail
On Jan 14, 2009, at 08:30, Sergey Kuleshov wrote: I am very new to cocoa development and just making my first steps. What I am trying to do is something like a To field in Mail.app which provides autocomplete and once matched replaces the text with the button. After hours of googling it looks like I have to put NSButtonCell inside NSTextView (or NSTextField?) however, I can't understand how do it though it should be fairly common as I see same kind of behavior in many applications! (especially for list of tags) Maybe there is a ready made component? See: http://developer.apple.com/documentation/Cocoa/Reference/ApplicationKit/Classes/NSTokenField_Class/Reference/Reference.html The first sentence of this reads: NSTokenField is a subclass of NSTextField that provides tokenized editing similar to the address field in the Mail application. I have a feeling that's probably what you're looking for. ;) ___ 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
drag and drop with snapping
I'd like to implement drag and drop behavior with snapping to guides in my application. I have looked at both the NSDraggingSource and NSDraggingDestination protocols and there seems to be no obvious way of modifying the position of the dragged image. Do I need to make my own transparent window and handle all drag-drawing myself to accomplish this? Thanks, -- Francisco Tolmasky www.tolmasky.com tolma...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Dealing with exceptions in a drawing stack
Suppose I have a drawing stack that is called from a view's drawRect: method. This stack can end up quite deep, with objects calling other objects to implement parts of their drawing. Some of these methods will save and restore the graphics context. If an exception is thrown from somewhere down inside this stack, I'd prefer it if I could continue to draw what can be drawn, rather than have to abort the drawing altogether which usually leaves the whole view in a weird state (at which point I'm hosed - I have to abandon that document and create a new one, which is hardly user friendly). Short of using try/catch around every place where the graphics context is saved and restored, is there a way to unwind the graphics context stack or at least put it into a good known state? (i.e. discard all the saved states and start over). If so, I can then implement a try/ catch at a higher level and fix up the graphics context there. Anyone had some experience of handling this? --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/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Dealing with exceptions in a drawing stack
On Wed, Jan 14, 2009 at 11:34 PM, Graham Cox graham@bigpond.com wrote: If an exception is thrown from somewhere down inside this stack, I'd prefer it if I could continue to draw what can be drawn, rather than have to abort the drawing altogether which usually leaves the whole view in a weird state (at which point I'm hosed - I have to abandon that document and create a new one, which is hardly user friendly). This is known as a continuation. Scheme, for example, often uses them instead of exceptions. Much like a jmp_buf, It contains all of the information necessary to resume execution at the point the continuation is created. In the event of an error, an error handler is invoked with the continuation object, at which point the error handler can inspect the error condition and potentially modify the state such that execution can continue to proceed. It then executes the continuation and the program resumes where it left off. This isn't the model that exceptions follow, unfortunately. By the time your @catch block has caught the exception, the stack has already been unwound and you can't get it back. --Kyle Sluder ___ 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
what the hell is Error (1007) creating CGSWindow?
So I have an application that has been bedeviled by a nasty, nasty bug
for years that manifests itself as an exception being thrown with:
Error (1007) creating CGSWindow
At that point, the user has no choice but to reboot the machine - the
Window Server is pretty much hoarked.
Since the app is so complex, and the code path for tickling this bug
is so arbitrary, I've been loathe to file a bug on it, since I didn't
have a simple example to repro it.
Completely separately, I decided this AM to write a simple Cocoa app
to generate a bunch of test files for my group. Lo and behold, I
managed to tickle the bug.
Error (1007) creating CGSWindow
I just filed the radar bug as 6498067
but if anyone wants to take a look, I'm happy to send you the source
for the app - it's only 60KB, but the core of the thing looks like
this (it falls apart after making about 340 images, on my machine with
32GB of RAM, and a NVidia FX5600 w/1.5GB of VRAM):
where we're going from 0 to 1,000, and the width and height are
1920x1080 - seems to crap out after 340 files for me.
