I think if you have really good undo support people find autosave less of an
annoyance. Most people took advantage of the save-before-experimenting method
because several apps had bad or nonexistent undo support so saving was the only
way to reliably go back.
On Sep 21, 2012, at 10:05 AM,
official and unofficial docs
and through trial and error.
On Aug 18, 2011, at 10:08 AM, Sean McBride wrote:
On Wed, 17 Aug 2011 03:17:30 -0600, Michael Vannorsdel said:
Apologies if this has been covered in the past but my searches did not
turn up anything as specific as I'm looking
Apologies if this has been covered in the past but my searches did not turn up
anything as specific as I'm looking for.
Is there a way to refine sandbox entitlements to allow read/write access to
specific files and directories instead of just all or none? For instance, only
allowing RW to
I had this happen a couple times when dealing with threading accidents. If the
tracking rects are added from a secondary thread or the window is ordered front
by a secondary thread, then the tracking areas will be stuck in a pending
state. There's probably more reasons for this but these are
I have a CALayer that is larger than the window and I move that layer around to
view various sections of it. The layer is also masked by another CALayer. The
problem I have is if I move half of the layer out of view the mask stops
working right. As if the mask is transparent so none of the
I've been working on some 64bit code cleaning but need some advice on float
constants. Currently on 32bit builds I use -fsingle-precision-constant since
most of my code uses CGFloats and I'd like the constants to be the correct
precision. But in some places I do work with doubles and would
This is exactly why I do it for the 32bit builds. What I'm trying to avoid in
the sections that use doubles is having the constant generate a float load,
double store, double load for each constant. I'm still poking through the
assembly to see if gcc is indeed doing this but I'm also checking
When I have a process that I really don't want running more than once
(some people will run copies or use LaunchServices to launch
multiples) I register a named mach port with NSConnection. When when
an instance launches it checks for that connection and if it exists
the new instance
I would try and reset the target just after setting the action:
[myMenuItem setAction:@selector(myFunction:)];
[myMenuItem setTarget:self];
I remember a long time ago I had a problem where if I reset the item's
action, it would set the target to nil for some reason. So in
practice I
I found that I can hotkey any keys and then use CGEventPost to post
the key to the front application. This effectively lets me track all
the keys the user presses from a non-privileged application while
still sending input to the key window/process. I was also able to see
my admin pass
Can RegisterEventHotKey be used to log an admin password or other
passwords? I accidentally hotkey'd some regular characters and had
them trigger when typing in my admin pass on 10.5.8.
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Ya makes sense and pretty much what I saw; input was getting absorbed
by the hotkey app. I'm just thinking if a userland processed hotkeyed
all keys when the pass window popped. I'm guilty of quickly typing my
pass so I could get 3-4 characters into it before I'd notice input
isn't
Thanks for the confirmation. I did some testing using Rosetta and
didn't see endian problems but I wanted to make sure this was intended
and reliable on actual PPC and not just an anomaly leading me to
believe the byte swapping was automatic for explicit types.
On Jun 24, 2009, at 10:54
To reiterate, this is probably the best solution if you want to blur
all images that will be set on this image view. From IB you just
activate the Core Animation layer for the image view and attach the
gaussian blur filter to it. Then every image it displays will be
blurred
What Mr. Farmer was saying is this problem is a known bug with no
workaround or quick fix. You will have to wait for Apple to fix this
in a future system update. There is nothing you can do to your
application to make this work properly.
If you make a separate status item plugin, that
Generally you subclass the control and implement whatever data
association you want.
On Jun 11, 2009, at 4:41 AM, Jo Meder wrote:
I'm pretty sure I know the answer to this question, but I'll ask it
anyway :-)...
Is there a Cocoa equivalent to Carbon's Get|SetControlProperty()
family of
Subclass NSView and use poseAsClass to replace the default
implementation. You'll need to do the posing before running
NSApplicationMain. You'll have to use a different approach if you're
building for 64-bit.
On Jun 11, 2009, at 7:25 AM, Jo Meder wrote:
Hi Graham,
On 11/06/2009, at
Could be one of many network programs like LittleSnitch or
NetBarrier. Could also be he's using a proxy.
On Jun 2, 2009, at 9:37 PM, Mr. Gecko wrote:
Hello. For some reason, one of my customers is having a problem
which I determined to be NSDictionary dictionaryWithContentsOfURL
can't
Great glad you found it. Those are pretty hard to track down sometimes.
