go back in source control and dredge up my old code
and try to repurpose it.
On Mon, Jun 16, 2008 at 6:29 PM, Nick Zitzmann <[EMAIL PROTECTED]> wrote:
>
> On Jun 16, 2008, at 7:16 PM, John Stiles wrote:
>
> I've got some code which subclasses NSTextView in order to provid
I've got some code which subclasses NSTextView in order to provide a custom
field editor for NSTextFields. It doesn't do anything too extraordinary,
just allows for filtering out some characters, watching for certain events
(like deleting characters or changing the selection), etc. It all works
qui
Note that these delegate methods will not work properly until your
Info.plist has been updated to reflect the file types which your app
supports.
I've recently opened a radar on this because it doesn't appear to be
well-documented and I spent a while trying to figure out why my
delegates were n
None of this really refutes what Ricky posted.
You are just lucky that it works in the one-display case. It really
isn't designed to work, and on some configurations, it just won't.
Is there anything preventing you from following Ricky's advice?
Dennis Munsie wrote:
In this case, what I am tr
There are a couple of gotchas... trying to disable the right-click menu
is one issue (maybe you can subclass the view for this, I haven't
tried), and having the Flash app send messages back to the main app is
another problem.
Julia Rixon wrote:
Ferhat Ayaz wrote:
Why don't load flash via We
We have an internal app which sends emails in a similar fashion (all for
automated stuff, not for users to see/touch) and we shell out to a Perl
script which uses MIME::Lite to handle this.
It works perfectly for us. I'm not sure how applicable it will be for
you, but it's handled everything w
Though I guess he could make a LSUIElement application to handle this.
Kinda clunky, but it would work.
Michael Ash wrote:
On Fri, May 9, 2008 at 8:17 AM, Daniel Parnell <[EMAIL PROTECTED]> wrote:
Thanks for that. I was wondering if that might be the case, although I've
not been able to fi
I think they are focusing on polishing up 2.0. I submitted an
enhancement for FSA a while back and that seemed to be where all the
focus was.
Dave Dribin wrote:
On May 8, 2008, at 7:53 PM, David Carlisle wrote:
My file directory name says "FScriptBin-20070421" My startup warning
message say
t with the same number of bytes, it's
somewhere in the 0.1 second range.
So very confused...
Ben
On May 7, 2008, at 8:21 PM, John Stiles wrote:
Have you tried running Shark? That might give you some insight as to
what's going on.
Ben Einstein wrote:
Hi All,
I have an enterprise DB
Have you tried running Shark? That might give you some insight as to
what's going on.
Ben Einstein wrote:
Hi All,
I have an enterprise DB application that once used DO to move some
files around (images and zip files, mostly). After some serious
testing and lots of reading, I decided to move
This was actually in a WWDC WebKit demo when WebKit was first
announced—they set up a web browser on stage with no code, IIRC.
Derek Chesterfield wrote:
I don't think it's a bug. I think Apple just decided they didn't want
to expose WebView with all the useful bindings.
I used to love that I
I'm going to guess that your preferences window is set to release-on-close.
This can be set in Interface Builder; check the Inspector for this
window and make sure that checkbox is not set.
Patrick Altman wrote:
I am very new to cocoa development but a rather seasoned software
engineer in gen
Recursive menu tracking? I can't say I'm surprised that this doesn't work :)
Honestly I have to say that this sounds like a really weird UI. I would
consider a more normal approach (submenus?)
Angel Todorov wrote:
Hi,
I have a regular menu (File menu) for my app. One of the menu items (New
..
An NSView doesn't have an associated cell, so "size to fit" doesn't
really have a meaningful answer. Size-to-fit works by asking the cell
what size it should be.
For a box, you could just enumerate the subviews and get the union of
all the subview frames.
Randall Meadows wrote:
On May 1, 2
There is in fact an API, -sizeToFit.
