Re: Stealing settings from Mail.app
Thanks, it was the iTools bit I needed. But looks like there is a bug in this
code it won't notice me.com and could wrongly think say macdonald.com is mobile
me. But thanks, it gives me what I needed.
--
Chris
On 31/05/2010, at 8:11 PM, Dante Palacios wrote:
NSString *GetPassword(NSString *userName, NSString *hostName) {
if (!hostName.length || !userName.length) return nil;
const char *serviceName = [hostName UTF8String];
const char *serviceUserName = [userName UTF8String];
const char *path = "";
UInt32 passwordLength = 0;
char *pass = nil;
OSStatus status = noErr;
SecKeychainItemRef itemRef;
if ([hostName rangeOfString:@"mac"].location != NSNotFound) {
serviceName = "iTools"; //<= note this
status = SecKeychainFindGenericPassword(NULL,
(UInt32)strlen(serviceName),
serviceName,
(UInt32)strlen(serviceUserName),
serviceUserName,
&passwordLength,
(void **)&pass,
&itemRef);
}else {
status = SecKeychainFindInternetPassword(NULL,
(UInt32)strlen(serviceName),
serviceName,
0,
NULL,
(UInt32)strlen(serviceUserName),
serviceUserName,
(UInt32)strlen(path),
path,
0,
kSecAuthenticationTypeAny,
kSecAuthenticationTypeAny,
&passwordLength,
(void **)&pass,
&itemRef);
}
if (status != noErr) {
CFStringRef errMess = SecCopyErrorMessageString(status, NULL);
NSLog(@"%@", (NSString *)errMess);
CFRelease(errMess);
return nil;
}
return [[[NSString alloc] initWithBytes:pass length:passwordLength
encoding:NSUTF8StringEncoding] autorelease];
}
On May 31, 2010, at 12:11 AM, Jn wrote:
Would you have the API code to retrieve it? Not sure what to pass to get the
mobile me password.
--
Chris
On 31/05/2010, at 2:43 AM, Jens Alfke wrote:
On May 30, 2010, at 12:55 AM, Chris Idou wrote:
OK, I see. So is Mobile-Me the only special case, or is there a more general
rule about other places to find smtp server passwords?
AFAIK it’s the only special case. In general, an SMTP server’s password is
stored under that server name in the keychain, as you’d expect.
—Jens
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/palacios.dante%40gmail.com
This email sent to palacios.da...@gmail.com
All the best,
Dante.
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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
Can't set header on NSURLRequest
Hello, I'm having an issue with setting two headers for my NSURLRequest: [theRequest setValue: [NSString stringWithFormat:@"/principals/ __uids__/%@/\r\n",self.userGUID] forHTTPHeaderField:@"Originator"]; [theRequest setValue: [NSString stringWithFormat:@"/principals/ __uids__/%@/\r\n",roomGUID] forHTTPHeaderField:@"Recipient"]; When I send the request they don't appear in the headers and the server fails to respond correctly as they are required headers. I also tried addValue:forHTTPHeaderField, with the same result. If I set them without the NSString (@"string...") it works fine. Am I missing something here? 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
Calendar Store- iCal Notifications
I know you can use the Calendar Store API to receive notifications if iCal data has changed. Can you use the same API to receive notifications if the ical user has received a notification related to a calDAV account (i.e. new meeting invitations, rejected meeting invitations, etc.). Thanks! --Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Cookies Problems
Hello
I am trying to write a test program that will send a value through a
web form, get the cookie information and then relay that cookie
information back to the sever. I have setup 3 php pages, which work
correctly. The first page displays a web form with one field named
"name". The second page "validates" the input, puts the "name"
variable into a session variable, then forwards you to the third page.
The third page checks for the session variable and then either
displays "You are logged in as: Bob" (using the $name / session
variable) or displays "failure". Each page also displays the SessionID
at the top of the screen.
When I run these PHP pages through a browser, it works as expected.
When I try to run this through Cocoa, I am able to retrieve the
Session Cookie, but when I call the data from the third page, I always
get a "failure" response, even though it seems to be relaying the
correct Session ID back to the server.
I have found many different articles online, but they all seem to be
giving the same basic code, which doesn't seem to work. I've tried
several different variations but I'm not sure where to go from here.
You can access the form at http://www.abovetheclamor.com/testform.php
Here are all the files I'm using:
### PHP 1 (testform.php) ###
';
echo '
';
?>
### PHP 2 (testform2.php) ###
';
$_SESSION['sessionvariable'] = $_POST['name'];
header("Location: testform3.php");
?>
### PHP 3 (testform3.php) ###
';
if ($_SESSION['sessionvariable'] == NULL) {
echo 'failure on page 3';
exit;
} else {
echo 'You are logged in as: ';
echo $_SESSION['sessionvariable'];
}
?>
### Cocoa / Objective-C ###
- (void)applicationDidFinishLaunching:(UIApplication *)application {
NSHTTPURLResponse *response;
NSError *error;
NSString *post = @"name=Bob"; // send the value "Bob" to the
form
NSData *postData = [post
dataUsingEncoding:NSASCIIStringEncoding allowLossyConversion:YES];
NSString *postLength = [NSString stringWithFormat:@"%d",
[postData length]];
NSMutableURLRequest *request = [[[NSMutableURLRequest alloc]
init] autorelease];
// setup the request to submit the data to the form
[request setURL:[NSURL URLWithString:@"http://www.abovetheclamor.com/testform.php
"]];
[request setHTTPMethod:@"POST"]; // run it as a POST
[request setValue:postLength forHTTPHeaderField:@"Content-
Length"];
[request setValue:@"application/x-www-form-urlencoded"
forHTTPHeaderField:@"Content-Type"];
[request setHTTPBody:postData];
// run the request
NSURLConnection *theConnection = [[NSURLConnection alloc]
initWithRequest:request delegate:self];
// run the request to get the cookies
[NSURLConnection sendSynchronousRequest:request
returningResponse:&response error:&error];
// get allCookies from the request
NSArray *allCookies = [NSHTTPCookie
cookiesWithResponseHeaderFields: [response allHeaderFields]
forURL:[NSURL URLWithString:@"http://www.abovetheclamor.com/testform.php
"]];
NSLog(@"How many Cookies: %d", allCookies.count);
for (NSHTTPCookie *cookie in allCookies) {
NSLog(@"Name: %@ : Value: %@, Expires: %@",
cookie.name, cookie.value, ookie.expiresDate);
}
[[NSHTTPCookieStorage sharedHTTPCookieStorage]
setCookies:allCookies
forURL:[NSURL URLWithString:@"http://www.abovetheclamor.com/";]
mainDocumentURL:nil];
NSArray * availableCookies = [[NSHTTPCookieStorage
sharedHTTPCookieStorage] cookiesForURL:[NSURL URLWithString:@"http://www.abovetheclamor.com/
"]];
NSDictionary * headers = [NSHTTPCookie
requestHeaderFieldsWithCookies:availableCookies];
NSMutableURLRequest *req = [[NSMutableURLRequest alloc] init];
// set the request to call the final page
[req setURL:[NSURL URLWithString:@"http://www.abovetheclamor.com/
testform3.php"]];
[req setAllHTTPHeaderFields:headers];
[req setHTTPShouldHandleCookies:NO];
// send the request to the final page
NSData *data = [NSURLConnection sendSynchronousRequest:req
returningResponse:&response error:&error];
// what did the server get? probably "failure". why?
NSLog(@"The server saw: %@", [[[NSString alloc]
initWithData:data encoding: NSASCIIStringEncoding] autorelease]);
[window makeKeyAndVisible];
}
Any help or guidance that you can provide would be greatly appreciated
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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: Apple Developer Update
Vincent, I respectfully disagree. If we encourage anyone who feels entitled to "discharge overflow of emotional frustration" to vent here, what's to stop anyone who disagree with Apple's policy to spam our mailboxes with tirades. I am frustrated too, but do something constructive. Blogging is a soapbox for anyone with an opinion or need to vent. – Chris On Jul 25, 2013, at 15:56, Vincent Habchi wrote: > Kyle, > >> Following that line of thought, how many of you actually think griping >> on this list is going to accomplish anything other than filling up >> everyone else's inboxes from what would otherwise be one of the >> remaining useful channels? > > While I agree with you, I think it is also essential to keep a thread open to > sort of discharge the overflow of emotional frustration – we’re not Vulcans! > :) I admit it is an off-topic and maybe off-color and whatever-off-you-like > thread, but it acts as a safety valve and helps people overcome this > unexpected downtime. And since, as you clearly state, this list is one of the > few remaining open media… > > Cheers! > Vincent > > > > > ___ > > Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) > > Please do not post admin requests or moderator comments to the list. > Contact the moderators at cocoa-dev-admins(at)lists.apple.com > > Help/Unsubscribe/Update your Subscription: > https://lists.apple.com/mailman/options/cocoa-dev/mailbox20040630%40gmail.com > > This email sent to mailbox20040...@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: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
NSPredicateEditorRowTemplate
I've got a NSPredicateEditor and I'm inheriting from NSPredicateEditorRowTemplate to make a custom template. It implements copyWithZone as the doco seems to imply I should. In Interface builder I have a number of standard row templates, and I've added my custom one at the end by setting the class in the identity pane. When I load the nib containing the NSPredicateEditor, for some reason it calls copyWithZone on my custom NSPredicateEditorRowTemplate during the nib loading process. Let's call this object passed to copyWithZone as object "A", and I return from copyWithZone as instance "B". So now there are two instances created during the nib loading process. Nothing in the documentation seems to explain why this would happen during nib loading. From my reading of the doco, it should only get called when I do a setObjectValue on the NSPredicateEditor. Anyway, when I call setObjectValue, it does indeed call copyWithZone again with object "A" and creates another object. Let's call this one "C". It then calls setPredicate on object "C" as you'd expect, whereupon I populate the fields of MyCustomPredicateRowTemplate. When the user clicks ok, then I call objectValue on the NSPredicateEditor and it calls predicateWithSubpredicates not on object "C", but on object "B", which is always going to be blank, because it is in fact object "C" which is the one displayed. Thus I can never retrieve values from the row. This phantom and unexplained object "B" that is created during NIB loading suddenly seems to be the one it cares about. I'm not sure where to go next. Has anyone got any advice? ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Allocating outlets from NIB file
Hi,
You should not be allocating either SpriteView or SpriteController if
they are referred to in the NIB. (which is the normal case).
Instead you go to the File's Owner object in interface builder, and go
to the Identity tab, and set the Class to be whatever class contains
your loadNib statement. ONLY load the NIB, don't attempt to create
windows, views or controllers. (Let's call this class that loads the
NIB as class "A".). Now you add some outlets to class "A" called
spriteView, spriteWIndow, spriteController or whatever else you need.
Then in interface builder you connect these outlets of "File's Owner"
to those objects. Now when you load the NIB, you'll have pointers to
all the objects you care about in the same object that loaded the NIB.
On 24/06/2008, at 10:17 PM, Joseph Ayers wrote:
I am quite confounded with regard to how/when to allocate outlets
which are classes existing as instances in
another class. Consider
@interface SpriteController : NSWindowController {
IBOutlet SpriteView* spriteView;
IBOutlet NSWindow* spriteWindow;
}
SpriteController* spriteController;
- (id)spriteWindow;
- (id)spriteView;
-(void)setSpriteView:(SpriteView*)view;
}
SpriteView is defined in another subclass as:
@interface SpriteView: NSView
{
NSImage *spriteImage;
}
SpriteController and SpriteView are defined and connected in the NIB
When I open the NIB with
-(void)loadSpriteController{
if (spriteController == NULL) {
spriteController = [[SpriteController alloc] init];
if (![NSBundle loadNibNamed:@"spriteWindow"
owner:spriteController]) {
NSLog(@"Error loading SpriteController");}
else{
NSLog(@"SpriteController NIB Loaded"); }
}
}
spriteController gets a pointer, but spriteView is NIL. All
subsequent messages to spriteView are
messages to NIL
How/where should I be allocating spriteView. I've tried adding:
[self setSpriteView: [[SpriteView alloc] init]];
to loadSpriteController with no consequence. spriteView gets a
pointer but
[[spriteController spriteView] drawRect:[[spriteController
spriteWindow] bounds]];
does not get to drawRect
ja
--
Joseph Ayers, Professor
Department of Biology and
Marine Science Center
Northeastern University
East Point, Nahant, MA 01908
Phone (781) 581-7370 x309(office), x335(lab)
Cellular (617) 755-7523, FAX: (781) 581-6076 Boston Office 444RI,
(617) 373-4044
eMail: [EMAIL PROTECTED]
http://www.neurotechnology.neu.edu/
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/idou747%40gmail.com
This email sent to [EMAIL PROTECTED]
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
Re: Seconds since system startup
Do a man 3 sysctl in the terminal and look for KERN_BOOTTIME On 24/06/2008, at 10:51 PM, Stefan Hafeneger wrote: Hi, I'm looking for a Cocoa function like GetCurrentEventTime() for Carbon to get the interval since system startup. Any ideas? With best wishes, Stefan___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/idou747%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: NSPredicateEditorRowTemplate
Hi! This is very interesting information. Wish it was in the doco! I have a custom view which wasn't responding to setObjectValue / objectValue. When I add those methods I find that on startup it does indeed copy the values from object "C" to object "B". This means that when I retrieve the values later on, instead of returning blank like before, it now returns the old value instead of blank. However I need the new value! If it were to copy the values across AFTER the user made changes or prior to me calling objectValue, then it would work. I thought maybe [rulePredicateEditor reloadPredicate] sounded like it might do it perhaps, but that doesn't help either. Once the user hits ok, we are still left with bogus values from object "B", albeit now old values instead of nil values. On 24/06/2008, at 11:02 PM, Jim Turner wrote: On Tue, Jun 24, 2008 at 2:02 AM, Chris <[EMAIL PROTECTED]> wrote: When the user clicks ok, then I call objectValue on the NSPredicateEditor and it calls predicateWithSubpredicates not on object "C", but on object "B", which is always going to be blank, because it is in fact object "C" which is the one displayed. Thus I can never retrieve values from the row. This phantom and unexplained object "B" that is created during NIB loading suddenly seems to be the one it cares about. I'm not sure where to go next. Has anyone got any advice? Hi Chris, I saw this very same behavior and was fortunate enough to get the following explanation from Peter Ammon: "A single row may be composed of views from multiple templates. When it's time to construct a predicate for that row, we pick one template, and if its views are not actually in the row, we call objectValue on the view in the row, and then setObjectValue: on the corresponding view in your template. Due to a bug, this happens more often than it should :( What this means for your NSTextField subclass is that it should do the right thing for objectValue and setObjectValue:. The object value of your view should encapsulate all the state your template needs to compute that portion of the predicate." The short answer, for me at least, was to make sure my custom NSTextField in my template handled objectValue/setObjectValue: properly. That way, when predicateWithSubpredicates: is called, the internals of the editor can pass around the values needed to properly compute the predicate. -- Jim http://nukethemfromorbit.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 [EMAIL PROTECTED]
Re: NSPredicateEditorRowTemplate
Hi,
No, I'm doing that all correctly. But now I seem to have changed
something minor, but I'm not sure what, and now its working. Now its
copying the object across correctly when I call objectValue.
Anyway, thanks for the setObjectValue tip which was the key. I never
would have guessed that one.
On 25/06/2008, at 12:52 AM, Jim Turner wrote:
On Tue, Jun 24, 2008 at 8:46 AM, Chris <[EMAIL PROTECTED]> wrote:
Hi!
This is very interesting information. Wish it was in the doco!
I have a custom view which wasn't responding to setObjectValue /
objectValue.