// eventually put this in a background thread with a cancel button on
a progress sheet
- (IBAction)generateFiles:(id)sender {
int start = [self.startingNumber intValue];
int howMany = 1 + [self.endingNumber intValue] - start;
NSSize imgSize = [self size];
int i, digits = [self.endingNumber digits];
int fontSize = [self determineFontSizeForNumber:self.endingNumber];
NSString* formatString = [NSString stringWithFormat:@%%0%dd,
digits]; // end up with %03d
NSRect rect = NSMakeRect(0, 0, imgSize.width, imgSize.height);
for (i = 0; i howMany; i++) {
NSAutoreleasePool* pool = [[NSAutoreleasePool alloc] init];
NSString* str = [NSString stringWithFormat:formatString, i];
NSFont* font = [NSFont systemFontOfSize:fontSize];
NSDictionary* dict = [NSDictionary
dictionaryWithObjectsAndKeys:font, NSFontAttributeName,
self.textColor, NSForegroundColorAttributeName, nil];
NSAttributedString* aStr = [[[NSAttributedString alloc]
initWithString:str attributes:dict] autorelease];
NSImage* image = [[NSImage alloc] initWithSize:imgSize];
[image lockFocus];
[aStr drawInRect:rect];
[image unlockFocus];
NSString* imgFormatString = [NSString stringWithFormat:@%...@_%
%0%dd.tif, digits]; // end up with %03d
NSString* imgName = [NSString stringWithFormat:imgFormatString,
self.prefix, i];
NSString* imgPath = [self.directory
stringByAppendingPathComponent:imgName];
NSData* tiffData = [image
TIFFRepresentationUsingCompression:NSTIFFCompressionLZW factor:1.0];
[tiffData writeToFile:imgPath atomically:YES];
[pool drain];
}
}
Thanks for any insight.
-- Michael B. Johnson, PhD
-- http://homepage.mac.com/drwave (personal)
-- http://xenia.media.mit.edu/~wave (alum)
-- MPG Lead
-- Pixar Animation Studios
___
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: what the hell is Error (1007) creating CGSWindow?
NSDictionary* dict = [NSDictionary dictionaryWithObjectsAndKeys:font, NSFontAttributeName, self.textColor, NSForegroundColorAttributeName, nil]; NSAttributedString* aStr = [[[NSAttributedString alloc] initWithString:str attributes:dict] autorelease]; NSImage* image = [[NSImage alloc] initWithSize:imgSize]; You're leaking this NSImage. -- Joe Ranieri ___ 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: what the hell is Error (1007) creating CGSWindow?
On Jan 14, 2009, at 9:13 PM, Joe Ranieri wrote: NSDictionary* dict = [NSDictionary dictionaryWithObjectsAndKeys:font, NSFontAttributeName, self.textColor, NSForegroundColorAttributeName, nil]; NSAttributedString* aStr = [[[NSAttributedString alloc] initWithString:str attributes:dict] autorelease]; NSImage* image = [[NSImage alloc] initWithSize:imgSize]; You're leaking this NSImage. sigh. I knew I should have explained this more. I know, in the code snippet I sent, that the NSImage is being leaked. But it's tickling the bug (where in the app, I have 100s of NSImages, and I need to keep them around). Why, on a machine with 32GB of RAM, and 1.5GB of VRAM, would having 350 1920x1080 NSImages live in the app be a problem? Make my app page, sure, but kill the window server so bad that it needs to get restarted? Not acceptable. -- Michael B. Johnson, PhD -- http://homepage.mac.com/drwave (personal) -- http://xenia.media.mit.edu/~wave (alum) -- MPG Lead -- Pixar Animation Studios ___ 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 exceptions in a drawing stack
On 15 Jan 2009, at 3:53 pm, Kyle Sluder wrote:
This is known as a continuation. Scheme, for example, often uses
them instead of exceptions. Much like a jmp_buf, It contains all of
the information necessary to resume execution at the point the
continuation is created. In the event of an error, an error handler
is invoked with the continuation object, at which point the error
handler can inspect the error condition and potentially modify the
state such that execution can continue to proceed. It then executes
the continuation and the program resumes where it left off.
OK, interesting...
This isn't the model that exceptions follow, unfortunately. By the
time your @catch block has caught the exception, the stack has already
been unwound and you can't get it back.
That's OK for the stack in general. The problem seems to be it leaves
the graphics context stack out of whack. I can obviously deal with
this by @catching wherever I have used saveGraphicsState and calling
restoreGraphicsState in my catch block before rethrowing, but it's
really clunky.
What I tried was this:
// at my very top level, i.e. [NSView drawRect:]
NSGraphicsContext* topContext = [[NSGraphicsContext currentContext]
retain];
@try
{
// make all my other drawing calls, any of which might save/restore
the graphics context but also could throw an exception
}
@catch( NSException* exc )
{
// we're back at the top, just ignore the exception here
}
[NSGraphicsContext setCurrentContext:topContext];
[topContext release];
Which actually seems to work - the view is no longer screwed up and
unusable following an exception. However, I'm not sure this is safe,
as it's blowing away all the stacked graphics contexts. While that's
what I want, I don't know if I'm allowed to just forget they ever
existed in this way.