Randall Meadows wrote:
I am creating a bunch of controls (at least NSTextfield,
NSPopupButton, NSSlider, and perhaps others) programmatically (that
will eventually be shown in an NSTableView), and would like to apply
the "Size To Fit" feature that IB provi
Yup, that was the problem. Thanks for figuring this out, Graham.
rdar://5903558[IB3] Items in pull-down-style NSPopUpButtons enable
"hidden" for no reason
John Stiles wrote:
Oh, is that the problem? I didn't see a Hidden checkbox. I'll look again.
Guess it's Rada
If you don't notice this then the items obviously don't show up at
runtime. You can select all and set them to visible, then go back and
set the first item back to hidden as a workaround.
G.
On 1 May 2008, at 10:37 am, Peter Ammon wrote:
On Apr 30, 2008, at 3:58 PM, John Stiles wrote
I've been trying to create an NSPopUpButton in IB3, in pull-down mode,
that has a submenu. It doesn't seem to work properly though. After I
drag in the "Menu >" item into the pop-up button's menu, the pop-up
seems to be irrevocably broken-it randomly neglects to display some
items, or fails to show
Manfred Schwind wrote:
You probably shouldn't be putting up a modal dialog from within the
menu-tracking runloop mode. You can use
-performSelector:withObject:afterDelay:inModes: to defer the call
that opens the modal dialog; use a delay of 0.0 but a modes array
that includes only the default
You could probably use NSMenu's -cancelTracking to work around this. (I
guess you could send it to every menu in the menubar, or see if just
sending it to mainMenu is good enough.)
Or, you can use -menuWillOpen: and -menuDidClose: to watch for when the
user accesses the menus, and postpone you
Man, what a bummer. I wish you had been around when I asked the question
before :)
Keith Blount wrote:
Really? Does this actually work?
I needed to find word boundaries to implement a Find panel which
supports searching for "Whole Words" and I ended up using
UCFindTextBreak based on advice fr
Really? Does this actually work?
I needed to find word boundaries to implement a Find panel which
supports searching for "Whole Words" and I ended up using
UCFindTextBreak based on advice from this list. It was a pain to
implement (since it's not designed to mesh with Cocoa at all).
Graham C
Kyle Sluder wrote:
On Tue, Apr 29, 2008 at 3:59 AM, Yann Disser <[EMAIL PROTECTED]> wrote:
Giving a minimum size to my drawer worked.
However I think this absolutely is a bug. I would expect my views to vanish
if the window is made too small and to reappear normally once the window is
again
You could also show any other thread stuck trying to lock a mutex? I
assume there's another thread holding this lock or stuck trying to lock
something else.
Colin Cornaby wrote:
Hi,
I'm currently trying to track down a deadlock. After adding a series
of layers, CoreAnimation deadlocks here
Is the $ usage an extension? That doesn't sound like regular C to me.
Alastair Houghton wrote:
On 24 Apr 2008, at 16:10, Dave Jewell wrote:
On 24 Apr 2008, at 8:42 am, Graham Cox wrote:
Aside: your ivars shouldn't start with an underscore - Apple reserves
such names for its own classes.
In
Nick Zitzmann wrote:
On Apr 23, 2008, at 4:04 PM, John Stiles wrote:
I have a class declared in code which, until recently, didn't have
any IBActions in it.
Recently I added some, and went to IB3, but it didn't notice that I
had added the actions. I had to manually add them via th
I have a class declared in code which, until recently, didn't have any
IBActions in it.
Recently I added some, and went to IB3, but it didn't notice that I had
added the actions. I had to manually add them via the + button.
Is there any way I can give IB3 a hint so it can find the class'
@interf
Thirded.
Matt Gough wrote:
I'd second that. The OS (well, Finder) also adds things to the
resource fork of files (custom icons, info about which app to open a
file with when you changed it from the default etc). Just as long as
you respect the existing contents this is exactly where you should
Mattias Arrelid wrote:
On Mon, Apr 21, 2008 at 7:05 PM, John Stiles <[EMAIL PROTECTED]> wrote:
In "The Path Of Key Events" in the URL you posted, the #1 item in the list
is "key equivalents."