When I add those methods I find that on startup it does indeed copy
the
values from object "C" to object "B". This means that when I
retrieve the
values later on, instead of returning blank like before, it now
returns the
old value instead of blank. However I need the new value! If it
were to copy
the values across AFTER the user made changes or prior to me calling
objectValue, then it would work. I thought maybe [rulePredicateEditor
reloadPredicate] sounded like it might do it perhaps, but that
doesn't help
either. Once the user hits ok, we are still left with bogus values
from
object "B", albeit now old values instead of nil values.
How are you accessing your custom view's object value? It almost
sounds like you're asking your original custom view for it's value
each time instead of the object currently being displayed.
When you create your template and insert your custom view, make sure
to keep a reference to that specific object so you can query it later
on.
@interface CustomPredicateEditorRowTemplate :
NSPredicateEditorRowTemplate
{
CustomTextField *myTextField;
}
-(CustomTextField *) myTextField;
@end
@implementation CustomPredicateEditorRowTemplate
-(CustomTextField *) myTextField
{
if( !myTextField )
// init your view
return( myTextField );
}
- (NSArray *)templateViews
{
return( [[super templateViews] arrayByAddingObject:[self
myTextField]] );
}
-(void) dealloc
{
[myTextField release];
[super dealloc];
}
- (NSPredicate *)predicateWithSubpredicates:(NSArray *)subpredicates
{
id objectValue = [[self myTextField] objectValue];
// Do magical things with objectValue
}
@end
--
Jim
http://nukethemfromorbit.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 [EMAIL PROTECTED]
NSPredicateEditor
Let's say I create a NSPredicateEditor and it looks like this: [All] of the following are true: [Name] equals [ ] --- So the user enters say "Fred" and the predicate returned is "Name == Fred". Later on, I reload that predicate into the NSPredicateEditor and it looks like this: [Name] equals [ Fred ] --- But now there is no option to change it to say: [Some] of the following are true: because that line isn't shown. How do I make that line reappear the next time? ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: NSPredicateEditor
Cool. Now I notice that if you manually plug in a complex predicate like "a = b or c = d and e = f", that it is capable of displaying it correctly. Do you know if there is any way to allow the user to create more complex expressions? By default it only seems to allow either AND or OR, but not both from the user, even though the machinery seems to know how to display things more complex. On 25/06/2008, at 11:57 PM, Jim Turner wrote: On Wed, Jun 25, 2008 at 1:08 AM, Chris <[EMAIL PROTECTED]> wrote: Let's say I create a NSPredicateEditor and it looks like this: [All] of the following are true: [Name] equals [ ] --- So the user enters say "Fred" and the predicate returned is "Name == Fred". Later on, I reload that predicate into the NSPredicateEditor and it looks like this: [Name] equals [ Fred ] --- But now there is no option to change it to say: [Some] of the following are true: because that line isn't shown. How do I make that line reappear the next time? By default, the NSPredicateEditor object in IB creates a compound predicate. Wrap your "Name == Fred" predicate in a compound predicate before reloading it in the editor and the Any/All/None row should return. -- Jim http://nukethemfromorbit.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 [EMAIL PROTECTED]
NSPredicateEditor
Consider the following code, any NSPredicateEditor gurus: NSPredicateEditorRowTemplate *tmpTemplate = [[NSPredicateEditorRowTemplate alloc] initWithCompoundTypes: [NSArray arrayWithObjects: [NSNumber numberWithInt: NSNotPredicateType], nil]]; NSPredicate * tmpPredicate = [NSPredicate predicateWithFormat:@"mykey = 'foo'"]; NSPredicate * tmpCompPredicate = [NSCompoundPredicate notPredicateWithSubpredicate:tmpPredicate]; double result = [tmpTemplate matchForPredicate:tmpCompPredicate]; NSLog(@"RESULT: %f", result); This code prints 0.0. The net effect is that NSPredicateEditor can't display a predicate like NOT (foo = "bar") A bug in NSPredicateEditor system perhaps? But surely someone would have seen it before. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: NSPredicateEditor
On Fri, Jun 27, 2008 at 10:09 AM, Peter Ammon <[EMAIL PROTECTED]> wrote:
>
> On Jun 25, 2008, at 7:27 PM, Chris wrote:
>
>>
>> The net effect is that NSPredicateEditor can't display a predicate like
>>
>> NOT (foo = "bar")
>>
>>
>> A bug in NSPredicateEditor system perhaps? But surely someone would have
>> seen it before.
>>
>
> Hi Chris,
>
> NOT type compound predicates only support exactly one subpredicate, or at
> least they did when NSPredicateEditor was written. For this reason, the
> default implementation of NOT expects the sole subpredicate to be an OR
> type. So set this:
>
> NOT(OR(foo="bar"))
Ahh, this would explain why it hasn't been reported before.
However the BNF description of predicates does not impose that restriction:
http://developer.apple.com/documentation/Cocoa/Conceptual/Predicates/Predicates.pdf
And also, when you convert the predicate to a string, it doesn't show the
OR. The OR is apparently optimized away by the predicateFormat routine.
Since the only way to feasibly store a predicate is as a string, and then
parse it back in, its not really consistent.
I've worked around it by replacing the matchForPredicate routine of
NSPredicateEditorRowTemplate, which is where I think the actual problem
lies:
- (double)matchForPredicate:(NSPredicate *)predicate {
if ([predicate class] == [NSCompoundPredicate self]) {
return 0.5;
}
return 0.0;
}
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
NSExpression
NSExpression defines this method: + (NSExpression *)expressionForFunction:(NSString *)name arguments: (NSArray *)parameters and the doco provides this example: [NSExpression expressionForFunction:(@selector(random)) arguments:nil]; Isn't that wrong? Can you really pass a selector to a NSString argument? The compiler is complaining, and the program crashes when I attempt 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 [EMAIL PROTECTED]
Re: NSExpression
If anyone has a clue how to use it, I'd be grateful. This was my unsuccessful attempt: NSExpression * ex = [NSExpression expressionForFunction:[NSExpression expressionForConstantValue:@"BAR"] selectorName:@"length" arguments:nil]; NSPredicate * predicate = [NSCompoundPredicate andPredicateWithSubpredicates:[NSArray arrayWithObject: ex]]; [predicate evaluateWithObject:@"FOO" substitutionVariables:nil]; [NSFunctionExpression evaluateWithObject:substitutionVariables:]: unrecognized selector sent to instance 0x68b8af0 On 29/06/2008, at 12:43 AM, Shawn Erickson wrote: On Jun 28, 2008, at 12:13 AM, Chris <[EMAIL PROTECTED]> wrote: NSExpression defines this method: + (NSExpression *)expressionForFunction:(NSString *)name arguments: (NSArray *)parameters and the doco provides this example: [NSExpression expressionForFunction:(@selector(random)) arguments:nil]; Isn't that wrong? Can you really pass a selector to a NSString argument? The compiler is complaining, and the program crashes when I attempt. Yeah it is wrong. I believe you just pass a string with the same contents as what you would put in @selector. -shawn ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Disabling column selection NSTableView
Hi, I've disabled column sorting. When I click on the header it seems to select the entire row. I want to disable this functionality, partly because it is meaningless for my app, and partly because I would like to make doubleclick action on the header do something to the selected row, and since single click on the header unselects the selected row, that isn't working. How do I disable single click on the header selecting the entire row? ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
NSPredicateEditorRowTemplate
I have a custom view in a NSPredicateEditorRowTemplate. It seems like the NSPredicateEditor isn't aware when I change the state of the custom view, because when I ask it for the predicate it just returns the old one without even calling predicateWithSubPredicates on my template. I'm guessing that NSPredicateEditor installs some kind of callback on the components and needs to know when they change, but being inexperienced with Cocoa, I'm not sure what it might be or how to implement it. I tried calling sendActionTo: with the action on the custom component when it changes, but that didn't seem to help. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: IB3 and custom controls: problem with actions
I think I've seen this a few times in IB3. But I fiddled a bit, saved, restarted and then it was working as I expected. I wrote it off to a bug at the time. On 04/07/2008, at 4:19 PM, Vitaly Ovchinnikov wrote: Hello all, I'm porting my project from XCode 2.5 to 3.0 and have a big trouble with custom controls. I have some custom controls (subclassed from NSControl), that I use in Interface Builder to create GUI. When I used IB 2.5, I placed "Custom View" object to the window, changed it's class and then I was able to Ctrl+Drag an action to controller. After upgrade to XCode 3.0 I found out that now it is impossible. Old links are still alive (marked with yellow triangles), but I can't create a new one. I don't see "Sent action" group by right-clicking my controls. If I perform Ctrl+Drag from my control to the controller - it doesn't highlited. But if I do the same for NSButton - all works fine. Did I miss something? 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/idou747%40gmail.com This email sent to [EMAIL PROTECTED] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
NSTableView Issue
Hello,
This is an issue I've been stuck on for a while, I'm trying to load an
instance of NSHTTPCookieStorage into a table view. I have it far
enough so that when I build the app and load it, it displays this http://i38.tinypic.com/315en3c.png
. However when I attempt to scroll up or down, the app does nothing
and I receive these error message in the console:
*** -[NSCFArray objectAtIndex:]: index (2) beyond bounds (2)
and
[ valueForUndefinedKey:]: this class is not key
value coding-compliant for the key name.
Here are my two table methods for reference:
- (int)numberOfRowsInTableView:(NSTableView *)tableView
{
return ([cookies count]);
}
- (id)tableView:(NSTableView *)tableView
objectValueForTableColumn:(NSTableColumn *)tableColumn
row:(int)row
{
NSHTTPCookie *cookie = [cookies objectAtIndex: row];
NSString *identifier = [tableColumn identifier];
return [cookie valueForKey: identifier];
}
Thank you, any guidance would be appreciated!
--Chris
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
Creating a NSHTTPCookie
Hello, I'm trying to create a NSHTTPCookie with this code: //dictionary of attributes for the new cookie NSDictionary *newCookieDict = [NSMutableDictionary dictionaryWithObjectsAndKeys:@".example.com", NSHTTPCookieDomain, @"Test Cookie", NSHTTPCookieName, @"/", NSHTTPCookiePath, @"test12345", NSHTTPCookieValue, @"2010-05-03 21:02:41 -0700", NSHTTPCookieExpires, nil]; //create a new cookie NSHTTPCookie *newCookie = [NSHTTPCookie cookieWithProperties:newCookieDict]; //add the new cookie [[NSHTTPCookieStorage sharedHTTPCookieStorage] setCookie:newCookie]; When I try to set the newly created cookie to the sharedHTTPCookieStorage it doesn't ever get set. Am I doing something wrong? 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
Creating a NSHTTPCookie
Hello, I'm trying to create a NSHTTPCookie with this code: //dictionary of attributes for the new cookie NSDictionary *newCookieDict = [NSMutableDictionary dictionaryWithObjectsAndKeys:@".example.com", NSHTTPCookieDomain, @"Test Cookie", NSHTTPCookieName, @"/", NSHTTPCookiePath, @"test12345", NSHTTPCookieValue, @"2010-05-03 21:02:41 -0700", NSHTTPCookieExpires, nil]; //create a new cookie NSHTTPCookie *newCookie = [NSHTTPCookie cookieWithProperties:newCookieDict]; //add the new cookie [[NSHTTPCookieStorage sharedHTTPCookieStorage] setCookie:newCookie]; When I try to set the newly created cookie to the sharedHTTPCookieStorage it doesn't ever get set. Am I doing something wrong? 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
Re: Getting key data out of the keychain
> On 1 Jan 2016, at 13:09, Andreas Mayer wrote: > > But I *still* don't know how to get at the key bytes of a SecKeyRef. :P Try asking on the apple-cdsa mailing list. It covers the security frameworks in OS X, including (hence the historical name) CDSA. Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Deleting an file that's been NSData memory mapped - safe?
> On 14 Feb 2016, at 11:45, dangerwillrobinsondan...@gmail.com wrote: > > Would the file itself be accessible by another process before your process > exits? Only if it manages to open it before it gets unlinked. After it gets unlinked, there’s no way to open it because there’s no longer any associated filename. Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Where are my bytes hiding?
> On 5 May 2016, at 20:01, Martin Wierschin wrote: > >> Those files are compressed by the filesystem. In HFS+/MacOS Extended that >> means that the data fork is empty and the file contents are stored in the >> resource fork or extended attributes structure. >> >> http://wiki.sleuthkit.org/index.php?title=HFS#HFS.2B_File_Compression > > Huh, that's interesting and surprising, thanks for the link. Is this method > of stashing compressed data in the xattrs something that's currently commonly > used by OSX? Or is it just some weird infrequently used trick? I see this on > the linked page: > >> Compression is most often used for files installed as part of Mac OS X; user >> files are typically not compressed (but certainly can be!) > > Are a lot of system files compressed like this? Is there any way a user file > might be compressed in such a way through normal user actions? You can do it explicitly using /usr/bin/ditto. Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: NSFileWrapper
> On 4 Jun 2016, at 23:02, Daryle Walker wrote: > >> On Jun 4, 2016, at 12:18 PM, Peter Hudson wrote: >> >> Hi Mike >> >> You're right - I want the behaviour of a package. Do you happen to know how >> I can achieve it? I had presumed it might be some mix of flag setting at >> system level. If for example i set the extension on an NSFileWrapper to >> .bundle i get some of what I need - but I cant find out how to get an icon >> of my choosing associated with it. > > You’ll have to work at it; it won’t be completely transparent to your users > to experience nor you to program. > > The “.bundle” extension is already in use for system plug-ins. You have to > come up with a new extension. For instance, if you’re already using “.xyz” > for you single-file format, you could use “.xyz-pkg” for your package format. Are you sure that’s true? Apps like OmniGraffle have a flat file format *and* a bundle format, both using the .graffle extension. In OmniGraffle Pro's document inspector you can switch between formats. If you have the app or a demo you can investigate its Info.plist. Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: "Live" NSFetchRequest?