On Jun 2, 2009, at 2:36 AM, Kevin Ross wrote:
Turns out I did have an spurious NSNotificationCenter registration
in a loaded view that was causing the crash. Thank you for your
help Michael!
Are you registering for any notifications? It looks like it's
crashing while trying to notify an object.
On Jun 1, 2009, at 12:31 AM, Kevin Ross wrote:
Hi everyone, I have a Core Data document based app that I've been
working on for a while and I now want to change some of the view
You're creating an NSImage and pretending it's an NSImageView which
it's not. You'll need to create a new NSImageView and then the
NSImage and set the image as the imageview's image.
On May 30, 2009, at 8:27 AM, cocoa learner wrote:
Thanx Nick for your reply.But in my window I am not
You'd really be better off making an NSView subclass and having it
draw the image you want in drawRect:.
- (void)awakeFromNib
{
myImage = [[NSImage alloc] init
[self setNeedsDisplay:YES];
}
- (void)drawRect:(NSRect)rect
{
NSSize isize = [myImage size];
[myImage
NSString's stringWithFormat:, or using sprintf/asprintf with the %d
format:
char * str;
if(asprintf(str, %d, intval) 0)
//error
CGContextShowTextAtPoint...
free(str);
or if you know the possible max size of the text:
char strbuffer[MAX];
if(snprintf(strbuffer, MAX, %d, intval)
rawFileName = [[NSString alloc] initWithString:[[files
objectAtIndex:i] stringByStandardizingPath]];
Be sure to release rawFileName before you replace it with a new path.
On May 26, 2009, at 7:33 AM, vinai wrote:
think I am missing something really basic here with NSString. I am
trying
You need to retain destinationPath when you create it so it doesn't go
away. You properly retain dataDict, now do the same with
destinationPath and you should be ok. Also you don't need to retain
newRecord; the dictionary will retain it when you add it.
On May 11, 2009, at 3:10 AM, Ben
The UIImage is the object (inherits from NSObject), so yes you'd pass
the pointer to it as the dict's object. And objectForKey: will pass
back that pointer to the UIImage again.
On May 12, 2009, at 10:00 PM, Eric E. Dolecki wrote:
I like the cache without writing to the disk (for now
Does anyone know if NSArchiver's archive to file methods do atomic
file writes?
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at
I have a custom shaped window that refuses to use VRAM as its backing
buffer. It's a relatively small borderless window with a semi
transparent rounded rect shape. Are there certain things that will
prevent a window from using VRAM on supported hardware? Other stock
windows do this just
Guess I should clarify. Has anyone verified it; I've had an issue
with partial writes archiving built-in classes. The program used to
get signals to terminate and archive some basic classes (NSArray,
NSString, NSNumber) to disk. But these files would only contain some
of the archive
Yes no surprise..
But if NSArchiver is indeed atomic I would think the saved file would
either be saved complete with new data or complete old data and not
with partial data in situations like an untimely crash or resource
loss. Unless I have the wrong expectation of atomic file writes,
Alright thanks, what I was looking for.
On Mar 29, 2009, at 1:41 PM, Clark Cox wrote:
All bets are off if you do unsupported things inside of a signal
handler. For instance it is possible that the routines used to write
the file *seemed* to complete successfully when they, in fact, didn't.
In
I've found the reason for it is that the window is not opaque
(setOpaque:NO). So are all non-opaque windows ineligible for VRAM
backing? I don't see any documentation on window attribute
requirements for it, only GPU requirements.
On Mar 29, 2009, at 2:23 PM, Michael Vannorsdel wrote
You'll need to use an atomic operation to access tempSum in the main
thread as well and make sure tempSum is on a 4 byte boundary (required
by atomic ops).
But a cleaner method would be to keep a local running total in the
worker thread and after your time period send a message to the main
You could also just make your own wrapper class for NSDictionary with
the usual set/remove methods and sub the key in the wrapper.