AppKit checks keyDown events to see if command is held; if it is, it
This "solution" will also kill all the user's Dashboard widgets, so it's
not to be taken lightly.
Mike Abdullah wrote:
Quite simply there is no public API to control the dock. The closest
(and hacky) solution is to modify the dock's plist and restart it.
Mike.
On 22 Apr 2008, at 07:02, Oliv
NM that last part, I found the plugin. For the archives, it's at
http://brockerhoff.net/src/RBSplitView.ibplugin.zip
Sorry, all, for the multiple replies-to-self here :)
John Stiles wrote:
Oh, there was a fourth problem actually. This doesn't work in all cases:
NSRect cha
lugin for RBSplitView? I've seen web posts indicating
that it exists but I can't find it.
John Stiles wrote:
I found thee things which, in conjunction, solve the problem:
1 - Rewrite synchronizedViewContentBoundsDidChange as follows.
-scrollToPoint seems broken. (I am clamping the scr
cocoa view that is shrunk to a degenerate rectangle
loses its position & size)
John Stiles wrote:
Has anyone gotten this example to work in Leopard?
http://developer.apple.com/documentation/Cocoa/Conceptual/NSScrollViewGuide/Articles/SynchroScroll.html
I just tried it and I
Has anyone gotten this example to work in Leopard?
http://developer.apple.com/documentation/Cocoa/Conceptual/NSScrollViewGuide/Articles/SynchroScroll.html
I just tried it and I'm having terrible luck. The view is just blanking
itself out immediately. I can get it to sort-of work if I replac
Corbin Dunn wrote:
On Apr 18, 2008, at 3:37 PM, John Stiles wrote:
Ben Lachman wrote:
> Well, you should be able to just override the drawing code, since
> thats really your problem. Going directly against the docs, while
it > may work fine now, is playing with fire in my opin
In "The Path Of Key Events" in the URL you posted, the #1 item in the
list is "key equivalents."
AppKit checks keyDown events to see if command is held; if it is, it
tries to match it against the menus before passing it through the
responder chain.
I think your best bet is to dim your menu item
Could you do something else to cause VoiceOver to explicitly say the
string you want it to?
Martin Wierschin wrote:
The "fake temporary item" solution actually works pretty well. It's
the last thing I'd call elegant, but here's how you can blink a menu
title in Cocoa.
Unfortunately that solu
No, that's not the problem. The problem is that some combinations simply
don't work. For instance, just as a random example, AppKit does not
match "option+," or "shift+`". (Haven't tried adding command but offhand
I don't have any reason to think that this would fix it.)
My app supports arbitr
Ben Lachman wrote:
> Well, you should be able to just override the drawing code, since
> thats really your problem. Going directly against the docs, while it
> may work fine now, is playing with fire in my opinion.
Yeah… that's why I posted :) I was hoping to get a "oh yeah, that only
applies
select the full row/column by clicking on the header.
Fortunately I don't need to worry about that in this case. Not a problem.
But this should get you started, hopefully.
->Ben
--
Ben Lachman
Acacia Tree Software
http://acaciatreesoftware.com
[EMAIL PROTECTED]
740.590.0009
On A
erformKeyEquivalent:f35Event];
[myMenu removeItemAtIndex:position];
}
John Stiles wrote:
Reading the list archives a little more, it looks like there may be
two ways to do this:
- _NSHighlightCarbonMenu and _NSUnhighlightCarbonMenu are SPIs which
take an NSMenu* and do exactly what you'd expect
- Yo
e key equivalent and no target or action. Then use NSMenu
-performKeyEquivalent: to simulate its selection.
Wow, great choices here :| I'm going to try #2 first since it's not SPI.
I'll inform the list of the results.
John Stiles wrote:
John Stiles wrote:
Randall Meadows wrote:
The docs for -editColumn:row:withEvent:select: ominously claim:
The row at rowIndex must be selected prior to calling
editColumn:row:withEvent:select:, or an exception will be raised.