--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/archive%40mail-archive.com
This email sent to arch...@mail-archive.com
Re: What determines VSIZE allocation?
On Wed, Jan 14, 2009 at 5:31 PM, Tobias Zimmerman automa...@gmail.com wrote: I know there are people who will say VSIZE doesn't matter if there is no paging, This is kind of like that famous question, Have you stopped beating your wife? VSIZE doesn't matter, period. Paging doesn't enter into it. but if I ever sought to distribute my app people will not want to install a statusitem that looks like a memory hog. The solution there is to educate the foolish people who think that VSIZE is any indication of being a memory hog, not to artificially reduce a number that has no bearing on any kind of system resource usage. Mike ___ 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: what the hell is Error (1007) creating CGSWindow?
On Thu, Jan 15, 2009 at 12:16 AM, Michael B Johnson w...@pixar.com wrote: On Jan 14, 2009, at 9:13 PM, Joe Ranieri wrote: NSDictionary* dict = [NSDictionary dictionaryWithObjectsAndKeys:font, NSFontAttributeName, self.textColor, NSForegroundColorAttributeName, nil]; NSAttributedString* aStr = [[[NSAttributedString alloc] initWithString:str attributes:dict] autorelease]; NSImage* image = [[NSImage alloc] initWithSize:imgSize]; You're leaking this NSImage. sigh. I knew I should have explained this more. I know, in the code snippet I sent, that the NSImage is being leaked. But it's tickling the bug (where in the app, I have 100s of NSImages, and I need to keep them around). Why, on a machine with 32GB of RAM, and 1.5GB of VRAM, would having 350 1920x1080 NSImages live in the app be a problem? Make my app page, sure, but kill the window server so bad that it needs to get restarted? Not acceptable. 350 32-bit images of that size works out to 2.7GB of memory. That's getting dangerously close to the limit of a 32-bit process. That limit is nominally 4GB, but system libraries decrease it significantly, and of course the window server has other stuff to keep track of than your images. NSImages created in this way use invisible windows which live in the window server, so the memory usage counts against it, not your app. When the window server runs out of address space and allocations begin to fail, it probably can't recover nicely and down you go. Now, should this happen? No. But UNIX is notoriously bad about recovering from low-resource situations. The standard answer to just about any resource allocation failure is to crash. Ideally the window server should not be killed by one runaway process, but such is not the world we live in. As to the solution, draw into something you *know* lives in your process and not on the window server, like an NSBitmapImageRep wrapped in an NSGraphicsContext. Then you'll still kill your program (unless you're building a 64-bit app), but you won't kill the system. Mike ___ 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: NIB's owner doesn't get released because of NSArrayController
Well, OK. I am ready to release all root objects myself. Just tested with subclassed NSView from that NIB - it didn't get released automatically. So I added several -release calls to -dealloc method of my NIB's owner and now it will release everything. But the problem is still there: -dealloc of my view controller didn't get called because array controller still retains it! As I wrote above, I added special killer method that breaks binding and after calling it I can call -release and it will call NIB owner's -dealloc. But this is ugly... On Thu, Jan 15, 2009 at 1:15 AM, Jason Foreman ja...@threeve.org wrote: On Wed, Jan 14, 2009 at 3:51 PM, Vitaly Ovchinnikov vitaly.ovchinni...@gmail.com wrote: Hello all, I have a custom-loaded NIB file with NSView and NSArrayController. Array controller is bound to file's owner (NSObject's derived class) and use some keypath for contentArray. The problem is when I close a window and main window controller tries to release this file's owner - array controller still holds a reference to it and doesn't allow to do that. So I have two questions: 1. What is the correct way to catch closing of the view to unbound array controller? 2. Do I really need to catch something? Maybe there is a simpler solution? Any ideas? Top-level objects in a nib need to be released. You can do this in the File's Owner object when the window is closed, for example. See this page for some additional info on nib object memory management: http://developer.apple.com/documentation/Cocoa/Conceptual/MemoryMgmt/Articles/MemMgmtNibObjects.html Jason ___ 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/vitaly.ovchinnikov%40gmail.com This email sent to vitaly.ovchinni...@gmail.com ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Quick (?) NSCell question: getting the drawing (inner) rectangle
I'm trying to determine what a cell's drawing rect will be (particularly the 'interior' size eventually expected to be sent to - drawInteriorWithFrame:inView: when the cell is actually drawn). I can get the 'prototypical' cell object by asking for the pertinent NSTableView column, and then asking for the cell. I can see that NSCell supports a range of methods under the heading Determining Cell Size in the docs. The method -drawingRectForBounds: looks promising, but it takes another rectangle, i.e. the bounds. Assuming that this method would give me what I want, how do I first obtain the appropriate 'bounds' to feed it? Perhaps this method is only intended to provide information to the system by the NSCell implementation and is not intended for the purpose I need. In which case, is there a way to achieve what I want without having to ask for the NSTableColumn width and working out what proportion of this width is used for non-'interior' furniture (if any)? Cheers -- Luke ___ 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