On Mar 16, 2009, at 9:37 AM, Bill Bumgarner wrote:
Wile certainly a creative solution, this suggestion is an 11 on the
scale of 1 to 10 of bad ideas. Even if
Use the NSDecimalNumber class (subclass of NSNumber) and create it
like you would any other NSNumber. Then use decimalValue to get the
NSDecimal struct from it. There is no NSDecimalFromString, only the
other way NSDecimalString. The only other way is to do it is manually
by setting the
I could be remembering wrong but I think you can append unichars to
mutable strings with [mstr appendFormat:@%C, theunichar]. Or use %S
for a null terminated array of unichars.
On Mar 11, 2009, at 4:00 AM, Ken Thomases wrote:
If a string literal in your source code actually contains the
Use FSSetCatalogInfo and set the finderInfo (interpreted as a
FolderInfo struct) and add the kHasBundle bit to the finderFlags of
the FolderInfo.
On Mar 6, 2009, at 8:31 PM, Graham Cox wrote:
'm implementing a plug-in scheme based on Rainer Brockerhoff's
design at
This is the manual way to do it I might add. Not sure why the
document type settings are not being used. Sometimes you need to
rebuild launch services database.
On Mar 6, 2009, at 9:02 PM, Michael Vannorsdel wrote:
Use FSSetCatalogInfo and set the finderInfo (interpreted
You're trying to print an NSDate object returned by dateFromString:
with a %i format which is for integers. So the number you're seeing
is the memory address of the returned date converted to an int.
On Mar 4, 2009, at 4:15 AM, Jacob Rhoden wrote:
Anyone experience this weird behaviour
I have an NSTextFieldCell that draws attributed strings which have non-
default kerning and line multiple heights (to tighten line spacing).
This all draws as expected in the window but when these cells draw to
the print operation context the line multiple is ignored (using
default) while
float f;
[myData getBytes:f length:sizeof(float)];
On Mar 3, 2009, at 9:51 AM, Jay Kickliter wrote:
hat's what I'm doing. But don't understand how to get the 4 bytes I
have in NSData into a float variable.
___
Cocoa-dev mailing list
Make the main thread method something like this:
- (void)myThreadStart
{
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];
[NSTimer scheduledTimerWithTimeInterval:1.0 target:self
selector:@selector(myFucntion) userInfo:nil repeats:YES];
[[NSRunLoop currentRunLoop] run];
I'm having a problem I'm not sure how to solve. I have a custom view
which uses an NSAttributedString to draw text which works fine except
when it draws to a printer context. The NSAttributedDictionary
created inside the NSAttributedString's initWithString:attributes:
gets over released
Bingo, exactly what I was doing. Thanks a lot.
On Feb 28, 2009, at 11:05 AM, Shawn Erickson wrote:
I suggest you correct your (very likely is yours) over release
issue. :)
With that said I know of at least one quirk when working with
attributed strings that can cause you something like
You can check which tracking area triggered the entered/exited by
examining the NSEvent's trackingArea and make sure it's yours.
On Feb 26, 2009, at 7:06 PM, Xu ZiZhan wrote:
thanks Douglas,but that's not the point. rather I override the -
mouseMoved,
and called [super mouseMoved:], the
It should be more like:
//create timer
//add timer to runloop
CFRunLoopRun(); //code blocks here (not really since the loop is
running) until runloop exits due to no longer having observers or timers
//do thread cleanup and exit
When you start a new thread it has a
Really it would be best to malloc the space, use it, and free it.
Once you get to huge stack usage you gamble that you won't run out
when there can be other higher up calls also consuming some
(frameworks, libs, 3rd party code, ect). Also if you only use the
large amount once in a while
If your original algorithm is scalar, you can get upto 4x increase by
moving to SIMD (SSE2/3) if your algorithms can be paralleled. Add
that to the 7 extra cores and you can get upto 32x speed up. In the
real world you're very (very) unlikely to reach max throughput because
of data
I've looked into this quite some time ago and have seen
NSURLConnection keep FTP and HTTP connections open even after the
originating NSURLConnection had been deallocated. The same connection
was reused for subsequent NSURLConnections to the same destination. I
never did see these close
Checkout CGEventCreateKeyboardEvent and CGEventPost to synthesize
keyboard events. You have to make and post an event for the key down
and then the key up. So: command key down, V key down, V key up,
command key up. There's also CGPostKeyboardEvent which is easier to
use but not
Have a look at vImagePermuteChannels_ARGB in the Accelerate
framework. Should be the fastest and easiest option for reordering
color components in pixel streams.