Are you asking about an NSFetchRequest that notices when objects are added to or changed in your persistent store by another process, or by another thread in the same process? In the latter case you can use the notifications posted by the NSManagedObjectContext that's being saved, and use that to refresh the objects in the context you're using for your human interface. Multi-process coordination is a much harder problem. -- Chris > On Sep 17, 2016, at 2:37 PM, Frank D. Engel, Jr. wrote: > > Before I go reinventing the wheel to try to code around this... > > Is there any way I can set up a "live" NSFetchRequest in a Core Data model, > so that is matches against the predicate change (for example, predicate > filters on a field, the value of that field changes such that an object > suddenly matches that did not before, or no longer matches when it previously > did), I can get some sort of notification of the change - or even that there > /was/ a change? > > > Notes: > > * The predicate is being created by the user using an NSPredicateEditor and I > am allowing a large selection of fields. > > * The data set could grow to include several thousand objects (maybe > tens/hundreds of thousands in some cases) > > * I am using the result set in "real time" in some cases and need to know > almost immediately when a change takes place - not a good thing to be > rerunning the query multiple times per second if I don't know that a change > happened - and changes to some of the attributes could be happening multiple > times per second in some cases. > > * Some of the attributes are calculated from other attributes, but I am > updating them in such a way that bindings fire. > > * MacOS X, not iDevices > > > ___ > > Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) > > Please do not post admin requests or moderator comments to the list. > Contact the moderators at cocoa-dev-admins(at)lists.apple.com > > Help/Unsubscribe/Update your Subscription: > https://lists.apple.com/mailman/options/cocoa-dev/cmh%40me.com > > This email sent to c...@me.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: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: SecStaticCodeCheckValidity fails when app is lauched from Terminal
How are you getting the URL that you pass to represent your application? Could it be that you’re constructing the URL from a relative path when run from the command line, rather than the full path? (You can’t depend on being run from any particular working directory.) -- Chris > On Sep 26, 2016, at 2:44 AM, Markus Spoettl wrote: > > I'm using SecStaticCodeCheckValidity() to self check the signature of my own > app when it is launched. This works fine and always has. > > All of a sudden, the call to SecStaticCodeCheckValidity() fails if (and only > if the application) is started from the Terminal. When I start the very same > app from the Dock or from the Finder the check succeeds (iow. the call > returns noErr). > > I don't know exactly when it started failing. I only know it definitely > worked before on previous versions of El Capitan but now it no longer does (v > 10.11.6). > > Any ideas? > > Regards > Markus > -- > __ > Markus Spoettl > ___ > > Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) > > Please do not post admin requests or moderator comments to the list. > Contact the moderators at cocoa-dev-admins(at)lists.apple.com > > Help/Unsubscribe/Update your Subscription: > https://lists.apple.com/mailman/options/cocoa-dev/cmh%40me.com > > This email sent to c...@me.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: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: SecStaticCodeCheckValidity fails when app is lauched from Terminal
On Sep 27, 2016, at 1:54 PM, Markus Spoettl wrote: > > On 27/09/16 22:39, Chris Hanson wrote: >> How are you getting the URL that you pass to represent your application? >> >> Could it be that you’re constructing the URL from a relative path when run >> from the command line, rather than the full path? (You can’t depend on being >> run from any particular working directory.) > > Not sure what you mean by URL, I'm merely executing the app's executable from > the command line. Assuming the My.app bundle is located in > "~/Projects/My/Debug", I cd into "~/Projects/My" and execute > "./Debug/My.App/Contents/MacOS/My”. I mean, how are you constructing the SecStaticCodeRef that you pass to SecStaticCodeCheckValidity()? -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: CoreData headaches
On Oct 27, 2016, at 9:02 PM, Steve Mills wrote: > > Yes, the Asset is an NSManagedObject. In this call chain, there is no > NSManagedObjectContext in sight. There is always an NSManagedObjectContext involved; an NSManagedObject doesn’t exist outside one. Fortunately, you don’t need to pass one along with an NSManagedObject, you can just ask the NSManagedObject for the context it’s a part of. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: CoreData headaches
On Oct 28, 2016, at 9:44 PM, Steve Mills wrote: > > On Oct 27, 2016, at 23:35:39, Dave Fernandes wrote: >> >> The managed objects exist in a MOC whether you have a reference to that MOC >> or not. You can get a reference to the MOC that an MO “belongs to" from the >> -[NSManagedObject managedObjectContext] instance method. Since the >> properties you need are so few and simple, why don’t you just pass these in >> to the NSOperation when you create it on the main thread instead of giving >> it the managed object? Then the MO will never be accessed off the main queue. > > I moved the CGImageSource creation to outside the block, which is where I > needed to access the managed object's folder and name properties. The block > now just loads the image from the source, converts it to an NSImage, and sets > the managed object's thumb property. It's no longer crashing, but the setting > of the thumb property seems like that shouldn't happen inside the block. You’re correct in this, reading a property of an NSManagedObject from another thread isn’t safe, and writing it isn’t either. > So would that be the right place to use the NSManagedObjectContext's > performBlock:? What you’ve done actually looks correct, in that you’re only interacting with the managed object’s managed (Core Data) properties (“thumb” in this case) within the block passed to -performBlock:. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Representing an object graph
On Dec 5, 2016, at 4:18 PM, Daryle Walker wrote: > I've heard that Core Data is a object graph and persistence library. What if > you want just the first part? The graph seems like a neat way to save on > modeling code, the external format is not database-ish at all (so the > capability for custom export formats won't help). Can I just not use the > persistence part and use custom save & load functions? Or do I have to (or > should) give up on Core Data? This is exactly the sort of thing that subclassing NSAtomicStore <https://developer.apple.com/reference/coredata/nsatomicstore> lets you do. There are only a few methods to override, and then you can just use one of your own documents as if it were one of the built-in persistent store types. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Representing an object graph
On Dec 15, 2016, at 7:24 PM, Daryle Walker wrote: > > On Dec 6, 2016, at 10:18 PM, Chris Hanson mailto:c...@me.com>> > wrote: >> >> On Dec 5, 2016, at 4:18 PM, Daryle Walker > <mailto:dary...@mac.com>> wrote: >> >>> I've heard that Core Data is a object graph and persistence library. What >>> if you want just the first part? The graph seems like a neat way to save >>> on modeling code, the external format is not database-ish at all (so the >>> capability for custom export formats won't help). Can I just not use the >>> persistence part and use custom save & load functions? Or do I have to (or >>> should) give up on Core Data? >> >> This is exactly the sort of thing that subclassing NSAtomicStore >> <https://developer.apple.com/reference/coredata/nsatomicstore> lets you do. >> >> There are only a few methods to override, and then you can just use one of >> your own documents as if it were one of the built-in persistent store types. > > I have looked into that, that’s why I wrote in: > >>> the external format is not database-ish at all (so the capability for >>> custom export formats won't help) > > The store format assumes there’s at least one field in a record that can be > used like a database(-ish) index. Custom stores can’t help if the data is > too dumb to have such a qualifying field. What do you mean? Most atomically read/written file formats should be supportable as an NSAtomicStore. Object IDs can be purely an in-memory construct for such a store, for example based on the order an object was (initially) read or instantiated. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Weird NSPredicate structure when using "first" in keypath
On Jan 9, 2017, at 8:29 PM, Jens Alfke wrote: > On Jan 9, 2017, at 6:38 PM, Aaron Tuller wrote: >> Try doing >> name.#first >> as FIRST and LAST are reserved words and need escaping: > > Thanks for the answer. Unfortunately this is a library that will allow > developers to provide their own predicate strings, and I can imagine this > causing confusion (and therefore support requests for us.) > > I added a workaround to my code that will recognize the weird predicates > formed by using unescaped ‘first’ and ‘last’, and handle them appropriately. > I hope that’ll be robust enough! If that's the case, why not just provide your library's API in terms of NSPredicate objects? Then if your users want to use a predicate string (rather than an NSPredicateEditor or other form of direct instantiation) they still trivially can, but it's on them to get the syntax correct rather than on you to try to figure out their intent. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Building .ipa in XC6.01
We're having a hard time building an IPA file in XC6. From what I have read online, it seems to be a bug. We can make a .pkg file or an XCarchive file from our app. Any confirmation this really is a bug? Seems really unusual for such an important feature. We have already tried to add the LSRequiresiPhoneOS key in the Plist but that didn't do anything for us. Any other tips or related links appreciated. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Multiple Today Widgets
Would there be any issue with creating an app that contains multiple Today Widgets? I've actually done it, and it works on the simulator and on my phone. But now with XC6, I can't build an archive so I am wondering if it is related to the .IPA build problem, or Apple won't allow multiple widgets for distribution? I see no technical reason why it would not. Thanks, Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
[MODERATOR] End of Thread: Re: Blurry is the New Sharp
Please stick to technical discussion on cocoa-dev. If there are remaining technical questions in this thread, please ask them in their own threads. (And avoid off-topic derails.) Thanks. -- Chris (cocoa-dev co-mod) ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: All buttons lost focus ring on Yosemite
Hi all, We had a similar issue. When running on Yosemite the same thing would happen to us with the focus rings not displaying. Don't know if this helps in your case, but the "fix" for us was to stop building with the 10.7 SDK. Moving to the 10.8 SDK fixed the problem for us. (Don't ask why we are using such old SDKs, and only moved to 10.8 instead of 10.9 or 10.10, that's another story for another time.) You can see this in a simple Cocoa app too started from an Xcode template. Thanks, Chris _ Chris Cianflone www.makemusic.com On Tue, Mar 10, 2015 at 5:13 AM, wrote: > What happens if you turn on full keyboard access? > > Sent from my iPhone > > > On 2015/03/10, at 18:45, Dragan Milić wrote: > > > > My previous message sent to the list apparently didn’t get admin > approval, there were two attached images showing standard alarm panels with > buttons which don’t draw their focus ring. Therefore, in this message I’ll > combine answers to both Kyle and Graham. > > > >> On pon 09.03.2015., at 17.48, Kyle Sluder wrote: > >> > >> I'd start by looking at that -[NSView(NTExtensions) drawFocusRing] > method. If you comment it out, what happens? > > > > No changes :-( I even went that far to remove the category completely > and ignore all compiler warnings where added methods were used. Now, that > would normally throw exceptions, but before even loading any application > window and using any view requiring added category methods, the application > shows two alert panels (one by one, attached images of those panels is the > reason my previous message didn’t get through ), depending on current user > preferences. That enables me to see buttons' behaviour even before any > exception is thrown. Here is the link to the image of the first alert panel: > > > > http://media.cocoatech.com/first_alert.png > > > > What’s interesting here is that, as you can see, the alert’s > suppressionButton checkbox DOES draw its focus ring, but other two buttons > don’t! Believe my words, if I use tab/shift+tabs to change focus on them, > they are focused (I can then activate them with the space), but the focus > ring is not drawn, only the checkbox gets it correctly. > > > > Immediately after that first alert is dismissed, the second one appears, > here is the link: > > > > http://media.cocoatech.com/second_alert.png > > > > With this one, even the focused suppressionButton checkbox doesn’t draw > its focus ring. Other two buttons don’t too. And all this is with the > NSView category completely removed. There are other NSView or NSControl or > NSButton or NSButtonCell categories defined. The code defining both alarm > panels is pretty standard, I can post it if necessary. > > > > Everything worked well in Mavericks, I’ve got no idea what happened in > the meanwhile. > > > >> For safety, you really ought to be prefixing all of your category > methods with some unique-to-you prefix. > > > > Yeah. As I’ve mentioned already, this is not my code, it was first > created back in 2002 and until this problem I haven’t even looked into it. > > > > > >> On uto 10.03.2015., at 00.44, Graham Cox wrote: > >> > >> In Xcode, add OBJC_PRINT_REPLACED_METHODS (value: YES) to your scheme's > environment variables. Then all of the methods replaced by categories are > logged when your app launches. While the list can be long and might take > some time to go through, it will show you if any of your category methods > are replacing anything - it's a much more reliable way to check than doing > a class dump. > > > > Thanks for the suggestion Graham, I’ll do that. It’s be interesting to > see if and what methods are replaced, although I don’t think it’ll solve > this particular problem. I mentioned above it still happens even with the > NSView category completely removed. > > > >> If you're not using those category methods, remove them. Most of them > seem to be convenience methods that are possible "nice to haves" rather > than vital to use NSView. Some appear to me to wholly misunderstand how a > view stack (involving semi-transparancy for example) is actually drawn. > Others are things that could be useful in particular circumstances but you > probably wouldn't want to apply to every view your app ever instantiates > including framework ones. For those custom views of yours that use these > things, relocate that code to the custom view. It may mean a small > duplication of code across a few different views, but it will be a lot > safer than swapping out NSView wholesale. NSView just may be Cocoa's single &
MODERATOR: End of Thread (was Re: I am reluctant to file any more bugs until those already reported are fixed)
This is the place for neither rants nor personal attacks. Please keep it technical. Thanks. -- Chris Hanson (cocoa-dev co-moderator) ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: MODERATOR: End of Thread (was Re: I am reluctant to file any more bugs until those already reported are fixed)
Once again, this thread is over. Please do not post in it again. This list is for asking and answering questions about development with the Cocoa, Cocoa Touch, and related frameworks. There are appropriate venues for feedback to Apple, including the site at <http://www.apple.com/feedback/ <http://www.apple.com/feedback/>> and the bug reporter at <http://bugreport.apple.com/ <http://bugreport.apple.com/>>. -- Chris Hanson (cocoa-dev co-mod) ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: system(root) locale
[NSLocale systemLocale] returns the root NSLocale. If you are referring to the BSD/Unix-level locale, you can start by looking in /usr/include/xlocale.h, but I don't know anything else about that. Chris Kane Cocoa Frameworks, Apple On Aug 31, 2009, at 12:53 AM, Maggie Zhang wrote: Hi, Does anyone know how to get the root locale in Objective-c or C or using Unix commands? I tried NSLocale but it didn't give what I want. Though I can get the user's current locale in different ways. Thanks for you help. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/ckane%40apple.com This email sent to ck...@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: What about revamping OpenUp.app for Snow Leopard?
On Sep 3, 2009, at 10:14 PM, Scott Anguish wrote: Looking inside zip/tar/tar.gz files (something OpenUp never did) would best be accomplished by writing a QuickLook plug-in (which probably exists. I don’t know about any Quick Look plug-ins, but the ZipBrowser example code both provides a skeleton of browsing zip archives and demonstrates a number of modern Cocoa development techniques: This example shows two versions of a single application: the original version of the application, in ZipBrowserBefore, and the polished version, in ZipBrowserAfter. ZipBrowser is a simple document-based application used for perusing the contents of a zip archive without having to unarchive it. The modifications between ZipBrowserBefore and ZipBrowserAfter fall into six categories: 64- bit readiness, performance and responsiveness, security, localization and internationalization, usability, and accessibility. Download ZipBrowser 1.1 here: http://developer.apple.com/mac/library/samplecode/ZipBrowser/index.html You can also search for ZipBrowser in the Snow Leopard documentation, and download & open the project right within Xcode. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Code Signing
Hi Peter, On Sun, Sep 6, 2009 at 9:00 PM, Peter Hudson wrote: > I notice that when I sign code ( the whole app ) I land up with 2 > additional items in my /Contents directory in the app bundle. > These items are a folder called _CodeSignature and an alias called > CodeResources. > > If I delete these two items, the code still identifies itself as signed ( > when I attempt to run codesign on it again ). > Also, the app still runs. > > I was wondering if these files are meant to remain or if removing them is > not an issue ? Within an application bundle there are signatures for all the files within the bundle and then there's a signature for the binary itself. The signature for the binary is stored within executable and you can see it if you use the otool -l command (look for the LC_CODE_SIGNATURE). The signature for all the resources is stored within the file you've discovered. As far as I know, except for static validation (using the codesign tool), the signature for the resources isn't used by much (or at least it wasn't for Leopard). The signature for the binary is used by the system for various things. For example you can set the kill flag to have your application terminated if there's an invalid signature, and it's used for Keychain access (so that if you upgrade an application, you won't be prompted again for password access provided the signature remains valid). It's important to remember that code signatures are not really there to prevent malware from running, or make it harder for hackers (since it's trivial to remove a code signature or replace them other valid signatures). At any rate, you shouldn't be deleting those files. If you remove them, the signature will be invalidated. By the way, the best list for code signing questions is the apple-cdsa list. Kind regards, Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
usb notification
Hi all,
Can't get a simple example working. AppearedNotificationHandler() should be
called in the code below when a USB device (any device for now) is plugged in.
I see the NSLog(@"registerDeviceCallbackHandler") message appear in the console
when I start the application, but never see the
NSLog(@"AppearedNotificationHandler") message when I plug in a device. What am
I missing?
Note: The project's MainMenu.nib Interface Builder file has an NSObject whose
class is "USBDriver".
static void AppearedNotificationHandler (void * refCon, io_iterator_t iterator);
@interface USBDriver(Private)
- (void) registerDeviceCallbackHandler;
@end
@implementation USBDriver(Private)
- (void) registerDeviceCallbackHandler
{
IOReturnkernErr;
CFRunLoopSourceRef runLoopSource;
CFMutableDictionaryRef matchingDict = IOServiceMatching
(kIOUSBDeviceClassName);
CFRunLoopRefrunLoop;
// Create the port on which we will receive notifications. We'll wrap
it in a runLoopSource
// which we then feed into the runLoop for async event notifications.
deviceNotifyPort = IONotificationPortCreate (kIOMasterPortDefault);
if (deviceNotifyPort == NULL)
{
return;
}
// Get a runLoopSource for our mach port.
runLoopSource = IONotificationPortGetRunLoopSource (deviceNotifyPort);
matchingDict = (CFMutableDictionaryRef) CFRetain (matchingDict);
kernErr = IOServiceAddMatchingNotification (deviceNotifyPort,
kIOFirstMatchNotification,
matchingDict,
AppearedNotificationHandler,
(void *) self,
&deviceAppearedIterator);
kernErr = IOServiceAddMatchingNotification (deviceNotifyPort,
kIOTerminatedNotification,
matchingDict,
DisappearedNotificationHandler,
(void *) self,
&deviceDisappearedIterator
);
// Get our runLoop
runLoop = [[NSRunLoop currentRunLoop] getCFRunLoop];
// Add our runLoopSource to our runLoop
CFRunLoopAddSource (runLoop, runLoopSource, kCFRunLoopDefaultMode);
NSLog(@"registerDeviceCallbackHandler");
}
@end
@implementation USBDriver
- (id) init
{
[super init];
if (self) {
deviceNotifyPort= IO_OBJECT_NULL;
deviceAppearedIterator = 0;
[self registerDeviceCallbackHandler];
}
return self;
}
@end
static void AppearedNotificationHandler (void * refCon, io_iterator_t iterator)
{
NSLog(@"AppearedNotificationHandler");
}
Any help is much appreciated.
