On Jun 21, 2010, at 1:33 AM, Steve Steinitz wrote:
> Hi Kyle,
>
> On 20/6/10, Kyle Sluder wrote:
>
>> Be aware that as of 10.6, this is an officially unsupported
>> configuration, prone to breaking in point releases as happened in
>> 10.6.2.
>
> Yes, we are more careful with Mac OS X updates n
Hi Kyle,
On 20/6/10, Kyle Sluder wrote:
Be aware that as of 10.6, this is an officially unsupported
configuration, prone to breaking in point releases as happened in
10.6.2.
Yes, we are more careful with Mac OS X updates now.
See Ben Trumbull's post here for the nitty-gritty:
http://lists.a
Hi John,
not to question your own algorithms, but you may also want to
consider checking the MAC address (ethernet)
I see what you are saying but I want to keep specific hardware
out of the equation. I want to be able to swap a new Mac Mini
in at any time. The computer name is a convenienc
On Sun, Jun 20, 2010 at 6:25 PM, Steve Steinitz
wrote:
> Each machine runs idle-time code which ensures it has recent data from the
> shared store. It was hard to get right but after a year of tweaking, it
> works well. SQLite has just enough locking capability to make it work.
Be aware that as
On Jun 20, 2010, at 6:26 PM, Steve Steinitz wrote:
> Thanks Jonathan Mitchell, John Joyce, Paul Sanders and Jens Alfke.
>
>
> John:
>
> Bingo! CSCopyMachineName() works perfectly.
>
> To answer your question, I want a human-readable machine identifier, but one
> that's not tied to the hardwa
Hi Kyle,
On 20/6/10, Kyle Sluder wrote:
to temporarily hide sales that are in progress on other machines.
I currently do that but in a more awkward way,
Might you instead want to make a sale an atomic thing? Perform the
sale on a scratch MOC, and then when the sale is complete (or voided),
me
On Sun, Jun 20, 2010 at 4:26 PM, Steve Steinitz
wrote:
> 1. to temporarily hide sales that are in progress on other machines.
> I currently do that but in a more awkward way,
Might you instead want to make a sale an atomic thing? Perform the
sale on a scratch MOC, and then when the sale
Thanks Jonathan Mitchell, John Joyce, Paul Sanders and Jens Alfke.
John:
Bingo! CSCopyMachineName() works perfectly.
To answer your question, I want a human-readable machine
identifier, but one that's not tied to the hardware, the way,
say, serial number is.
Our Core Data point-of-sale sys
On Jun 20, 2010, at 11:01 AM, John Joyce wrote:
> This is different from a localhost name. The host name can be very different.
And each active network interface will have a different hostname, just as it
has a different IP address. So I might simultaneously be “jens.foocorp.com” on
Ethernet w
On Jun 20, 2010, at 8:46 AM, Steve Steinitz wrote:
> Hi Paul,
>
> On 20/6/10, Paul Sanders wrote:
>
Are you #including ?
>>
>>> I wasn't. I added it. Were you wondering if it would make a difference?
>>
>> Well, yes. Default return type is int. In fact I'm suprised you
>> didn't get
> The correct import is #import
> .
So it is, sorry. It was half-way down my source file.
Regards,
Paul Sanders.
___
Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)
Please do not post admin requests or moderator comments to the list.
Contact the
On 20 Jun 2010, at 14:46, Steve Steinitz wrote:
> Hi Paul,
>
> On 20/6/10, Paul Sanders wrote:
>
Are you #including ?
>>
>>> I wasn't. I added it. Were you wondering if it would make a difference?
The correct import is #import
.
Omitting it causes the compiler to assume integer return
Hi Jonathan,
Thanks for replying.
On 20/6/10, Jonathan Mitchell wrote:
>> Could I have done something to my project to break toll-free bridging?
> You can't break toll free bridging.
> The NSObject and CF type collaborate to route objects and function calls
> appropriately.
Thanks for that
Hi Paul,
On 20/6/10, Paul Sanders wrote:
Are you #including ?
I wasn't. I added it. Were you wondering if it would make a difference?
Well, yes. Default return type is int. In fact I'm suprised you
didn't get a warning about the function being undefined. My Mac is
powered off or I wou
>> Are you #including ?
> I wasn't. I added it. Were you wondering if it would make a difference?
Well, yes. Default return type is int. In fact I'm suprised you didn't get a
warning about the function being undefined. My Mac is powered off or I would
check. Does adding the #include sort
Hi Paul,
On 20/6/10, Paul Sanders wrote:
It looks like SCDynamicStoreCopyComputerName is not prototyped
correctly so the compiler assumes it returns an int. In what I assume
is a 64-bit build this will lose the top 32 bits of the CFStringRef.
Yes, your diagnosis holds water. I'm building an
nders ; cocoa-dev@lists.apple.com
Sent: Sunday, June 20, 2010 11:52 AM
Subject: Re: Getting Computer Name
Here is my method:
+ (NSString *)
computerName
{
CFStringRef temp = SCDynamicStoreCopyComputerName (NULL, NULL);
NSString* name = [NSString stringWithString: temp];
Hi Paul,
Thanks for your reply.
On 20/6/10, Paul Sanders wrote:
I use #1 and it works fine for me. Note that
SCDynamicStoreCopyComputerName might return NULL, and don't forget to
CFRelease temp.
OK, thanks.
In the case of #1, what is the NSLog statement that is failing? And I
take it tha
On 20 Jun 2010, at 11:10, Steve Steinitz wrote:
>
>
> According to the docs CFStringRef is toll-free-bridged with NSString and so
> interchangeable. The authors of the examples cite no issues. Could I have
> done something to my project to break toll-free bridging? I confess, I
> haven't t
ive
until the current autorelease pool 'pops'. You can look at name (and probably
temp, but I'm not sure) in the debugger, of course.
Regards,
Paul Sanders.
- Original Message -
From: Steve Steinitz
To: cocoa-dev@lists.apple.com
Sent: Sunday, June 20, 2010 11:10 AM
Hello,
I've found several examples of getting the computer Name but
some give warnings about making a pointer from an integer and
they all fail with signal EXC_BAD_ACCESS here
objc_msgSend_vtable5
_NSDescriptionWithLocaleFunc
_CFStringAppendFormatAndArgumentsAux
_CFStringCreat
21 matches
Mail list logo