On Jan 7, 2009, at 4:27 PM, Eric Gorr wrote:
Well, I wrote the code to change the pixel format from BGRA to ARGB.
Running
vImage_Buffer struct has a field for row bytes as well as image
dimensions.
On Jan 8, 2009, at 6:54 AM, Eric Gorr wrote:
Looks nearly perfect ... unfortunately, it does not look like it
will take the rowBytes of the GWorldPtr data into account. But, it
may still be faster to create a
Just start at inputTime 1.0 and increment down to 0.0. This will
rotate opposite from 0.0 to 1.0.
On Aug 25, 2008, at 11:42 PM, Seth Willits wrote:
CIRotatingCubeTransition is semi-public which confuses me. IB3
exposes it, but it otherwise seems to be private. I'd like to change
the
Ah I see what you're doing. Off the top of my head you might be able
to do this by rotating the two images 90 degrees and adding an affine
transform filter to rotate back 90 (so images are upright again). In
essence rotating the animation 90 degrees so it rotates on the
horizontal axis.
In my office we usually call objects returned directly without an
autorelease as short returned and an retain-autoreleased object as
pooled. Though these sound more slang than something that could be
official.
On Jul 18, 2008, at 3:10 PM, Andy Lee wrote:
Unfortunately, there's no such
I don't think it's the cause but you should probably use tanf to avoid
value casting to and from a double.
On Jul 13, 2008, at 7:30 AM, Patrick Walker wrote:
I can't post the whole thing because it's sort of large and
integrated but here is what I have originally done.
float radians;
You can do:
if([theName isEqualToString:@John Lenon])
//do stuff
NSString has some comparison methods in the class listing. The above
example will ask two different NSString objects if they have the same
string values.
On Jun 25, 2008, at 8:24 AM, Papa-Raboon wrote:
Hi All,
If you get your image data into a bitmap form you can use vImage's
(Accelerate Framework) permute functions to quickly rearrange the
channel ordering. There might be another way using color spaces, but
none that I know off hand.
On Jun 16, 2008, at 5:09 PM, Rodrigo Gutierrez wrote:
I'm
I thought I saw a CRC32 implementation in zlib at one time.
On Jun 14, 2008, at 9:25 AM, Jens Alfke wrote:
On 14 Jun '08, at 4:59 AM, Ilan Volow wrote:
No mention at all I can find (in the 20 seconds I scanned the first
two result pages) of any cocoa CRC implementations. If a newbie
were
In short yes. The long answer is the textfield will release the
string when it's done with it and if no one else retained the string
it will be deallocated.
On Jun 14, 2008, at 9:35 PM, Christopher J Kemsley wrote:
I'm new to Obj-C, and I'm trying to make sure I start off writing
good
NSFileHandle itself. Use seekToFileOffset: and readDataOfLength: to
get the bytes you want. Or you can drop down to the BSD layer and use
open/close, lseek, and read.
On Jun 14, 2008, at 8:05 PM, Angelo Chen wrote:
Hi,
NSFileManager's contentsAtPath can read the entire file, is there a
Ah ok. I was bit by something similar a few years ago. I had
forgotten to put the prototype in my class interface and believed the
compiler was using a prototype from an unrelated class which had
different arg and return types. But the twist was the missing
prototype caused the default
Something like:
messageString = [[NSString alloc] initWithData:(NSData*)data
encoding:NSASCIIStringEncoding];
the encoding type will depend on what encoding you expect the data to
be in.
On Jun 10, 2008, at 4:08 AM, Angelo Chen wrote:
What is the correct way to cast from CFDataRef to
[[NSWorkspace sharedWorkspace] launchApplication:@Safari];
On Jun 10, 2008, at 4:55 PM, Memo Akten wrote:
Hi all, i'm writing an app that launches some default apps like
safari, itunes, iphoto etc using NSTask. I was wondering if there is
a way of writing the launch url not fully
textShouldBeginEditing: and textDidBeginEditing: are generally methods
you'd override in a subclass. By default they call the
control:textShouldBeginEditing: and controlTextDidBeginEditing: of the
delegate (if there is one). The latter are the methods your delegate
needs to implement.
I suspect since the method had no prototype the compiler just assumed
the default id return type, but due to a bug didn't generate a
warning. The problem is most likely the calling method was expecting
the return value to be an integer (id; pointer) but instead is a
float. Even with a
Try using NSURL's fileURLWithPath:, you won't need to preappend
file://.