Thanks,
Chris
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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: usb notification
> >IOReturnkernErr;
>
> You never seem to check for errors. Please ensure
> your calls are
> succeeding by checking the value of this variable. If
> they're
> failing, use macerror(1) to look up the error number.
IOServiceAddMatchingNotification() returns no errors (0).
> > - (id) init
> > {
> >[super init];
> >
> >if (self) {
>
> This is not the correct initializer pattern. You need
> to assign to self here.
>
> --Kyle Sluder
Changed this to self = [super init];
However, the issue still persists.
Chris
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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
testing ppc on intel
Every program that I build universal but run on intel (OS 10.5) with "arch -ppc" option, crashes with a report like the following, and I've tested quite a few, even simple ones. Is it unreasonable to try to test ppc programs on intel, or am I doing something wrong, or what? Version: ??? (???) Code Type: PPC (Translated) Parent Process: bash [1652] Interval Since Last Report: 83 sec Crashes Since Last Report: 1 Per-App Interval Since Last Report: 0 sec Per-App Crashes Since Last Report: 1 Date/Time: 2009-09-14 15:05:30.700 +1000 OS Version: Mac OS X 10.5.8 (9L30) Report Version: 6 Anonymous UUID: FA165A8B-95AA-4C9A-9296-C70D9F3138FC Exception Type: EXC_CRASH (SIGTRAP) Exception Codes: 0x, 0x Crashed Thread: 2 Thread 0: 0 ??? 0x816a5e28 0 + 2171231784 Thread 1: 0 ??? 0x800bc286 0 + 2148254342 1 ??? 0x800c3a7c 0 + 2148285052 2 translate 0xb818b6ea CallPPCFunctionAtAddressInt + 202886 3 ??? 0x800ed155 0 + 2148454741 4 ??? 0x800ed012 0 + 2148454418 Thread 2 Crashed: 0 ??? 0x8019d402 0 + 2149176322 1 translate 0xb80b6b00 0xb800 + 748288 2 translate 0xb80b7007 0xb800 + 749575 3 translate 0xb80d49c0 0xb800 + 870848 4 translate 0xb813d75f spin_lock_wrapper + 4259 5 translate 0xb8011b64 0xb800 + 72548 Thread 2 crashed with X86 Thread State (32-bit): eax: 0x ebx: 0xb80b6c78 ecx: 0xb009adcc edx: 0x8019d402 edi: 0xb8208980 esi: 0x0005 ebp: 0xb009adf8 esp: 0xb009adcc ss: 0x001f efl: 0x0246 eip: 0x8019d402 cs: 0x0007 ds: 0x001f es: 0x001f fs: 0x001f gs: 0x0037 cr2: 0xf00fffd0 Binary Images: 0xb800 - 0xb81d7fe7 translate ??? (???) /usr/libexec/oah/translate Translated Code Information: NO CRASH REPORT __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: testing ppc on intel
10.5.8. Now that I try, a new empty Cocoa app in XCode works. My app is just a one window Cocoa program that links with two frameworks (both of which are universal with ppc). arch -ppc works with /bin/cat, iTunes and so forth. Just not with my programs. I'm pretty sure I have no input managers, I cleared the contents of ~/Library/InputManagers and /Library/InputManagers, and I don't think I have kexts, I haven't installed those sort of programs, or any haxies. If there is something wrong with my program, I'm not sure how to debug it. From: Greg Guerin To: list-cocoa-dev Sent: Monday, 14 September, 2009 5:33:48 PM Subject: Re: testing ppc on intel Chris Idou wrote: > Every program that I build universal but run on intel (OS 10.5) with "arch > -ppc" option, crashes with a report like the following, and I've tested quite > a few, even simple ones. How simple is simple? Hello world, or simpler? What about programs you didn't write, such as /bin/cat? Exactly which version of Leopard? 10.5 covers at least nine revisions. Do you have any input managers, haxies, kexts, etc. active? -- GG ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/idou747%40yahoo.com This email sent to idou...@yahoo.com __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: Grand Central Dispatch vs CFRunLoop
eneral danger in all this is that some API you are using refers to the "current run loop" and does something to it which is crucial later. This has always been true for background threads, but the pitfall becomes more pronounced as more developers move to multithreading, particularly using the "indirect" kinds of mechanisms like dispatch queues and NSOperationQueues. With respect to NSTask, there's no reason to believe that it will work if you just run the main dispatch queue rather than running the NSRunLoop. I highly doubt it would. I think it will work fine as long as you don't want the NSTaskDidTerminateNotification and related (ie, -isRunning to return the right answer) and anything else the NSTask might do as a result of the child process dying. However, if you are also using NSFileHandle "in-background" operations, then you will need the run loop run (on the thread(s) which start those background reads) to get those notifications. That's a separate matter from NSTask, but people may confuse the two together so I mention it here. Chris Kane Cocoa Frameworks, Apple ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Is a serial NSOperationQueue FIFO?
On Sep 6, 2009, at 9:32 PM, Eric Hermanson wrote: When working with a serial NSOperationQueue (i.e. setMaxConcurrentOperationCount==1), are the operations added to the queue guaranteed to run in FIFO order if all operations have the same priority and no operation dependencies are involved? My guess is that the NSOperationQueue will run the operations in FIFO order, however, I don't believe the NSOperationQueue documentation guarantees this. The documentation says operations will run based on priority, then on dependency structure. Because all operations in this aforementioned serial queue have the same priority (i.e normal) and no dependencies, then I suppose the NSOperationQueue could sort the operations any way it wanted, not necessarily in the order I added them to the queue. Does anyone have insight into this, or is there documentation anywhere that can guarantee to me an NSOperationQueue with a concurrent operation count of 1 will indeed process the operations first come first serve? If all operations have the same priority (which is not changed after the operation is added to a queue) and all operations are always - isReady==YES by the time they get put in the operation queue, then a serial NSOperationQueue is FIFO. The second condition I mention there is stronger than "no operation dependencies" -- dependencies don't directly matter to a queue, just the readiness of the operations, and dependencies are just one thing that can affect readiness. If you are adding NSOperations to a given queue from multiple threads, keep in mind that it is very tricky (well, impossible) to accurately observe from the outside the actual order of events (ie, who got in first). Chris Kane Cocoa Frameworks, Apple ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Need -[NSTask waitUntilExitOrTimeout:] (was NSTask "Thread Unsafe"...)
On Sep 14, 2009, at 6:29 PM, Jerry Krinock wrote:
The reason why I asked this is because I need:
-[NSTask waitUntilExitOrTimeout:(NSTimeInterval)].
But there is no such method. So, I wrote a wrapper around NSTask
that does this.
On 2009 Sep 14, at 07:41, Adam R. Maxwell wrote:
Ordinarily, I'd guess that it's supposed to mean that you can't
share instances between threads. However, guesswork when dealing
with the threading docs drives me crazy.
For NSTask in particular, I recommend that you interpret that more
conservatively...
Also, I presume it means that you should not invoke the same
instance from more than one thread. I'm not sure how much more
conservative you could get, other than to run it on the main thread.
I wrote an implementation of NSTask from scratch to work around the
problem.
Whew, Adam. Since I'm 10.5+, and don't think I'll ever be running
more than one NSTask at a time, I won't be quite that conservative.
In my NSTask wrapper, I detach a new thread. In this new thread, I
set up the task, then run the run loop once...
[task launch] ;
if ([task isRunning]) {
// Write data to stdin file/pipe
// ...
// ...
}
NSDate* limitTime = [NSDate dateWithTimeIntervalSinceNow:timeout] ;
[[NSRunLoop currentRunLoop] runMode:NSDefaultRunLoopMode
beforeDate:limitTime] ;
// The above will block and be run to here either due to
// the posting of an NSTaskDidTerminateNotification, or the
// passing of limitTime, whichever occurs first.
That is not a valid assumption (described in the comments and apparent
from the rest of the code).
Just do the loop, checking the time and isRunning. Running the run
loop once is not necessarily sufficient.
Keep in mind that isRunning can start returning NO before (or after)
the notification gets delivered. There's no necessary correlation
between the two. If you actually care that the notification has been
posted, you should register an observer and change some shared state
in the handler method, which the looping can check for.
if (![task isRunning]) {
taskResult = [task terminationStatus] ;
// Read stdout and stderr
// ...
// ...
}
else {
taskResult = SSYShellTaskerErrorTimedOut ;
// Clean up
kill([task processIdentifier], SIGKILL) ;
// Set NSError indicating timeout
// ...
// ...
}
Actually, I originally was doing this on the main thread, except
that the -[NSRunLoop runMode:beforeDate] was in a while() loop.
When the loop ran, I would check and see if there was a timeout,
task was finished, or neither (presumably some other input source
triggered the run). But then I found an obscure bug: If I had
disabled undo registration in order to do some behind-the-scenes
changes, it would re-enable undo registration. Uh, yeah, it took me
a while to track that one down. Apparently, running a run loop that
is already running is not a good idea. I suppose I could also have
done it with a timer.
If I interpret your explanation of the previous code correctly and it
was similar to the code above, the problem here, in a sense, is more
your desire to block until the task is finished or a timeout expires.
Go back to the main thread. Setup a oneshot NSTimer for the timeout
period. Setup a notification handler to listen for the
NSTaskDidTerminateNotification. If the timer fires first, kill the
task, unregister the notification handler, etc. If the notification
happens first, invalidate the timer, unregister the notification
handler, etc. Don't run the run loop yourself. Let your code be
event-driven.
A run loop can be run re-entrantly, but as Jens said, usually it is
done in a private mode, to prevent (say) the default mode stuff from
happening at times which are surprising to that stuff (breaking
assumptions or requirements it has), and possibly surprising to your
app (breaking assumptions or requirements it has). However in this
case, the task death notification, if you need that, requires the
default run loop mode to be run to get delivered.
Chris Kane
Cocoa Frameworks, Apple
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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: #pragma to suppress a warning message
On Sep 16, 2009, at 6:09 PM, Jay Boyer wrote: Finally, I am doing this because the preferredLanguages method is specified to be an OS 10.5+ method and the project I am working on is being built on the 10.4 SDK. The proper way to set up this project, as you’ve described it, is to build the project using a Base SDK of Mac OS X 10.5 (or later), and a Mac OS X Deployment Target of 10.4. - Base SDK is the latest operating system from which you wish to use features - Deployment Target is the earliest operating system on which your software will run This is the case for both Mac OS X and iPhone OS. You do not need to use the SDK specific to an operating system just because that is the minimum operating system your application will support. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Static Analyzer and Core Foundation
On 18 Sep 2009, at 4:33 PM, Steve Cronin wrote:
Alert - potential boneheaded-ness lies ahead - please be gentle.
I'll be the first to admit that I'm much happier in ObjC than C…
I thought I would try static analysis and see what turned up.
On the whole I must say I'm pleased but this one has me questioning
my basic understanding
if (![[NSFileManager defaultManager] fileExistsAtPath:thisPath]) {
…
41 } else {
42 MDItemRef mdi = MDItemCreate( nil, (CFStringRef)thisPath );
43 if ( mdi != nil ) {
44 CFDictionaryRef dictRef = MDItemCopyAttributes( mdi,
MDItemCopyAttributeNames(mdi));
…
CFRelease(dictRef);
} else {
… handle error
}
}
Static Analysis tells me I have a potential leak of object allocated
on line 44 -- that's all.
My questions are:
1) Why is there not a warning for the object allocated on line 42 -
mdi ?
Seems like it should be; you're not releasing it and it's going out of
scope.
2) Why is the CFRelease(dictRef) not sufficient?
In this case, it's probably complaining about the
MDItemCopyAttributeNames() call, which (because it has "Copy" in the
name) returns an object you'd have to release.
I.e., line 44 allocates two objects, the dictionary which you *are*
releasing and the attribute names array, which you're not.
.chris
--
Chris Parker
Apple Inc.