Center image in NSTableView header
I am creating an NSAttributedString containing an image in an NSTextAttachment, setting an NSTableView headerCell to the attachmentCell. This works great, the image is shown nicely on the table view column header. But I am unable to get the image centered in the column header. Anyone have a hint to how I can do that? Here is my code: NSImage *warningImage = [NSImage imageNamed:@danger]; NSTextAttachment *ta = [[[NSTextAttachment alloc] init] autorelease]; [(NSCell *)[ta attachmentCell] setImage:warningImage]; NSAttributedString *headerAS = [NSAttributedString attributedStringWithAttachment:ta]; [[[oactiveElementsTable tableColumnWithIdentifier:@warningword] headerCell] setAttributedStringValue:headerAS]; Ivan ___ 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: Localized resources not loading from external frameworks
On Jan 13, 2009, at 6:03 PM, Samir Patel wrote: All the files are in the right locations, but when I run my app which is linked against the framework, the English version of my UI from the framework always loads no matter what language I have set in System Preferences. Any ideas how to get the localized resources to load? Which macro are you using? You must use NSLocalizedStringFromTableInBundle() within non-app bundles and frameworks, and the NSBundle passed in must point to the bundle holding the localized strings. Nick Zitzmann http://www.chronosnet.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: What determines VSIZE allocation?
On Jan 14, 2009, at 9:40 PM, Michael Ash wrote: On Wed, Jan 14, 2009 at 5:31 PM, Tobias Zimmerman automa...@gmail.com wrote: I know there are people who will say VSIZE doesn't matter if there is no paging, This is kind of like that famous question, Have you stopped beating your wife? VSIZE doesn't matter, period. Paging doesn't enter into it. but if I ever sought to distribute my app people will not want to install a statusitem that looks like a memory hog. The solution there is to educate the foolish people who think that VSIZE is any indication of being a memory hog, not to artificially reduce a number that has no bearing on any kind of system resource usage. Generally true (though I might choose a slightly different descriptor of the uselessness), but not always. VSIZE *can* be a very useful indication that an application is consuming address space. This is not the same as consuming memory and, to the user, is an utterly useless distinction to make. For example, an application's VSIZE might be growing over time because it is mmap()'ing a bunch of files (or a few small files). If the app fails to unmap, the VSIZE will grow and the app may likely exhaust its address space without any paging activity. Example: for applications that are processing large files -- ID3 tag editors come to mind -- watching the VSIZE can be a useful way of determining if your code is properly managing the mapping of said files. b.bum ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
completely messed up with bindings...
Hello all, I have array of Filter objects in my document. Filters are used in two ways: 1. I can select one of them to filter my data 2. I can browse them all and add, delete or/*modify any of them My document exposes this array via -filters/-setFilters pair. I have two NSArrayControllers: one for the first task and another for the second task. Main window has NSPopUpButton that is bound to the first array controller. It is used to select filter. Window controller receives messages from this button and call -setFilter method of my document. Document filters data and all works fine, but I think there is a simpler way to connect my document's -filter property and to controller's selection. Is it possible? This is the first question. Other array controller used for editing filters. I have NSTableView bound to that controller that shows filter names and NSPredicateEditor that is bound to selection of the second array controller. It is used to edit filter. All works fine, but I can't receive a notify that current filter was changed. NSPopUpButton somehow knows that name of filter was changed and displays it immediately. But my document's -setFilter method didn't get called. -setFilters method didn't get called too. I tried to bind filter property of my document to the first controller's selection but it didn't work too. I wrote something like this in the main window controller: [[self document] bind:@filter toObject:pFiltersController1 withKeyPath:@selection options:nil] I even added +initialize method to the document and exposed filter binding. Nothing... Just some errors in log about NSControllerObjectProxy predicate: unrecognized selector sent to instance .. So the questions are: 1. How to bind some property of my document to selected object of NSArrayController? Or at least how to monitor changes of the selected object? 2. What is the correct way to do what I want (two different types of operations on one array)? Thank you. ___ 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