On Jun 5, 2008, at 8:37 PM, Mark Bateman wrote:
I'm currently hardcoding the file location just for debugging my
app. It works great.
I'm now trying to use NSBundle to insert the resource location
string for a
If the method is defined above the place you use it, you can avoid
compiler warnings. But the most common and more correct thing to do
is declare the method in the header with the rest of your class so
anyone that imports that header will know the specifics of that method
(and the
You can't really upgrade an already running process's privilege
level. What I'd suggest is make a small launcher program that the
user opens. This would ask for the admin password and then launch
your main application using AuthorizationExecuteWithPrivileges and
friends.
On Jun 3,
This will happen if the window is deallocated. It's probably getting
cleaned up by garbage collection.
On Jun 2, 2008, at 2:36 AM, Francis Perea wrote:
i Wayne, first of all thanks for your quick reply.
Sorry to say that the Console says nothing :-(
--
[Session started at
Retain and release have no effect on ObjC types when using garbage
collection. If your code is written relying on retain counting then
you should turn off garbage collection since you're trying to manage
the memory yourself (and probably designed the code as such). Garbage
collection has
I've never tried it personally, but you might make a CFMutableArray
with NULL callbacks and then cast it to an NSMutableArray since
they're bridged types.
On Jun 2, 2008, at 10:36 AM, Todd Ransom wrote:
Unfortunately I need to target Tiger also. Thanks for the info,
though, this will be
The retained option for windows in IB applies to the window's back
buffer settings. Retained means it only buffers sections offscreen
and onscreen drawing is done directly instead of to a back buffer
first. Not related to the window's retain count.
On Jun 2, 2008, at 5:24 AM, Francis
If you method is something like:
- (void)func1
then you should use @selector(func1) without the colon.
On May 31, 2008, at 2:06 AM, Markus Spoettl wrote:
NSInvocationOperation *theOp = [[NSInvocationOperation alloc]
initWithTarget:self
I think you want to schedule the connection for the
NSModalPanelRunLoopMode runloop mode. This is the mode that is used
for modal windows.
On May 31, 2008, at 7:28 AM, Torsten Curdt wrote:
A bit puzzled ...seems like I need some advise here.
From a NSWindowController I create a modal
I saw you using that in your included code so I thought you were ok
with it. For Tiger I can only suggest using secondary thread
processing as the cleanest approach. Or you can reconsider using a
modal window.
On May 31, 2008, at 9:07 AM, Torsten Curdt wrote:
You mean with
Basically NSString's floatValue and doubleValue methods only work when
the numbers use the US style dot separator (as opposed to other locale
separators such as a comma). If your NSString might contain non-US
separators (or other formatting differences) you'll have to use
NSScanner to do
This is from the CGColorSpaceCreateDeviceRGB docs:
In Mac OS X v10.4 and later, this color space is no longer device-
dependent and is replaced by the generic counterpart—
kCGColorSpaceGenericRGB—described in “Color Space Names”. If you use
this function in Mac OS X v10.4 and later, colors
You can't init an object more than once (well not an intended use).
You do:
SimParamnames = [NSMutableDictionary dictionary];
and later:
[SimParamnames initWithContentsOfFile: aFile];
which are both initializing the object. The second init is either
being ignored or perhaps corrupting
What I'd suggest is to assign each selection with a tag number in IB
(or programatically if need be). Then in your code you can get the tag:
int colorID = [[colorButton selectedItem] tag];
Assign each its corresponding ID as the tag like 4 for black, 1 for
red, ect.
On May 29, 2008, at
I think this is right as well. The image can be cached in a form best
suited for the first draw of it. Any further draws will very likely
use the cached version and the original resolution is gone. If you
set it not to cache, the image will be drawn from the original data
each time.
Generally you'd use the mouseDown to get when and where the drag
started (may have). Then use the mouseDragged to track where the
mouse is dragging to. Then on mouseUp you know it's done. Tracking
area is only valid if a tracking rect you installed generated the event.
On May 27, 2008,
This is an important consideration. If you're relying on a specific
event to finish within a finite amount of time, you're going to have a
race condition where the event could finish later than expected and
you may end up in an invalid state.