I'm implementing a subclass of NSTableView which behaves a little more
like an Excel spreadsheet—it mainta
7;t be more than one character if dead key state
was zero.
// Accented characters take a single byte in legacy encodings.
UInt8 singleChar = chars & 0xFF;
result = [[[NSString alloc] initWithBytes:&singleChar
length:1
t 11:38 AM, John Stiles wrote:
Quick question: in Leopard, are there any keyboards left which don't
have a uchr?
I found some sample code which includes a fallback case for if no
'uchr' resource is found (it uses plain KeyTranslate in this case)
and I'm wondering whether
John Stiles wrote:
Randall Meadows wrote:
On Apr 17, 2008, at 11:54 AM, John Stiles wrote:
As previously explained here, I'm handling hotkeys in my app via
custom code in order to work around some AppKit bugs.
How can I simulate the menu-title blink effect using Cocoa? In
Carbon,
Randall Meadows wrote:
On Apr 17, 2008, at 11:54 AM, John Stiles wrote:
As previously explained here, I'm handling hotkeys in my app via
custom code in order to work around some AppKit bugs.
How can I simulate the menu-title blink effect using Cocoa? In
Carbon, it's FlashMenuBar(m
As previously explained here, I'm handling hotkeys in my app via custom
code in order to work around some AppKit bugs.
How can I simulate the menu-title blink effect using Cocoa? In Carbon,
it's FlashMenuBar(menuID) but I don't see a Cocoa equivalent.
_
t telling what shift-[ could be, and
it's even possible that [ is the shifted version of another character.
I'd go for the actual character and its meaning, not the key.
Best,
Hank
On Apr 17, 2008, at 12:02 PM, John Stiles wrote:
I have an NSEvent and I need to know what key the use
coming up with low-ASCII
values, i.e. numbers under 32.
I guess I could implement a hybrid approach where I only use
UCKeyTranslate if the character appears to be letters or punctuation...?
Seems doable but makes me wonder if I'm going down the wrong path. Is
there a better way?
John St
I'm currently cribbing from here:
http://lists.apple.com/archives/carbon-dev/2005/May/msg01062.html
And I got rid of the non-uchr section. I can require Leopard in this case.
From: Jean-Daniel Dupas [mailto:[EMAIL PROTECTED]
Sent: Thursday, April 17, 2008 9:37 AM
To: John Stil
he Leopard-and-above timeframe.
John Stiles wrote:
Sweet, I will take a look at this and post back when I have results or
questions. Thanks!
Greg Titus wrote:
I think you'd ask the NSEvent for its -keyCode, then pass that key
code to UCKeyTranslate() with all the modifier key state (includ
tring for what that key
would mean if the user hadn't been pressing any modifiers.
Hope that helps,
- Greg
On Apr 17, 2008, at 9:02 AM, John Stiles wrote:
I have an NSEvent and I need to know what key the user has pressed,
minus any of the modifiers. NSEvent -charactersIgnoringModifi
I have an NSEvent and I need to know what key the user has pressed,
minus any of the modifiers. NSEvent -charactersIgnoringModifiers seems
like a good place to start, but it has one serious flaw—it does not
ignore the Shift key. So, for instance, it won't change ~ to `, ! to 1
or { to [.
I ne
less. Core Foundation is just ugly in comparison... :)
Michael Ash wrote:
On Wed, Apr 16, 2008 at 1:20 PM, John Stiles <[EMAIL PROTECTED]> wrote:
The difference is, in STL, if you want an array where you can remove from
the beginning, you would never use vector<> anyway. You'd use d
Jean-Daniel Dupas wrote:
Le 16 avr. 08 à 00:07, John Stiles a écrit :
Hmm, set sounds a lot easier than this to me.
Just use insert to put all the doubles into the set (one line), then
use lower_bound to find the delineations between each group (another
one-liner, though of course you&#x
I never thought of using a hash function to do binning. Interesting
approach.