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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
building and running on Snow Leopard
If you build an application on Snow Leopard, but against the 10.5 deployment target, and then you run the program on Snow Leopard, do you get all the Leopard 10.5 bugs as if you ran it on Leopard, or do you get to benefit from Snow Leopard bug fixes only if you build against the 10.6 deplotment target? __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: building and running on Snow Leopard
Ok, so if I don't desire bug for bug 10.5 compatibility, but I do want it to run on 10.5, then I select 10.6 SDK and 10.5 deployment target, is that right? I'm curious what happens if you use some external frameworks that were built against 10.5 SDK, yet your main app is 10.6 - I assume it will have to go with what your app says, I guess? Also, since I'd like to take advantage of Snow Leopard bug fixes, but I don't want to blow up on Leopard, what is the correct way to test whether you are running on Leopard so that the code at runtime can implement Leopard specific work arounds? From: Ken Thomases To: Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Wednesday, 23 September, 2009 5:37:01 PM Subject: Re: building and running on Snow Leopard On Sep 23, 2009, at 2:20 AM, Chris Idou wrote: > If you build an application on Snow Leopard, but against the 10.5 deployment > target, and then you run the program on Snow Leopard, do you get all the > Leopard 10.5 bugs as if you ran it on Leopard, or do you get to benefit from > Snow Leopard bug fixes only if you build against the 10.6 deplotment target? Not quite either. At runtime, your app can only use the frameworks provided on the host system. Just because you linked against the 10.5 SDK doesn't change the fact that Snow Leopard only provides Snow Leopard's frameworks. In general, your app will get the benefit of bug fixes in the Snow Leopard frameworks when running on Snow Leopard. However, there are some bug fixes which Apple has determined would break backward compatibility with apps built and tested on Leopard or earlier. For those fixes, they test which SDK the app was built against. For those built against the pre-Snow Leopard SDK, they continue to provide the old, compatible-bug-buggy behavior. This is documented in the release notes for the affected frameworks. Note that this tests for the SDK that the app was built against. You mentioned deployment target. That's not the same thing. If you build against the 10.6 SDK but with a 10.5 deployment target, it is expected that you will be testing on both 10.6 and 10.5. It is then your code's responsibility to handle both the old framework behavior the new behavior. The Snow Leopard frameworks _do not_ go out of their way to maintain the old behavior in this case. Regards, Ken __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: building and running on Snow Leopard
From: Ken Thomases >Don't test whether you're running on Leopard. Just write your code to cope >with either behavior of the framework. Read the release notes for the >frameworks on which you rely for more guidance. I think when you see the >sorts of bugfixes which are special-cased in this way, it will be >obvious how you should code in response. Well I'm thinking of a particular bug on Leopard with MDItemCreate where you have to choose between your program crashing and leaking memory. Apple tells me it is fixed in Snow Leopard. On Leopard I choose to leak memory, but I'd rather not do that on Snow Leopard. __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: NSManagedObject Initialization Methods Not Called
On Sep 21, 2009, at 12:12 PM, Richard Somers wrote: I have a core data document based application. When a file on the disk is opened -awakeFromInsert and -awakeFromFetch are never called for one of my NSManagedObject objects. Why is not one of these methods getting called when the object is loaded into memory from disk? It sounds like you’re expecting adding a persistent store to a persistent store coordinator to result in all of its managed objects being sent either -awakeFromInsert or -awakeFromFetch. * -awakeFromInsert will only be sent to a managed object when it’s actually inserted into a managed object context, not when it’s realized from a fault. * -awakeFromFetch will only be sent to a managed object when it’s actually realized from a fault, not when the persistent store containing it is added to a persistent store coordinator — or even when it’s fetched via a fetch request or relationship traversal. Hope this clears things up a little. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Snow Leopard: unsupported PointerFunctions configuration was requested
I'm porting an app to Snow Leopard. Very early in application startup, even before main() is called, I get the following error: 2009-09-24 12:07:11.462 MyApp[5534:a0f] *** An unsupported PointerFunctions configuration was requested, probably for use by NSMapTable, NSHashTable, or NSPointerArray. The requested configuration fails due to integer personality not using opaque memory It happens twice on startup, and below are the stack traces when it happens: What does it all mean? #00x7fff84b342f7 in NSLog #10x7fff84ac84d1 in +[NSConcretePointerFunctions initializeSlice:withOptions:] #20x7fff84aea796 in -[NSConcretePointerFunctions initWithOptions:] #30x7fff84aea74d in +[NSPointerFunctions pointerFunctionsWithOptions:] #40x10004b633 in +[RKCache load] #50x7fff8764141b in call_load_methods #60x7fff87641160 in load_images #70x7fff5fc0315a in __dyld__ZN4dyldL12notifySingleE17dyld_image_statesPK11ImageLoader #80x7fff5fc0bcdd in __dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContextEj #90x7fff5fc0bc9d in __dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContextEj #100x7fff5fc0bda6 in __dyld__ZN11ImageLoader15runInitializersERKNS_11LinkContextE #110x7fff5fc0210e in __dyld__ZN4dyld24initializeMainExecutableEv #120x7fff5fc06981 in __dyld__ZN4dyld5_mainEPK12macho_headermiPPKcS5_S5_ #130x7fff5fc016d2 in __dyld__ZN13dyldbootstrap5startEPK12macho_headeriPPKcl #140x7fff5fc01052 in __dyld__dyld_start #00x7fff84b342f7 in NSLog #10x7fff84aea7bd in -[NSConcretePointerFunctions initWithOptions:] #20x7fff84aea74d in +[NSPointerFunctions pointerFunctionsWithOptions:] #30x10004b633 in +[RKCache load] #40x7fff8764141b in call_load_methods #50x7fff87641160 in load_images #60x7fff5fc0315a in __dyld__ZN4dyldL12notifySingleE17dyld_image_statesPK11ImageLoader #70x7fff5fc0bcdd in __dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContextEj #80x7fff5fc0bc9d in __dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContextEj #90x7fff5fc0bda6 in __dyld__ZN11ImageLoader15runInitializersERKNS_11LinkContextE #100x7fff5fc0210e in __dyld__ZN4dyld24initializeMainExecutableEv #110x7fff5fc06981 in __dyld__ZN4dyld5_mainEPK12macho_headermiPPKcS5_S5_ #120x7fff5fc016d2 in __dyld__ZN13dyldbootstrap5startEPK12macho_headeriPPKcl #130x7fff5fc01052 in __dyld__dyld_start __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: Controlling Spaces and retrieving information about the current Space?
>>> Secondly, is there any way in any of those environments to >>> programmatically switch Spaces? [ ... ] >> >> No, there is no API (in Cocoa or otherwise) to control the active space. >But how does Spaces do it? Is it via an internal, private API that >people like us don't have access to? There may not be any API, internal or otherwise. It might be all just built into the internal workings of the window manager process. One would presume there is an internal API for switching to the grid view, since Spaces.app does that. __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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
Snow Leopard, core data, read only and multiple threads
I've got an app that worked on Leopard. I ported it to Snow Leopard SDK 10.6, and now it works on Snow Leopard, but it doesn't work correctly on Leopard anymore. I haven't changed anything that ought to affect this. It's an app with a foreground gui that writes an XML coredata store. A background thread reads the repository and takes action. Both threads have the full core data stack with their own coordinators. As soon as I activate the background thread, the XML store gets set to zero bytes. When I encountered the problem I read the doco and I added the NSReadOnlyPersistentStoreOption when calling addPersistentStoreWithType in the background thread, but that hasn't helped. It wasn't necessary before. Has anyone got any thoughts? And also, how am I going to debug this? Xcode 3.2 doesn't run on 10.5 does it? And is an Xcode 3.2 project now going to work with XCode 3.1? __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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
NSDocumentController didCloseAllSelector
The doco for closeAllDocumentsWithDelegate:didCloseAllSelector:contextInfo: says that it should call the didCloseAllSelector. How would one do that? This is what I am doing: [delegate performSelector:didCloseAllSelector withObject:(id)YES withObject:contextInfo]; but I'm nervous about that typecast. Is that the way, or is there another way? __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: NSWindowController and GC
The right way is to make sure something in your program retains a pointer to your window controller. Sometimes there will be an obvious place, like a member of the object that created the window controller. That is most typical. Worst case is you retain a pointer in some global variable or similar, but that also has a greater danger that you will forget to zero it out when you are finished with it, and it won't get collected. Some of my temporary dialog box controllers inherit from a class that puts itself in a global dictionary when created, and removes itself when the window closes. From: Bryan Matteson To: cocoa-dev@lists.apple.com Sent: Thursday, 1 October, 2009 9:20:16 AM Subject: NSWindowController and GC I was just reminded of something. I use GC in my app, and unless I specifically disable collection for a window controller, it's destroyed as soon as it loses key. I solved it by setting the controller as the delegate of the window, disabling collection for the window controller in windowDidLoad, and re-enabling it in windowWillClose:. Is there another way? -B ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/idou747%40yahoo.com This email sent to idou...@yahoo.com __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.com/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: Best Design Advice
If it displays a window it should be an app. If its an app it should be in /Applications and it should be started from Login Items. You can't install it automatically by security design. But once you get the user to install and run it, there are no permissions issues in adding it to login items (although asking the user is polite), and connecting to your web site. From: David Blanton To: cocoa-dev List Sent: Sunday, 15 November, 2009 4:21:31 AM Subject: Best Design Advice I need to create an app the is auto downloaded and installed from a web site when the user clicks a web page button. The app needs to be installed so that it always runs when the user logs in. The app needs to periodically (by user preference setting) connect to a web server. The app needs to display a window with information gleaned from its last web server connect. = 1. Where should the app be installed. 2. Are there any permission issues. 3. Is this a background service or status like item. 4. Is it's periodic web connect fired by a timer or some other means. 5. Any other general tips / tricks. = Thanks in advance! db ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/idou747%40yahoo.com This email sent to idou...@yahoo.com __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: NSDocumentController didCloseAllSelector
What I have is a document based app, but it automatically saves everything - I don't want to ask the user. So I'm saving all my documents in closeAllDocumentsWithDelegate. It's working pretty good, the only major hassle is the app doesn't die on machine shutdown, presumably because I wasn't calling didCloseAllSelector. I agree the whole setup is pretty confusing but after ponding it extensively, I think it is more or less sensible, given the need for the RunLoop and so on. (albeit, poorly documented). I'm not sure what you mean you can "invoke the method easily enough". Maybe the right way to do it is with NSInvocation. But knowing what I know about objective-c, I can't see how setArgument:atIndex: would be able to know that the second argument is a BOOL which is one byte. Maybe it does know and it all just works. ____ From: Graham Cox To: Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Thursday, 1 October, 2009 11:42:05 AM Subject: Re: NSDocumentController didCloseAllSelector On 01/10/2009, at 11:00 AM, Chris Idou wrote: > The doco for closeAllDocumentsWithDelegate:didCloseAllSelector:contextInfo: > says that it should call > the didCloseAllSelector. How would one do that? This is what I am doing: Chris, this stuff is a right old mess. No wonder you're confused. It says *it* will call that method if all documents are closed. You don't necessarily have to call anything - looking at the byzantine interactions between all these methods all I can say is, if you can avoid it, do so! What do you want to do? Unless you're overriding - (void)canCloseDocumentWithDelegate:(id)delegate shouldCloseSelector:(SEL)shouldCloseSelector contextInfo:(void *)contextInfo to do something in a non-standard way there's nothing to do. If you are overriding it, you need to handle the whole kit-n-kaboodle, including presenting the alert to the user and saving the document to disk, and finally invoking the mysterious callback selector. I can't actually see an easy way to invoke the callback selector with the signature as given - you need to pass it the document (self), a BOOL and the contextInfo. You can invoke the method directly on the delegate easily enough, but since you're supposed to use the callback supplied, that might not work. The only way I can see to do it is to create an NSInvocation with the selector, set each argument and call invokeWithTarget: passing the delegate. Someone else might have a brighter idea - this seems very involved and badly thought-out. --Graham __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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: whether to use core data...
On Oct 4, 2009, at 10:39 AM, Jeffrey Oleander wrote: >> It's not a toy language or API by any stretch of the >> imagination. > > But Core Data seems to be. Please expand on this, and why you feel it to be the case. A great many people are using Core Data successfully in their applications on both Mac OS X and iPhone OS. It is by no means a toy. That you don’t currently think it meets your (unarticulated) needs does not make it one. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Snow Leopard: unsupported PointerFunctions configuration was requested
Below is the RegexKit code that causes the Snow Leopard error. After reading
the doco, I don't feel very enlightened about what the unsupported
configuration is, or what the solution might be. Can anyone give me a clue?
+ (void)load
{
RKAtomicMemoryBarrier(); // Extra cautious
if(RKCacheLoadInitialized== 1) { return; }
if(RKAtomicCompareAndSwapInt(0, 1, &RKCacheLoadInitialized)) {
if((cacheMapKeyCallBacks= dlsym(RTLD_DEFAULT, "NSIntegerMapKeyCallBacks"))
== NULL) { cacheMapKeyCallBacks= dlsym(RTLD_DEFAULT, "NSIntMapKeyCallBacks"); }
#ifdef ENABLE_MACOSX_GARBAGE_COLLECTION
id garbageCollector = objc_getClass("NSGarbageCollector");
if(garbageCollector != NULL) {
if([garbageCollector defaultCollector] != NULL) {
id pointerFunctions = objc_getClass("NSPointerFunctions");
RKCacheIntegerKeyPointerFunctions = [pointerFunctions
pointerFunctionsWithOptions:NSPointerFunctionsIntegerPersonality];
RKCacheIntegerKeyPointerFunctions.acquireFunction =
intPointerFunctionsAcquire;
RKCacheIntegerKeyPointerFunctions.descriptionFunction =
intPointerFunctionsDescription;
RKCacheIntegerKeyPointerFunctions.hashFunction=
intPointerFunctionsHash;
RKCacheIntegerKeyPointerFunctions.isEqualFunction =
intPointerFunctionsIsEqual;
RKCacheIntegerKeyPointerFunctions.relinquishFunction =
intPointerFunctionsRelinquish;
RKCacheIntegerKeyPointerFunctions.sizeFunction=
intPointerFunctionsSize;
RKCacheObjectValuePointerFunctions = [pointerFunctions
pointerFunctionsWithOptions:(NSPointerFunctionsZeroingWeakMemory|
NSPointerFunctionsObjectPersonality)];
[[garbageCollector defaultCollector]
disableCollectorForPointer:RKCacheIntegerKeyPointerFunctions];
[[garbageCollector defaultCollector]
disableCollectorForPointer:RKCacheObjectValuePointerFunctions];
}
}
#endif // ENABLE_MACOSX_GARBAGE_COLLECTION
}
}
____
From: Bill Bumgarner
To: Chris Idou
Cc: cocoa-dev@lists.apple.com
Sent: Thu, 24 September, 2009 12:36:04 PM
Subject: Re: Snow Leopard: unsupported PointerFunctions configuration was
requested
On Sep 23, 2009, at 7:09 PM, Chris Idou wrote:
Very early in application startup, even before main() is called, I get the
following error:
>
> 2009-09-24 12:07:11.462 MyApp[5534:a0f] *** An unsupported PointerFunctions
> configuration was requested, probably for use by NSMapTable, NSHashTable, or
> NSPointerArray. The requested configuration fails due to integer personality
> not using opaque memory
>
> It happens twice on startup, and below are the stack traces when it happens:
>
> What does it all mean?
>
>
> #00x7fff84b342f7 in NSLog
> #10x7fff84ac84d1 in +[NSConcretePointerFunctions
> initializeSlice:withOptions:]
> #20x7fff84aea796 in -[NSConcretePointerFunctions initWithOptions:]
> #30x7fff84aea74d in +[NSPointerFunctions pointerFunctionsWithOptions:]
> #40x10004b633 in +[RKCache load]
One change to the NSPointerFunction based collection classes (which support a
number of things that NSArray & the like do not) between Leopard and Snow
Leopard was to tighten up the validation of the pointer function validation
such that a handful of non-sensical configurations are detected that weren't
before.
I.e. RKCache is trying to create a set of pointer functions that doesn't make
sense (and for whose behavior is undefined).
As someone else pointed out, RK is most likely RegexKit and, thus, the source
is available and can be fixed. Given the relatively common use of RK*, please
provide a patch to the author so it can be fixed upstream.
b.bum
__
Get more done like never before with Yahoo!7 Mail.
Learn more: http://au.overview.mail.yahoo.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: real verses Virtual memory
On 10 Oct 2009, at 17:44, jon wrote: no, no control over the website, but on webView i turned off all the java script and flash and stuff, so that helps with the speed, not with it using virtual memory every 20 seconds, and it does need to be 20 seconds. So if you change your process to sleep 20 seconds and then repeat - instead of exiting and getting restarted - how does that change the disk activity? Cheers, Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Fw: Snow Leopard: unsupported PointerFunctions configuration was requested
Below is the RegexKit code that causes the Snow Leopard error. After reading
the doco, I don't feel very enlightened about what the unsupported
configuration is, or what the solution might be. Can anyone give me a clue?
+ (void)load
{
RKAtomicMemoryBarrier(); // Extra cautious
if(RKCacheLoadInitialized== 1) { return; }
if(RKAtomicCompareAndSwapInt(0, 1, &RKCacheLoadInitialized)) {
if((cacheMapKeyCallBacks= dlsym(RTLD_DEFAULT, "NSIntegerMapKeyCallBacks"))
== NULL) { cacheMapKeyCallBacks= dlsym(RTLD_DEFAULT, "NSIntMapKeyCallBacks"); }
#ifdef ENABLE_MACOSX_GARBAGE_COLLECTION
id garbageCollector = objc_getClass("NSGarbageCollector");
if(garbageCollector != NULL) {
if([garbageCollector defaultCollector] != NULL) {
id pointerFunctions = objc_getClass("NSPointerFunctions");
RKCacheIntegerKeyPointerFunctions = [pointerFunctions
pointerFunctionsWithOptions:NSPointerFunctionsIntegerPersonality];
RKCacheIntegerKeyPointerFunctions.acquireFunction =
intPointerFunctionsAcquire;
RKCacheIntegerKeyPointerFunctions.descriptionFunction =
intPointerFunctionsDescription;
RKCacheIntegerKeyPointerFunctions.hashFunction=
intPointerFunctionsHash;
RKCacheIntegerKeyPointerFunctions.isEqualFunction =
intPointerFunctionsIsEqual;
RKCacheIntegerKeyPointerFunctions.relinquishFunction =
intPointerFunctionsRelinquish;
RKCacheIntegerKeyPointerFunctions.sizeFunction=
intPointerFunctionsSize;
RKCacheObjectValuePointerFunctions = [pointerFunctions
pointerFunctionsWithOptions:(NSPointerFunctionsZeroingWeakMemory|
NSPointerFunctionsObjectPersonality)];
[[garbageCollector defaultCollector]
disableCollectorForPointer:RKCacheIntegerKeyPointerFunctions];
[[garbageCollector defaultCollector]
disableCollectorForPointer:RKCacheObjectValuePointerFunctions];
}
}
#endif // ENABLE_MACOSX_GARBAGE_COLLECTION
}
}
____
From: Bill Bumgarner
To: Chris Idou
Cc: cocoa-dev@lists.apple.com
Sent: Thu, 24 September, 2009 12:36:04 PM
Subject: Re: Snow Leopard: unsupported PointerFunctions configuration was
requested
On Sep 23, 2009, at 7:09 PM, Chris Idou wrote:
Very early in application startup, even before main() is called, I get the
following error:
>
> 2009-09-24 12:07:11.462 MyApp[5534:a0f] *** An unsupported PointerFunctions
> configuration was requested, probably for use by NSMapTable, NSHashTable, or
> NSPointerArray. The requested configuration fails due to integer personality
> not using opaque memory
>
> It happens twice on startup, and below are the stack traces when it happens:
>
> What does it all mean?