Perhaps if you told us what it is you're
On May 23, 2008, at 3:32 PM, Stephen Herron wrote:
The goal:
A window with two views. The view on the left, NSImageView, displays
the NSImage from - (void)openPanelDidEnd:. The right view, NSView,
displays the result of a CIFilter using the pixel data displayed in
the right-hand view.
You troubles are all in this method and how you handle cityArray.
First you don't need to alloc and init a new array for cityArray as
[NSArray arrayWithObjects: c0, ...c9, nil] gives you a new one already
with the objects you specified.
Another thing you should watch for is there's an
Would be like a cookbook including instructions on how to operate an
oven and use a knife to cut vegetables in every single recipe. I
think most would find that pretty redundant and diluting, especially
when the preface covers these topics in detail.
On May 21, 2008, at 8:37 PM, Jens
When you mention running count, are you keeping an int or something to
keep track of thread progress? If so remember to use atomic
operations to update the count from multiple threads to prevent data
corruption. As for NSAlert, a general rule is not to interact with
the UI from any
Even delegation is not a commonly known term in my experience (it's
used for several differing ideas in the US). I try my best to give
terms and examples with the best chance of grasping, especially with
beginner concepts. Even the truest term of delegate doesn't perfectly
fit Cocoa
There's nothing that guarantees a Cocoa delegate will act for another
object and that the represented object won't act how it wants as
well. Sometimes a delegate method is just a notification something
happened/happening without the delegate having any say on the matter
or affect on the
Ditto, I'll surrender the last word to you. Though I'm interested to
know if the OP is having any success or not. This thread got a little
sidetracked.
On May 19, 2008, at 2:23 PM, I. Savant wrote:
I think that's the last I'll comment on the dictionary definition
matter; it's a silly
How are you creating the stream?
On May 19, 2008, at 7:07 PM, John MacMullin wrote:
Continuing on my efforts re: my modified Echo Server, the following
code is crashing.
- (void)startStreamWrite:(NSOutputStream *)ostream
{
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc]init];
Well what can you do. Not sure why but lately many newcomers have
been showing up and complaining about Cocoa's difficulty. I'm not
sure if they've done GUI work before, but I remember my days with
PowerPlant and spending a massive amount of time just creating the GUI
and the code to
I'm also wondering if many of the people finding Cocoa difficult are
also lacking OO programming experience. The docs teach Cocoa really
well but if you're unfamiliar with OO design and concepts the Cocoa
docs are going to be very daunting.
On May 18, 2008, at 3:28 PM, Jason Stephenson
On May 18, 2008, at 7:39 PM, Julius Guzy wrote:
So I wouldn't have much to say about it except that it does have a
tendency to make things seem more exciting than they actually are.
For instance I can refer here to the idea of dynamic typing which
still requires us to have the header files
Checkout NSXML* and friends (NSXMLNode, NSXMLElement). Also there's
Introduction to Tree-Based XML Programming Guide for Cocoa and
Introduction to Event-Driven XML Programming Guide for Cocoa guides
in the docs.
On May 18, 2008, at 8:07 PM, David wrote:
Is there a Cocoa package that
Nothing wrong with saying you should read such and such. But RTFM
is the condescending way of saying it (just look what it stands for).
Would be like asking someone where the restroom is and getting look
at the building directory, you blind clueless moron. My point was
about posts that
The difficulty is methods in ObjC are dispatched messages rather than
hardcoded functions so going from call to method execution has some
hidden intermediate steps. And there can be more than one method with
the same name from different classes/protocols. This is one of the
pillars to
I'm thinking the window might be redrawing itself right after your own
drawing and erasing it. You could try disabling the window's auto
displaying and flush the window buffer after your draw to determine if
this is the case.
I don't know where you're doing the drawing to know if it's
Well you didn't present the secret handshake that would open immediate
access to such information.
Actually, it was probably not brought up earlier as it's kind of an
unsaid list etiquette rule exercised by some not to critique someone's
choice of APIs unless specifically asked for. Most
Set its cell as not editable in IB or programatically with [[indicator
cell] setEditable:NO].
On May 17, 2008, at 4:41 PM, Philip Bridson wrote:
I have an NSLevelIndicator in a utility window that does the
standard stuff, but I have just noticed that if I click on the
indicator I can
1 - 100 of 192 matches
Mail list logo