Army Research Lab wrote:
Have you looked at hash_multimap
(http://www.sgi.com/tech/stl/hash_multimap.html)? Note that the following
code was beaten out in entourage, without compiling, testing, etc.
struct eqdouble
AFAIK, you'd need to use the lower-level NSURL* APIs (e.g.
NSURLConnection) instead of stringWithContentsOfURL, which is going to
be a little more work. Unless there is some trick I don't know about...
Also, you don't really want ASCII encoding—probably UTF8 is more what
you had in mind. ASCII
Hmm, set sounds a lot easier than this to me.
Just use insert to put all the doubles into the set (one line), then use
lower_bound to find the delineations between each group (another
one-liner, though of course you'll need to loop over the number of
groups you want). Then the distance between
IS makes a good point. Moreover, if you have some method which has a
side effect or which might not return the same result every time, "po"
is no good for debugging it. These simplistic examples (creating a URL
from a string) will work the same no matter how many times you execute
them, but a l
er, and
(im 99% sure) it deals with things like if treeController returned nil
instead of a NSArray.
NSManagedObject *selectedTreeObject = [self
valueForKeyPath:@"delegate.mainWindowController.treeController.selectedObjects.lastObject"];
On Apr 14, 2008, at 10:05 PM, John Stiles wrote
The chained approach is tempting since it's short and convenient, so if
the code is not prone to failure, I'd say go for it.
If you expect that you might need to see intermediate values in the
debugger or there are weird edge cases where something might return nil,
I'd break it out into multip
Laimonas Simutis wrote:
On Fri, Apr 11, 2008 at 6:01 PM, Scott Anguish <[EMAIL PROTECTED]> wrote:
On Apr 10, 2008, at 9:04 PM, Laimonas Simutis wrote:
Hey,
This is my first cocoa projects so I am kind of finding my way around
the framework. The question I have is maybe more related t
I think Instruments could do a better job of telling you what's going
wrong than we could.
FWIW, the table view doesn't even know the contents of most of your 1500
rows. It asks for them from the data source as it needs them, and
probably only knows the values of the currently visible cells.
Even if you could change it, once your app is running it, changing the
plist won't have any effect until the next time the app is run. So it's
probably a moot point.
glenn andreas wrote:
On Apr 9, 2008, at 4:41 PM, Randall Meadows wrote:
[resending with my subscribed address--grrr]
On Apr
You cannot tell your app to hide from the dock once it has shown.
However, you can start hidden and become visible via the (IIRC)
TransformProcessType API.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On
Behalf Of Jere Gmail
Sent: Wednesday, April 09, 2008 2:11 PM
This is very cool and I would have used it if I had known about it
earlier, but it isn't a replacement for NSToolbar so much as a simpler
way to create one (which is awesome and should have been there all along).
Geoff Beier wrote:
On Tue, Apr 8, 2008 at 12:25 PM, John Stiles &l
Well, I was thinking more along the lines of an RBSplitView, something
you can just drop into your nib/code and suddenly your toolbars are
better :)
Thanks for the pointers, everyone.
[EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] wrote on
2008-04-07 17:47:30:
I'm looking for an open-source N
For any curious Apple engineers, I've just filed this issue as
rdar://5848023 and attached a test app (four lines of code).
John Stiles wrote:
I am trying to add some hotkeys to buttons in my app, and I've hit a
weird snag. Specifically, the shift modifier flag appears to be
i
I am trying to add some hotkeys to buttons in my app, and I've hit a
weird snag. Specifically, the shift modifier flag appears to be ignored
for anything other than alphanumeric keys—i.e. I can't make a button
that corresponds to "cmd+F1" and a second button that corresponds to
"cmd+shift+F1".
I'm looking for an open-source NSToolbar-like toolbar implementation...
do any exist? I haven't found any yet.