>
>
> #00x7fff84b342f7 in NSLog
> #10x7fff84ac84d1 in +[NSConcretePointerFunctions
> initializeSlice:withOptions:]
> #20x7fff84aea796 in -[NSConcretePointerFunctions initWithOptions:]
> #30x7fff84aea74d in +[NSPointerFunctions pointerFunctionsWithOptions:]
> #40x10004b633 in +[RKCache load]
One change to the NSPointerFunction based collection classes (which support a
number of things that NSArray & the like do not) between Leopard and Snow
Leopard was to tighten up the validation of the pointer function validation
such that a handful of non-sensical configurations are detected that weren't
before.
I.e. RKCache is trying to create a set of pointer functions that doesn't make
sense (and for whose behavior is undefined).
As someone else pointed out, RK is most likely RegexKit and, thus, the source
is available and can be fixed. Given the relatively common use of RK*, please
provide a patch to the author so it can be fixed upstream.
b.bum
Get more done like never before with Yahoo!7 Mail. Learn more.
__
Get more done like never before with Yahoo!7 Mail.
Learn more: http://au.overview.mail.yahoo.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: Appropriate dealloc and finalize actions
On Oct 12, 2009, at 8:25 AM, Jens Alfke wrote: > On Oct 12, 2009, at 4:26 AM, Karolis Ramanauskas wrote: > >> As you can see each box has one or more little "inputs" and "outputs" in >> fact these inputs and outputs are instances of one class (KROMPort). When I >> drag a connection from output to an input, I set each "port's" connection >> property to point to another "port". So Input points to Output and Output >> points to Input. Only one-to-one relationships are allowed (one connection >> per input/output). > > It's best not to do this kind of cleanup in dealloc/finalize. Instead, have > an explicit method like -disconnect that's called to remove the object from > the graph. That way you're in direct control of removing objects. And yes, > when [foo disconnect] is called it will need to tell its connected object to > clear the connection to itself. It’s also good in cases like this not to represent ownership of objects via the graph, but by some outer container. For example, if I were creating an application with documents that consisted of connected nodes, I’d have the document own (retain) the nodes and have the nodes just reference (assign) each other. That would allow me to keep “stop referencing node A from node B” distinct from “remove node A from the document.” The latter may imply the former, but the former generally doesn’t imply the latter. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: SelectedRowIndexes
Graham's solution is excellent. You could also have a couple of buttons between the two lists, with arrows (one right, one left) to move items to and remove them from, the sub-group list. And an additional pair to the right of the sub-group list (one up, one down) allowing reordering. This would allow a completely "control" (e.g. Buttons and lists) way of doing it. Finally, you could present the sub-group info above or below the sub-group list. > From: Graham Cox > Date: Thu, 15 Oct 2009 17:45:59 +1100 > To: Peter Hudson > Cc: > Subject: Re: SelectedRowIndexes > > > On 15/10/2009, at 5:24 PM, Peter Hudson wrote: > >> The user has an NSTableView in which is presented a number of >> elements ( 1 per row, typically 20 - 200 rows ) out of which they >> need to produce a number of sequenced sub groups. >> The user presses the 'Start Sub Group' key ( i.e. clear the sequence >> selection cache ) and then by picking a possible sequence ( one row/ >> element at a time ) they build a group. >> As the group is selected, further info about that particular sub >> group of elements is presented. >> They can close/abort/store a given sub group at any time. > > > That does make some sense, but falls into the trap of not avoiding > modes (in this case a "build subgroup" mode). > > If it were me, I'd have two lists: a master list and a sub-group list. > Drag items from the master to the group, and allow drag reordering of > the group to set the sequence. As a short-cut you could allow a double- > click on an item in the master list to add it to the end of the > subgroup, which almost duplicates the current approach but with a > double-click instead of a single, yet avoids the mode. > > Just a thought ;-) > > --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/chris%40clwill.com > > This email sent to ch...@clwill.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: Problem writing unit tests - "Incompatible types in initialization"
On Oct 25, 2009, at 1:08 PM, Ian Piper wrote: > testConverter = Converter.new; Dot syntax is for accessing state (properties) of objects, not for invoking behavior. This should be either testConverter = [Converter new]; or testConverter = [[Converter alloc] init]; to indicate that it’s invoking behavior rather than accessing state on the Converter class itself. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Odd window behavior LSUIElement=1
I have an app running without dock icon, and it exhibits odd behavior. Let's say I have existing applications running, "A" with a window on top of application "B". My program pops up a window and forces it to the front. So I have "MyApp" on top of "A" on top of "B". Now I click on Application "A", bringing it to the front. Now I have "A" on top of "MyApp" on top of "B". Now I quit "A". As I quit "A", "MyApp" suddenly pops behind "B". The same thing happens whether I programmatically forced "MyApp" to the front, or whether I bring it to the front by clicking on it. It doesn't happen when I run LSUIElement=0 Can anyone tell me what is going on? __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.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
Of FontManagers and responder chains
I am trying to implement choosing default fonts, as found in the OS-X 10.6 version of TextEdit's Preferences dialog. i.e., you press a button, and it brings up the Font Chooser, and you select a font and it shows the name of the font in the dialog. Now I thought I had it all working, but the trouble is I have the option in System Preferences, Keyboard prefs called "Full Keyboard Access" to allow focus in all controls. When I turn that option off, nothing in my dialog has explicit focus, and so it seems, changeFont: is never called, presumably because my delegate object for the window is not in the responder chain. I've tried to explicitly make my window the first responder, and to make the button the first responder, but apparently no dice, they are not allowed to be first responder, or even in the responder chain, because they have no focusable objects. I could change the action for NSFontManager, but this breaks things for other uses of the NSFontManager, and seems like a bad idea. I notice that TextEdit when you press a change font in TextEdit Preferences it explicitly makes the text fields to NOT be first responder, so it seems like it is also using the first responder mechanism to make it work. But I can't seem to. __ Get more done like never before with Yahoo!7 Mail. Learn more: http://au.overview.mail.yahoo.com/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: why use pow(x, 2)?
How completely rude of you, Greg, to confuse a good argument with facts :)
But it still does leave the style question: is pow(x,2) clearer than x*x?
In the case from the OP, I think that the pow is clearer, because it is
implementing an algorithm that calls specifically for x-squared. And in the
case where x is not a simple variable, but rather an expression, it's even
more clear (and less prone to typing errors).
My $0.02...
> From: Greg Parker
> Subject: Re: why use pow(x, 2)?
>
> This is easy to test empirically. In this simple case, the compiler
> does optimize pow(x, 2) directly to a single-instruction x*x.
>
> % cat test.c
> #include
> int main(int argc, char **argv) {
> return pow(argc, 2);
> }
> % cc -O3 test.c -o - -S
> [...]
> _main:
> LFB17:
> pushq %rbp// build stack frame
> LCFI0:
> movq %rsp, %rbp // build stack frame
> LCFI1:
> cvtsi2sd %edi, %xmm0 // convert int argc to float
> mulsd %xmm0, %xmm0 // pow(argc, 2)
> cvttsd2si %xmm0, %eax // convert float->int for return
> leave
> ret
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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
NSOperationQueue for NSXMLParser object
Let me start with what I'm trying to accomplish. I have an app that
is constantly running an animation, which's attributes are determined
after downloading and parsing some XML. The XML is parsed at a given
interval using an NSTimer. As expected, sometimes when the XML is
being parsed the animation stalls. I've determined that I'll need to
do the parsing on a separate thread from the main thread (Unless there
is a better solution?).
I'm running into some problems using NSOperation with a NSTimer. Here
are some portions of the code:
//From the App delegate
//create a timer to parse the XML
self.timer = [NSTimer scheduledTimerWithTimeInterval: 2
target: self
selector: @selector(parseXML:)
userInfo: nil
repeats: YES];
//Parse the XML
- (void) parseXML:(NSTimer *)timer {
XMLParser *parser = [[XMLParser alloc] initWithURL:@"Location of the
XML" delegate:self];
[queue addOperation:parser];
[parser release]; //this is where my problem is
}
XMLParser is a subclass of NSOperation, here is some of that class.
- (void)main {
[self startParsing]; //this method starts the parsing
}
When the parsing is complete, I use this code to send the result
back to the delegate object:
//sends the result back to the delegate
if ( [delegate respondsToSelector:@selector(status:)] ) {
[delegate status:self.result];
}
This code works, as long as I don't release the parser object after
this:
[queue addOperation:parser];
[parser release];
However, if I don't release the Parser object, the app leaks like
crazy, since it's creating new Parser objects on each fire of the
timer. When I release the parser, the second fire the app terminates:
"Program received signal: “EXC_BAD_ACCESS”.".
Thank you for any advice!
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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: Hardware UUID
Please remember that gethostuuid() has all the same caveats mentioned in TN1103. .chris On 8 Nov 2009, at 3:13 PM, Grigutis, John A wrote: Also look at gethostuuid: int gethostuuid(uuid_t id, const struct timespec *wait) I don't think it was around when that technote was last updated. -- John Anthony Grigutis Systems Analyst/Programmer/Mac Specialist Indiana University : UITS : Communication & Support : SDD On Nov 8, 2009, at 3:53 PM, Brent Smith wrote: Thanks Jerry, This is exactly what I needed. On Nov 8, 2009, at 12:14 PM, Jerry Krinock wrote: On 2009 Nov 08, at 11:05, Brent Smith wrote: Is there a simple way to obtain and hardware uuid of the mac? Read this first: http://developer.apple.com/mac/library/technotes/tn/tn1103.html ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/liarepmi %40hack3r.com This email sent to liare...@hack3r.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/grigutis %40indiana.edu This email sent to grigu...@indiana.edu ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/ctp%40apple.com This email sent to c...@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: Should I learn CoreData for this project?
On Nov 10, 2009, at 8:15 AM, Ruotger Skupin wrote: > True, but as I found out the hard way, using Core Data multithreaded and > Bindings at the same time is a very bad idea and will lead you into a world > of pain. This is part of the reason for Core Data’s recommended thread isolation policy: Use a separate NSManagedObjectContext per thread. If you follow this recommendation, instead of attempting to share objects from a single context between multiple threads, then there should be no problems using Core Data multithreaded. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Performance
Hi all, A performance related question: I've written a Cocoa app that continuously reads data from a USB device and plots it on a graph. It consists of three of my own classes. The first class is the model that submits asynchronous bulk reads to the USB device. The callback for these reads copies the received data from the buffer asynchronous filled by the request and into an NSData object that is allocated and added to an NSArray. The buffer that the asynchronous request was using is recycled into a new asynchronous request. So the pool of buffers is allocated once and reused while the NSData objects are continuously allocated and deallocated (when released by the controller). The second class is the view that uses an NSTimer to call [self setNeedsDisplay:YES] 24 times per second. The drawRect: method requests data from the data source (the third, controller class) and draws it with OpenGL. The third class is the controller class that serves as a data source for the view. When the view calls the giveMeData: method, the controller class calls the buffer: method in the model class, which removes and returns the oldest NSData object in its NSArray. The controller then does some computations on it and passes it off to the view in a new NSData object. The application runs pretty well, and running it through the Leaks instrument there are no leaks except for 16-bytes when the application is first starting caused by IOUSBLib. However, looking at it in the Activity Monitor, the real memory used starts off at 25 MB and steadily grows to 250+ MB while the virtual memory starts off at about the same and steadily grows to about the same or sometimes close to 500MB, over the course of several minutes. This especially happens if I don't move the mouse for a while, or don't have the application in focus. As soon as a move the mouse or bring the application into focus, it's as if an autorelease pool is drained and the memory drops back down to 30-40MB real and 30-40MB virtual. This is annoying since the application hangs for 5 seconds or so when this memory draining is occurring. Has anyone dealt with this before? Any ideas on what could be causing this and how to work around it? Another related question: The number of asynchronous requests that the driver submits at once and the maximum number of objects in the NSArray are set to some constant. Now in theory, the requests should be added and drained "simultaneously" (i.e. by separate threads), but if I make this constant larger, I notice a greater delay between when data entering the USB device and being plotted. It's as if the asynchronous requests are being filled and added to the NSArray all in a group, and then drained by the view all in a group... maybe? Is this something subtle from the way run loops work? Is the USB asynchronous bulk read run loop source on the same thread as the view class's drawRect method? My apologies for the length of this post and if the USB stuff isn't strictly Cocoa related, but since it's so intertwined with Cocoa classes, I figured this was the best place to ask. Advice is much appreciated. Please CC me on replies. Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