I want to check it out and see how some things are done... I don't
really need a ton of features outside the standard toolbar stuff, it's
more for learning/research to see what can be
Cocoa doesn't have ordered sets. It has arrays (NSArray), unordered sets
(NSSet), and unordered key-value tables (NSDictionary). Any of these can
optionally be mutable. NSCountedSet also allows for the equivalent of a
"multiset," but for some reason there is no NSCountedDictionary, for the
equi
You can't be KVC compliant if you use prefixes like m_, so this limits
your potential for using things like bindings.
When I write C++, I am also a fan of m_, but it's just not appropriate
for Cocoa since it basically means you're fighting against the frameworks.
Scott Andrew wrote:
See i stil
I'd suspect some sort of error in your app, e.g. perhaps "pool" itself
has never been alloc/inited and contains garbage. [pool release] on a
valid but empty NSAutoreleasePool is perfectly harmless.
Kevin Dixon wrote:
I have a method in my class that's a thread entry point, using
detachNewThrea
I'm not a hacker, but if I had to figure out how a Cocoa app worked in a
hurry, I'd check out F-Script Anywhere.
justin webster wrote:
just wondering how easy it is for would-be hackers to get inside my code.
how meaningful and human-readable is a reverse engineered version of
my app?
is the
I'd never heard the Smalltalk conventions before, but I have to admit I
really like the sound of them. I'd love to see a block of code written
to these rules to see how it plays out in practice. (ObjC or C++, that
is, not Smalltalk.)
Robert Claeson wrote:
On 3 Apr 2008, at 19:58, Rob Napier
Your delegate and data source can be the same object if you want.
Nothing prevents it.
Mr. Gecko wrote:
What do you mean by the "name"?
in the identity tab of Interface builder
You could make one class that has all of the common code, and then
subclass it for each table instance. That's a pr
What do you mean by the "name"?
You could make one class that has all of the common code, and then
subclass it for each table instance. That's a pretty common technique.
Mr. Gecko wrote:
How would I get the name of a table view? I am trying to make all of
my tables use one class for the data
Alastair Houghton wrote:
On 2 Apr 2008, at 17:19, John Stiles wrote:
I take it all back; in 2007 there was an MD5 attack discovered which
actually allows for completely different binaries that sign the same.
Check Wikipedia for the details, but basically MD5 is totally broken
now. Wow, times
John Stiles wrote:
Jens Alfke wrote:
Also, you're aware that MD5 shouldn't be used for anything
security-related anymore? Last I heard it's pretty close to being
fully broken. SHA-1 is a lot more secure, and has a larger output
which itself makes collisions less likely.
&q
Jens Alfke wrote:
Also, you're aware that MD5 shouldn't be used for anything
security-related anymore? Last I heard it's pretty close to being
fully broken. SHA-1 is a lot more secure, and has a larger output
which itself makes collisions less likely.
"Fully broken"? I don't know about tha
I haven't experimented with CC_MD5, but we do have code which calculates
MD5s (calculated via simple C code).
Is CC_MD5 optimized e.g. using SSE or AltiVec? Should I expect to see a
perf boost if I swapped in this code instead of our regular C code?
Jean-Daniel Dupas wrote:
Le 2 avr. 08 à 01
I have a vague recollection that the mechanism used by Activity Monitor
and friends is not public API, unfortunately.
Martin Redington wrote:
I went to check the referenced thread, as this is something I wanted
to do occasionally, when I noticed that I was the OP for it.
The discussion was
If it's rounding, that would be pretty severe. I'd expect at least 6 or
7 viable digits.
Aki Inoue wrote:
From the look of it, I think there is floating-point rounding involved.
After all, the graphics system is treating it as a scalable font so
the notion of bitmap (or pixel-based) font des
If you want to use OpenGL for your GUI, that's always an option, but
again you're on your own here. Cocoa provides very little support for
this (outside of the generic NSOpenGLView).
Honestly I think most people don't appreciate how much stuff you get
"for free" in Cocoa, but you learn pretty
Basically you get to rewrite all the controls from scratch :)
It's not a minor undertaking, so be sure it's worth the effort. Your
best bet is to look for open-source code which does what you want, since
there are already a ton of open-source Cocoa classes for custom UIs.