Advice about crash requested
I've got a user getting the following crash, and I don't know what to make of it because it all happens in Apple code. Has anyone got any advice? Process: XXX [242] Path:/Applications/XXX.app/Contents/MacOS/XXX Identifier: XXX Version: 1.19 (19) Code Type: X86-64 (Native) Parent Process: launchd [180] Date/Time: 2009-11-15 23:34:52.052 -0500 OS Version: Mac OS X 10.6.2 (10C540) Report Version: 6 Exception Type: EXC_BAD_ACCESS (SIGSEGV) Exception Codes: KERN_INVALID_ADDRESS at 0x000b Crashed Thread: 0 Dispatch queue: com.apple.main-thread Application Specific Information: objc_msgSend() selector name: invalidate NSTextAttachmentImageView objc[242]: garbage collection is ON Thread 0 Crashed: Dispatch queue: com.apple.main-thread 0 libobjc.A.dylib 0x7fff85a4210a objc_msgSend + 22 1 com.apple.AppKit 0x7fff8169871e -[NSImageCell(NSPrivateAnimationSupport) _stopAnimation] + 43 2 com.apple.AppKit 0x7fff819b60c2 -[NSImageView finalize] + 26 3 libobjc.A.dylib 0x7fff85a48f67 finalizeOneObject + 48 4 libauto.dylib 0x7fff85b113d5 Auto::foreach_block_do(auto_zone_cursor*, void (*)(void*, void*), void*) + 85 5 libobjc.A.dylib 0x7fff85a48b5c batchFinalize + 64 6 libobjc.A.dylib 0x7fff85a48e00 batchFinalizeOnMainThread + 87 7 libSystem.B.dylib 0x7fff879e3ce8 _dispatch_call_block_and_release + 15 8 libSystem.B.dylib 0x7fff879c287a _dispatch_queue_drain + 251 9 libSystem.B.dylib 0x7fff879c3127 _dispatch_queue_serial_drain_till_empty + 9 10 libSystem.B.dylib 0x7fff879f5e4c _dispatch_main_queue_callback_4CF + 37 11 com.apple.CoreFoundation 0x7fff86c99b00 __CFRunLoopRun + 2560 12 com.apple.CoreFoundation 0x7fff86c98c2f CFRunLoopRunSpecific + 575 13 com.apple.HIToolbox 0x7fff865b9a4e RunCurrentEventLoopInMode + 333 14 com.apple.HIToolbox 0x7fff865b9853 ReceiveNextEventCommon + 310 15 com.apple.HIToolbox 0x7fff865b970c BlockUntilNextEventMatchingListInMode + 59 16 com.apple.AppKit 0x7fff815f31f2 _DPSNextEvent + 708 17 com.apple.AppKit 0x7fff815f2b41 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 155 18 com.apple.AppKit 0x7fff815b8747 -[NSApplication run] + 395 19 XXX 0x00010eea main + 76 20 XXX 0x00010e7c start + 52 Thread 1: Dispatch queue: com.apple.libdispatch-manager 0 libSystem.B.dylib 0x7fff879c0bba kevent + 10 1 libSystem.B.dylib 0x7fff879c2a85 _dispatch_mgr_invoke + 154 2 libSystem.B.dylib 0x7fff879c275c _dispatch_queue_invoke + 185 3 libSystem.B.dylib 0x7fff879c2286 _dispatch_worker_thread2 + 244 4 libSystem.B.dylib 0x7fff879c1bb8 _pthread_wqthread + 353 5 libSystem.B.dylib 0x7fff879c1a55 start_wqthread + 13 Thread 2: 0 libSystem.B.dylib 0x7fff879a7e3a mach_msg_trap + 10 1 libSystem.B.dylib 0x7fff879a84ad mach_msg + 59 2 com.apple.CoreFoundation 0x7fff86c997a2 __CFRunLoopRun + 1698 3 com.apple.CoreFoundation 0x7fff86c98c2f CFRunLoopRunSpecific + 575 4 com.apple.Foundation 0x7fff82021a24 -[NSRunLoop(NSRunLoop) runMode:beforeDate:] + 270 5 au.com.tech.CocoaTools0x000101f152af -[InfiniteThread main] + 171 6 AutoFilerPlugin 0x00010352040c -[FilerExecutor main] + 72 7 com.apple.Foundation 0x7fff81fe6e99 __NSThread__main__ + 1429 8 libSystem.B.dylib 0x7fff879e0f8e _pthread_start + 331 9 libSystem.B.dylib 0x7fff879e0e41 thread_start + 13 Thread 3: Dispatch queue: Garbage Collection Work Queue 0 libSystem.B.dylib 0x7fff879e29ee __semwait_signal + 10 1 libSystem.B.dylib 0x7fff879e67f1 _pthread_cond_wait + 1286 2 libobjc.A.dylib 0x7fff85a48cd8 batchFinalizeOnTwoThreads + 250 3 libauto.dylib 0x7fff85b0b1e7 Auto::Zone::invalidate_garbage(unsigned long, unsigned long const*) + 71 4 libauto.dylib 0x7fff85afbbd1 auto_collect_internal(Auto::Zone*, unsigned int) + 481 5 libauto.dylib 0x7fff85afc16d auto_collection_work(Auto::Zone*) + 157 6 libSystem.B.dylib 0x7fff879e3ce8 _dispatch_call_block_and_release + 15 7 libSystem.B.dylib 0x7fff879c287a _dispatch_queue_drain + 251 8 libSystem.B.dylib 0x7fff879c26dc _dispatch_queue_invoke + 57 9 libSystem.B.dylib 0x7fff879c2286 _dispatch_worker_thread2 + 244 10 libSystem.B.dylib 0x7fff879c1bb8 _pthread_wqthread + 353 11 libSy
tiffs on pasteboard
I'm trying to read a TIFF off the pasteboard, but it doesn't seem to work for me. What I'm doing is going into Finder, finding a TIFF file and Command-C it. Among other types I then find on the pasteboard are: NeXT TIFF v4.0 pasteboard type and public.tiff If I take either of those from the pasteboard and write the NSData object to a file: [content writeToFile:@"/tmp/foo.tiff"atomically:YES]; or else if I make an NSImage from the data: [[NSImagealloc] initWithData:content] and display it in an NSImageView then I always see the generic TIFF icon, i.e. the one with a photo of a child scattered on the table, with a loupe on top of it. The same thing happens for JPEGs, GIFS and PNGs, but not for com.apple.icns. If I take the com.apple.icns, it displays correctly. What am I doing wrong? __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Advice about crash requested
Mainly stuff to do with examining files, and on the odd occasion it needs to do anything in the GUI it executes it on the main thread. From: David Duncan To: Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 11:21:58 AM Subject: Re: Advice about crash requested On Nov 19, 2009, at 4:18 PM, Chris Idou wrote: > I've got a user getting the following crash, and I don't know what to make of > it because it all happens in Apple code. Has anyone got any advice? What are you doing on Thread 2 ([FilerExecutor main])? -- David Duncan Apple DTS Animation and Printing __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
If I double click the tiff in Finder, and open it in Preview, click it in Preview, Command-A, Command-C, the same thing happens. From: Kyle Sluder To: Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 11:36:54 AM Subject: Re: tiffs on pasteboard On Thu, Nov 19, 2009 at 4:33 PM, Chris Idou wrote: > I'm trying to read a TIFF off the pasteboard, but it doesn't seem to work for > me. What I'm doing is going into Finder, finding a TIFF file and Command-C it. You've copied the file icon. If you want to copy the file's data, you need to open it in Preview and copy that. --Kyle Sluder __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
I notice that at least com.apple.icns works when I copy it from Preview, but it doesn't work when just copying from Finder. But either way TextEdit seems to be able to get the right thing if I paste into it. From: Chris Idou To: Kyle Sluder Cc: cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 11:43:17 AM Subject: Re: tiffs on pasteboard If I double click the tiff in Finder, and open it in Preview, click it in Preview, Command-A, Command-C, the same thing happens. From: Kyle Sluder To: Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 11:36:54 AM Subject: Re: tiffs on pasteboard On Thu, Nov 19, 2009 at 4:33 PM, Chris Idou wrote: > I'm trying to read a TIFF off the pasteboard, but it doesn't seem to work for > me. What I'm doing is going into Finder, finding a TIFF file and Command-C it. You've copied the file icon. If you want to copy the file's data, you need to open it in Preview and copy that. --Kyle Sluder __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/idou747%40yahoo.com This email sent to idou...@yahoo.com __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
OK, its working now as far as copying from Preview. Not sure what happened there. But I'm still confused about one thing: If copying from Finder actually copies the file icon, and not the actual tiff, how come it pastes ok into TextEdit? From: Jens Alfke To: Chris Idou Cc: Kyle Sluder ; cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 12:11:34 PM Subject: Re: tiffs on pasteboard On Nov 19, 2009, at 4:43 PM, Chris Idou wrote: > If I double click the tiff in Finder, and open it in Preview, click it in > Preview, Command-A, Command-C, the same thing happens. That's weird. Can you show us the code that reads the image data off the pasteboard? —Jens __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Advice about crash requested
This will help if I'm ever able to repeat it. Right now its just a user report and I can't repeat it. From: Sean McBride To: Jens Alfke ; Chris Idou Cc: cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 12:03:33 PM Subject: Re: Advice about crash requested On 11/19/09 4:54 PM, Jens Alfke said: >On Nov 19, 2009, at 4:18 PM, Chris Idou wrote: > >> Thread 0 Crashed: Dispatch queue: com.apple.main-thread >> 0 libobjc.A.dylib 0x7fff85a4210a objc_msgSend + 22 >> 1 com.apple.AppKit 0x7fff8169871e -[NSImageCell >(NSPrivateAnimationSupport) _stopAnimation] + 43 >> 2 com.apple.AppKit 0x7fff819b60c2 -[NSImageView >finalize] + 26 > >Looks like NSImageCell is calling a bogus object, probably one that's >already been freed. That's sort of unusual for a GC'd app -- usually this >indicates a ref-counting bug. You're not calling AppKit from a >background thread, are you? > >Turning on NSZombieEnabled might help shed some light on this. It won't, because NSZombieEnabled does nothing in GC apps. :( >There's a >"So, You've Crashed In objc_msgsend" FAQ out there on teh interwebz >somewhere that goes into more detail on troubleshooting. It's here: <http://www.sealiesoftware.com/blog/archive/2008/09/22/ objc_explain_So_you_crashed_in_objc_msgSend.html> -- Sean McBride, B. Engs...@rogue-research.com Rogue Researchwww.rogue-research.com Mac Software Developer Montréal, Québec, Canada __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
But what are the rules and conventions? Why would Finder put the icon in the pasteboard on the assumption that is what the user wants, but TextEdit would paste in the actual file on the assumption that is what the user wants? It seems schizophrenic. I just want a list of rules of how to use the Pasteboard in a way that conforms to what users expect. If I start ignoring the TIFF part of the pasteboard in favor of the URL on the assumption that is what the user wants, why is the tiff there anyway? And will that get me into trouble in other cases? From: Jens Alfke To: Chris Idou Cc: Kyle Sluder ; cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 12:29:51 PM Subject: Re: tiffs on pasteboard On Nov 19, 2009, at 5:24 PM, Chris Idou wrote: > But I'm still confused about one thing: If copying from Finder actually > copies the file icon, and not the actual tiff, how come it pastes ok into > TextEdit? The Finder also puts the location of the file on the pasteboard, and I believe TextEdit sees that, recognizes that it's an image file, and reads the file into an image. You can do the same fairly easily. —Jens __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
asl_log aborts the program?
I've got a report from a user of my program crashing. In the console they are getting this: 11/19/09 3:08:46 PM[0x0-0x18a18a]Progname[8699]Progname(8699,0x1167b) malloc: *** error for object 0x100563870: pointer being freed was not allocated 11/19/09 3:08:46 PM[0x0-0x18a18a] Progname[8699]*** set a breakpoint in malloc_error_break to debug And in the crash report: Process: Progname [11128] Path:/Applications/Progname.app/Contents/MacOS/Progname Identifier: Progname Version: 1.19 (19) Code Type: X86-64 (Native) Parent Process: launchd [180] Date/Time: 2009-11-19 17:40:42.570 -0500 OS Version: Mac OS X 10.6.2 (10C540) Report Version: 6 Exception Type: EXC_CRASH (SIGABRT) Exception Codes: 0x, 0x Crashed Thread: 2 Application Specific Information: abort() called objc[11128]: garbage collection is ON Thread 0: Dispatch queue: com.apple.main-thread 0 libSystem.B.dylib 0x7fff879a7e3a mach_msg_trap + 10 1 libSystem.B.dylib 0x7fff879a84ad mach_msg + 59 2 com.apple.CoreFoundation 0x7fff86c997a2 __CFRunLoopRun + 1698 3 com.apple.CoreFoundation 0x7fff86c98c2f CFRunLoopRunSpecific + 575 4 com.apple.HIToolbox 0x7fff865b9a4e RunCurrentEventLoopInMode + 333 5 com.apple.HIToolbox 0x7fff865b9853 ReceiveNextEventCommon + 310 6 com.apple.HIToolbox 0x7fff865b970c BlockUntilNextEventMatchingListInMode + 59 7 com.apple.AppKit 0x7fff815f31f2 _DPSNextEvent + 708 8 com.apple.AppKit 0x7fff815f2b41 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 155 9 com.apple.AppKit 0x7fff815b8747 -[NSApplication run] + 395 10 Progname 0x00010eea main + 76 11 Progname 0x00010e7c start + 52 Thread 1: Dispatch queue: com.apple.libdispatch-manager 0 libSystem.B.dylib 0x7fff879c0bba kevent + 10 1 libSystem.B.dylib 0x7fff879c2a85 _dispatch_mgr_invoke + 154 2 libSystem.B.dylib 0x7fff879c275c _dispatch_queue_invoke + 185 3 libSystem.B.dylib 0x7fff879c2286 _dispatch_worker_thread2 + 244 4 libSystem.B.dylib 0x7fff879c1bb8 _pthread_wqthread + 353 5 libSystem.B.dylib 0x7fff879c1a55 start_wqthread + 13 Thread 2 Crashed: 0 libSystem.B.dylib 0x7fff87a7784d usleep$NOCANCEL + 0 1 libSystem.B.dylib 0x7fff87a96e3c abort + 93 2 libSystem.B.dylib 0x7fff879ae155 free + 128 3 libSystem.B.dylib 0x7fff879fa16e asl_set_query + 572 4 libSystem.B.dylib 0x7fff87a16239 asl_send + 824 5 libSystem.B.dylib 0x7fff87a15e56 asl_vlog + 570 6 libSystem.B.dylib 0x7fff87a15c19 asl_log + 153 7 Progname.CocoaTools0x000104567101 -[Log vlogLevel:format:arguments:] + 270 8 Progname.CocoaTools0x00010456730c -[Log notice:] + 164 etc Does this look like an Apple bug? It looks to me like the asl subsystem is freeing something it shouldn't and aborting the program. __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
I thought the solution to big files was to use pasteboard promises... not to setup a pasteboard which is conflicted about what item it is trying to store. At least that's what I find documented by Apple. From: Alexander Spohr To: Chris Idou Cc: Jens Alfke ; cocoa-dev@lists.apple.com Sent: Fri, 20 November, 2009 7:00:18 PM Subject: Re: tiffs on pasteboard Am 20.11.2009 um 03:04 schrieb Chris Idou: > But what are the rules and conventions? Why would Finder put the icon in the > pasteboard on the assumption that is what the user wants, but TextEdit would > paste in the actual file on the assumption that is what the user wants? It > seems schizophrenic. Yeah. Copying a 20MB image to the pasteboard would be what the user wants. Or better one Gig of video. I think not. You think like a user (who would assume that the file is really on the pasteboard) and not like a developer (who should know that the pasteboard content has to be lightweight). > I just want a list of rules of how to use the Pasteboard in a way that > conforms to what users expect. If I start ignoring the TIFF part of the > pasteboard in favor of the URL on the assumption that is what the user wants, > why is the tiff there anyway? The TIFF is a preview. The URL is the meat. > And will that get me into trouble in other cases? That depends only on the pasteboard provider. There is no strict rule. The provider does not know the recipient and therefore puts everything on the pasteboard that will satisfy the task the provider offers. atze __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: asl_log aborts the program?
I just noticed something I didn't see before, namely that the doco says to call asl_open once for each thread, which I'm not doing. I guess this most likely is the cause. From: Jeremy Pereira To: Chris Idou ; Cocoa Forum Cc: Jeremy Pereira Sent: Fri, 20 November, 2009 9:35:01 PM Subject: Re: asl_log aborts the program? On 20 Nov 2009, at 06:33, Chris Idou wrote: > > > I've got a report from a user of my program crashing. > > In the console they are getting this: > > 11/19/09 3:08:46 PM[0x0-0x18a18a]Progname[8699]Progname(8699,0x1167b) > malloc: *** error for object 0x100563870: pointer being freed was not > allocated > 11/19/09 3:08:46 PM[0x0-0x18a18a] Progname[8699]*** set a breakpoint in > malloc_error_break to debug > > > Thread 2 Crashed: > 0 libSystem.B.dylib 0x7fff87a7784d usleep$NOCANCEL + 0 > 1 libSystem.B.dylib 0x7fff87a96e3c abort + 93 > 2 libSystem.B.dylib 0x7fff879ae155 free + 128 > 3 libSystem.B.dylib 0x7fff879fa16e asl_set_query + 572 > 4 libSystem.B.dylib 0x7fff87a16239 asl_send + 824 > 5 libSystem.B.dylib 0x7fff87a15e56 asl_vlog + 570 > 6 libSystem.B.dylib 0x7fff87a15c19 asl_log + 153 > 7 Progname.CocoaTools0x000104567101 -[Log > vlogLevel:format:arguments:] + 270 > 8 Progname.CocoaTools0x00010456730c -[Log notice:] + 164 > etc > > Does this look like an Apple bug? It looks to me like the asl subsystem is > freeing something it shouldn't and aborting the program. > Normally when I see crashes involving calls to methods with format strings, I immediately assume that the arguments following the the format string do not match the format specifiers in the format string. e.g. if -[Log notice:] is declared like this -(void) notice: (NSString*) format, ...; [log notice: @"blah blah blah %@ %s", "blah"]; The above line has an NSObject missing from its argument list. There's also [log notice: someNSStringofIndeterminateOrigin]; which may crash if the string contains format specifiers e.g. if it is a URL with % escapes in it. So you probably want to check your calls to -[Log notice:]. > > > __ > Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. > Enter now: http://au.docs.yahoo.com/homepageset/ > ___ > > Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) > > Please do not post admin requests or moderator comments to the list. > Contact the moderators at cocoa-dev-admins(at)lists.apple.com > > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/cocoa-dev/adc%40jeremyp.net > > This email sent to a...@jeremyp.net __ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email __ __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: tiffs on pasteboard
Well... now that in 10.6 the pasteboard types are UTI types, I'm not sure that one can say anymore that there is no mapping from files to pasteboard types. And 10.6 Finder puts a public.tiff into the pasteboard in this case, but I would argue it is the wrong tiff. On Nov 20, 2009, at 1:35 AM, Chris Idou wrote: > I thought the solution to big files was to use pasteboard promises... not to > setup a pasteboard which is conflicted about what item it is trying to > store. At least that's what I find documented by Apple. You're correct that promises would overcome some of the inefficiencies (although I can envision other situations where the size would be a problem.) I suspect the real reason is that there isn't a straightforward mapping between file types and pasteboard types. There is not in general any table that says "a file with extension ".foo" contains data that can be stored in a pasteboard using type 'FooPasteboardType'"; so the Finder would not be able to decide in all cases what pasteboard type to use for the contents of the dragged/copied file.* You could argue that images are a useful special case; but I suspect that the engineers responsible decided that it was better to remain consistent.. (And actually, it doesn't even work for all image types. The old Mac 'PICT' format is infamous for having a slightly different representation in memory than in a file, so you can't just slurp a PICT file into the pasteboard and have it be useable.) —Jens * Especially for package file types that are actually directories; would the rule be to create a Zip archive? :-P __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
core data and migration
If I have versions of my model, A, B and C, do I only need a mapping model from A->B and B->C, and if someone wants to upgrade A->C is core data smart enough to do the two migrations, or do I need a separate mapping model from A->C ? __ Win 1 of 4 Sony home entertainment packs thanks to Yahoo!7. Enter now: http://au.docs.yahoo.com/homepageset/ ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Better sorting using threads?
On 12 Mar 2010, at 20:20, Kyle Sluder wrote: > On Fri, Mar 12, 2010 at 11:55 AM, Andrew James > wrote: >> I've made the assumption that NSArray wraps an old school contiguous C-style >> array ( ala C++ std:::vector ) > > This assumption is false. There was a good analysis of NSArrays at <http://ridiculousfish.com/blog/archives/2005/12/23/array/> Cheers, Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
Re: App modal window and secondary thread
On Mar 20, 2010, at 4:24 AM, Philippe Sismondi wrote: Had I not been compelled to support OS X 10.4 I would have re- written the whole thing to use something better than threads - GCD or NSOperationQueue capabilities are said to be safer, although I have not yet learned them. Let's nip this in the bud: NSOperation and GCD are not "safer than threads" -they *are* threads. They are a more convenient API but they're still concurrent programming -and that's what's hard about programming against threads, not the API itself. In particular, they *do not* insulate you from AppKit or Core Data's threading model. And a particular operation or dispatch to a concurrent queue can be served by any thread, without any strong binding to one thread (modulo the binding of the main queue to the main thread). I would also not invoke abortModal on a background thread just as a matter of cleanliness. I'd implement a delegate/callback/notification for my background processor to tell whatever is controlling it that it's done. That's what would invoke abortModal - and it would do that on the main thread/queue. -- Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Allowing incoming connections.
On Mar 24, 2010, at 1:15 AM, Ken Thomases wrote: > You can work around this by signing your app properly as part of building it, > but I'm not sure it's common to sign development builds. It’s somewhat common to sign debug builds of Internet applications, using a self-signing certificate: http://www.red-sweater.com/blog/514/development-phase-code-signing Note that the article predates some Xcode support for code signing. With modern versions of Xcode — Xcode 3.1 and later, I believe — you don’t need to use a Run Script build phase to do the code signing. Specify the identity in your build settings just like you would to debug your application on the iPhone and Xcode will sign it. Release builds of applications that listen for connections and applications that use the Keychain really should be signed using a real certificate, to avoid prompting the user to re-authorize on every update to the application. — Chris ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
NSImageView Will Not Alias Images
I am trying to get NSImageView to alias dropped images, but it
refuses. Just spent an hour looking and trying several variations to
no avail. Here's what I have done in a subclass of NSImageView.
-(void)drawRect:(NSRect)rect
{
[[NSGraphicsContext currentContext]
setImageInterpolation:NSImageInterpolationHigh];
[[NSGraphicsContext currentContext] setShouldAntialias:YES];
[super drawRect:rect];
}
I have assigned the ImageView in IB my subclass and used NSLog to
verify that it is drawing through the above drawRect method of my
subclass - but the images (PNG screen captures) will not alias.
Any suggestions?
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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
Tricky binding and continuous update problem
I've got a problem that I had working, but my app suddenly seems broken, and I'm not sure if I did something or what. All I know is old versions of my app work, but now my code base doesn't. I've got a UI with a NSTableView at the top, and some individual fields at the bottom. Typical UI where you click on a table row and it shows you the values in the fields. Now one of the fields shown in the table is a derived value from the base fields. I have continuous update values set in the NSTextField at the bottom. So what I expect as you type in the field, the value in the table gets continuously updated also. (I'm using keyPathsForValuesAffecting etc for the derived field). But what's happening is if I type in the lower field, the value gets sent to the derived field in the table, but then Text field gets immediately set back to the original value. So the effect is you type one character and it is lost as far as the lower field is concerned, but the derived value in the table is updated. You can only ever change one character. And also the lower text field loses focus. I've put break points in all the setters and getters, and there seems like no reason for this odd behavior. I can't see how the field would get set back to the old value. If I turn off Continuously Update Value, it works sensibly, albeit not as nice since you've got to exit the field to have everything in synch. Any thoughts? ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: NSImageView Will Not Alias Images
On Apr 6, 2010, at 4:50 PM, Ken Ferry wrote: On Tue, Apr 6, 2010 at 2:58 PM, Jens Alfke wrote: On Apr 6, 2010, at 1:17 PM, Chris Tracewell wrote: I am trying to get NSImageView to alias dropped images, but it refuses. Nitpick: you mean "antialias". Aliasing is what creates the jaggies, antialiasing smooths them away. Just spent an hour looking and trying several variations to no avail. Here's what I have done in a subclass of NSImageView. I remember having to deal with this too, years ago. It's too bad AppKit hasn't added support for this yet :( The problem is that NSImageView internally keeps a scaled copy of the image. So the actual scaling that creates the aliasing isn't done in the drawRect: method at all. This was once true, but is out of date. I'd like to see a test app. For example, how do you know you aren't getting antialiasing? It may be that you just don't like the output. :-) Well - NSImageInterpolationHigh and NSImageInterpolationNone produce the same exact result - screen shot copy (control-command-shift-4) the imageView built once using NSImageInterpolationHigh and once using NSImageInterpolationNone then paste each into a Photoshop layer, align them perfectly and then turn the top layer off and on at 800% and there is not a single pixel that moves or changes color. The dropped image is roughly 1000 x 1000 (a screen shot PNG) and the imageView size is 200x200 To make sure the currentContext was correct - I log [[NSGraphicsContext currentContext] imageInterpolation] for each build and it shows the correct values 3 and 1 respectively. And then just to be super sure I I log [[NSGraphicsContext currentContext] isDrawingToScreen] inside drawRect of my NSImageView subclass and get YES. I did implement my own image sizing in drawRect and then used [self setImage] and it works great. Not sure what could be the issue. It would also be good to know what OS you are working on. -Ken I'm using 10.5.8. - XCode 3.1.2 - iMac Core 2 Duo and the app is GC. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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: Tricky binding and continuous update problem
I am using an NSArrayController.
I don't know, it seems to me like having everything always in synch is nicer.
The user can see immediately how changing one field is affecting the other. And
it used to work.
- Original Message
From: Jerry Krinock
To: Chris Idou
Sent: Wed, 7 April, 2010 10:24:18 AM
Subject: Re: Tricky binding and continuous update problem
On 2010 Apr 06, at 16:36, Chris Idou wrote:
> If I turn off Continuously Update Value, it works sensibly,
Turn it off. Look at any of Apple's Sample Code. Also, Cocoa Design Rule #1:
If something is off/on by default, don't change it unless you're knowingly
doing something weird.
> albeit not as nice since you've got to exit the field to have everything in
> synch.
That's the way most apps work.
You did not mention using an array controller. To insure data integrity in
case user tabs out, abruptly closes the window, etc., do not bind directly to
the model. Instead, bind your text fields ("detail views") to an
NSArrayController to which the table columns are also bound. As a matter of
fact, if you didn't use an array controller, you'd better look at Apple's
DepartmentAndEmployees sample code and be prepared for a little re-work. There
may be a way to make your design paradigm work, but why bother?
The way I understand it, the NSArrayController superclass NSObjectController
fulfills the same purpose of data integrity if you can't bind to an array
controller, for example if you have an Inspector in another nib.
NSObjectController seems like a heavy weight for this purpose, but it works.
For an example of this, in DepartmentAndEmployees's MyDocument.nib, look at the
NSObjectController named 'Department Controller'. It seems like it's useless
but it's not.
Probably someone else can give a more in-depth explanation.
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
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: Tricky binding and continuous update problem
How does it wreak havoc with undo? Undo seems to work still. In something like TextExpander, you change the textfield where your expansion goes, and the table in the left hand side changes as you type. That's what I was aiming for. But I've found why it suddenly stopped working. I stumbled upon the URL below. Apparently the "Automatically rearrange content" for the NSArrayController makes this all go awry. It seems like I can choose between auto rearrange content, or continuously update values, but not both. Don't know if this is a bug or if there is some reason for it. http://stackoverflow.com/questions/1028700/core-data-strange-bindings-error-on-re-opening-a-document-help - Original Message ---- From: Kyle Sluder To: Chris Idou Cc: Jerry Krinock ; cocoa-dev@lists.apple.com Sent: Wed, 7 April, 2010 11:21:28 AM Subject: Re: Tricky binding and continuous update problem On Tue, Apr 6, 2010 at 5:28 PM, Chris Idou wrote: > I don't know, it seems to me like having everything always in synch is nicer. > The user can see immediately how changing one field is affecting the other. > And it used to work. That will wreak havoc with Undo. Continuously updating values is not pretty, and it's not how text fields behave on the Mac. Rich text does behave as you describe, but you can't use bindings for that. Avoid fighting the framework. ;-) --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
Reordered columns and [tableView editColumn:0 row:ind withEvent:nil select:YES];
I've got some code that uses editColumn:0 to force the user into edit mode on the first column. But I've noticed that if user column reordering is allowed, this no longer edits the correct column. This seems odd to me because I thought Cocoa pretty much shielded the programmer from all the user reordering stuff. Even more oddly, if the user puts a checkbox as the first column, the checkbox seems to start behaving like a text box, albeit with invisible characters. Could I be doing something wrong, or is this some kind of bug? And do I have to somehow translate column 0 to find out the real column? I know Java tables have methods to translate between model and view column numbers because of user reordering, but I haven't noticed such a thing in Cocoa. ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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 Key Value Observing performance pickle
I seem to have got myself into some pickle with key value observing whereby things to do with NSTableView are freezing up for minutes at a time. Every time I pause the program there are things to do with adding and removing observers, but all of it is in Apple's code, so I'm not quite sure what I've done to cause this problem. #00x7fff8453d06b in append_referrer_no_lock #10x7fff8453ea58 in weak_register #20x7fff845366e2 in auto_assign_weak_reference #30x7fff81d6a36e in -[NSConcretePointerArray arrayGrow:] #40x7fff81c64cdd in -[NSConcretePointerArray addPointer:] #50x7fff81c6aa99 in -[NSKeyValueObservationInfo _initWithObservances:count:] #60x7fff81c70399 in _NSKeyValueObservationInfoCreateByRemoving #70x7fff81c6fcfe in -[NSObject(NSKeyValueObserverRegistration) _removeObserver:forProperty:] #80x7fff81c6fc3f in -[NSObject(NSKeyValueObserverRegistration) removeObserver:forKeyPath:] #90x7fff81c70522 in -[NSKeyValueNestedProperty object:didRemoveObservance:recurse:] #100x7fff81c7077f in -[NSKeyValueUnnestedProperty object:didRemoveObservance:recurse:] #110x7fff81c6fd5b in -[NSObject(NSKeyValueObserverRegistration) _removeObserver:forProperty:] #120x7fff81c6fc3f in -[NSObject(NSKeyValueObserverRegistration) removeObserver:forKeyPath:] #130x7fff81c70522 in -[NSKeyValueNestedProperty object:didRemoveObservance:recurse:] #140x7fff81c7077f in -[NSKeyValueUnnestedProperty object:didRemoveObservance:recurse:] #150x7fff81c6fd5b in -[NSObject(NSKeyValueObserverRegistration) _removeObserver:forProperty:] #160x7fff81c6fc3f in -[NSObject(NSKeyValueObserverRegistration) removeObserver:forKeyPath:] #170x7fff810e23c7 in -[_NSModelObservingTracker _registerOrUnregister:observerNotificationsForModelObject:] #180x7fff810e1a07 in -[_NSModelObservingTracker clearAllModelObjectObserving] #190x7fff811bc4b3 in -[_NSModelObservingTracker setIndexReferenceModelObjectArray:clearAllModelObjectObserving:] #200x7fff811bc44c in -[NSArrayController _setObjects:] #210x7fff81255d9f in -[NSArrayController _arrangeObjectsWithSelectedObjects:avoidsEmptySelection:operationsMask:useBasis:] #220x7fff811bc9ff in -[NSArrayController setContent:] #240x7fff8125ad66 in -[NSArrayDetailBinder _refreshDetailContentInBackground:] #250x7fff81c791a1 in -[NSObject(NSKeyValueObservingPrivate) _notifyObserversForKeyPath:change:] #260x7fff810de96e in -[NSController _notifyObserversForKeyPath:change:] #270x7fff811e8575 in -[NSArrayController didChangeValuesForArrangedKeys:objectKeys:indexKeys:] #280x7fff811e894b in -[NSArrayController _selectObjectsAtIndexesNoCopy:avoidsEmptySelection:sendObserverNotifications:forceUpdate:] #290x7fff81c6f43d in -[NSObject(NSKeyValueCoding) setValue:forKey:] #300x7fff81ccdf42 in -[NSObject(NSKeyValueCoding) setValue:forKeyPath:] #310x7fff8126a3ed in -[NSBinder _setValue:forKeyPath:ofObject:mode:validateImmediately:raisesForNotApplicableKeys:error:] #320x7fff8126a2a5 in -[NSBinder setValue:forBinding:error:] #330x7fff812655c8 in -[NSTableBinder tableView:didChangeToSelectedRowIndexes:] #340x7fff81261e8f in -[_NSBindingAdaptor tableView:didChangeToSelectedRowIndexes:] #350x7fff811368e2 in -[NSTableView _enableSelectionPostingAndPost] #360x7fff811f8556 in -[NSTableView mouseDown:] #370x7fff81199f1b in -[NSWindow sendEvent:] #380x7fff810cf662 in -[NSApplication sendEvent:] ___ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. 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