Also, be sure your c
epts the key. Apparently I believed that the
message to super didn't work precisely because the next responder in my main
application was accidentally swallowing all keyDowns. I apologize for the
misinformation, but thank you for helping me find a bug.
Allen
-Original Message-
From: Jo
n I
thought. Walking the menus and submenus recursively to build the table
is pretty simple, and I've already got a good hash table class which I
was able to put to good use.
Maybe I'll file a bug on this...
Allen Smith wrote:
On Mar 27, 2008, at 8:12 PM, John Stiles wrote:
hash table before handling -keyDown:,
maybe I'll just keep doing that. It's gross but I guess all the
potential solutions are gross.
Ken Thomases wrote:
On Mar 27, 2008, at 7:52 PM, John Stiles wrote:
I am implementing a custom NSView subclass (actually a simple
subclass o
d
command keys.
The view can be connected to other blobs of code, and I am responsible
for passing along the message, "hey, the user pressed space!" But it
will gleefully consume the event no matter what I tell it was pressed.
Hamish Allan wrote:
On Fri, Mar 28, 2008 at 12:5
I am implementing a custom NSView subclass (actually a simple subclass
of NSOpenGLView) that implements -keyDown: in order to respond to user
typing. Typically, this works great.
However, I have a few menu items which respond to atypical hotkeys (e.g.
one responds to "space", another to "optio
I've found that adjusting the font matrix to include a negative Y
transformation has the effect of decreasing the line height. This seemed
like a bug to me when I discovered it, since I just wanted the text to
be translated upwards in its cell, not shrink the line height. I wasn't
sure if it wa
FWIW, if you find yourself needing to do much more in the way of
distorting images, I'd recommend looking into OpenGL directly.
It's extremely fast since it goes straight to the hardware, and doing
something like "map an image to a trapezoid" is /very/ few lines of
code. Much simpler than trying
AppKit does an optimization where at app shutdown time, it stops
dealloc'ing things.
It does make app quitting faster, but it means that if your deallocs
have side effects, the side effects will never occur. It also makes leak
tracking difficult if you try to do a full-on leak check at applica
cal performance while the underlying guts
were actually doing a more sophisticated task (copy instead of
reference, sorted instead of unsorted). Other than that, you've been
giving credit pretty much where it's due.
Michael Ash wrote:
On Fri, Mar 21, 2008 at 4:51 PM, John Stiles <[EMAIL
Michael Ash wrote:
On Fri, Mar 21, 2008 at 12:13 PM, John Stiles <[EMAIL PROTECTED]> wrote:
Michael Ash wrote:
For what it's worth, I wrote a quick test program creating 50,000
random key/value pairs of NSStrings of around 500 characters each,
then inserted them into an NSDic
Michael Ash wrote:
On Thu, Mar 20, 2008 at 7:58 PM, E. Wing <[EMAIL PROTECTED]> wrote:
You really should profile to find your bottlenecks, especially when
the STL is concerned. My personal experience has been that gcc poorly
optimizes STL code automatically for you and you must go in and
p
You could always do this
float value;
if (1 == sscanf([myString UTF8String], "%f", &value))
{
// the string was a valid float
}
else
{
// it wasn't
}
For integer, replace "float" with "int" and "%f" with "%d".
Localization concerns still apply in theory but in practice there are
few loca
7;m worried that I'll still end up having to import the header file
for my C++ class in my Objective C source file though, which caused a
lot of compile errors previously.
On Mar 20, 2008, at 12:32 PM, John Stiles wrote:
> Without starting a religious war, I have to disag
Without starting a religious war, I have to disagree with this.
ObjC++ is probably a bad idea if you are a novice programmer in general,
but I think it also has some really good things going for it, and having
written huge amounts of ObjC++ code, I think it's perfectly
straightforward to use.
1 - 100 of 169 matches
Mail list logo
