August 2005 - Darwin-dev - lists.apple.com

Obtaining full Darwin source as image or tarballs
by wadeslists＠mac.com 10 Aug '05

10 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Is there any way to obtain the full Darwin source, as an image or in project tarballs, easily? I can't grab them off Apple's web page because they insist you login every 10 minutes, regardless of activity, which invariably screws up every automated downloader I've tried. I'd rather not use CVS because it's far too slow, but I am aware that's a last ditch option. Wade Tregaskis (AIM/iChat, Yahoo & Skype: wadetregaskis, ICQ: 40056898, MSN: wjtregaskis(a)students.latrobe.edu.au, AV iChat & email: wadetregaskis(a)mac.com, Jabber: wadetregaskis(a)jabber.org) -- Sed quis custodiet ipsos custodes? This email sent to site_archiver(a)lists.apple.com

1 0

Re: auto_ptr not behaving as per header comment
by Alexey Proskuryakov 10 Aug '05

10 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com struct foo { }; struct bar : public foo { }; std::auto_ptr<bar> get_bar() { return std::auto_ptr<bar>(new bar); } int main() { std::auto_ptr<foo> ptr = get_bar(); return 0; } void use_foo(std::auto_ptr<foo>); std::auto_ptr<bar> aBar; use_foo(aBar); - WBR, Alexey Proskuryakov _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Tue, 9 Aug 2005 18:56:24 -0700 Steve Checkoway <steve(a)cs.washington.edu> wrote: This code looks intuitively correct, so it *should* be OK, in general sense. A closely related issue in the C++ standard is described here: <http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#463> (may even be exactly the same, I'm not sure). When corrections to the standard introduce such severe regressions, I don't know who is to blame. CodeWarrior compiles this just fine (why am I not surprised?), but still doesn't compile other intuitively correct code (grrrr...): I'd expect this issue to have been extensively discussed, filed in the gcc bugzilla etc - I just haven't researched it a lot. Of Apple mail lists, xcode-users appears the most relevant. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Re: Changing Mach-O app load behavior
by Eric Shepherd 10 Aug '05

10 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Eric Shepherd Owner Syndicomm http://www.syndicomm.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Eric, you could have a small loader program run which then would load the rest of your binary as a dylib, using NSCreateObjectFileImageFromMemory. The problem is that one of the requirements of my project is that it not require any recompilation of the wrapped program, which may have been written previously by another company entirely. Otherwise making the app load as a dylib that way would be a great solution. Unfortunately, in my particular case, I have a very stringent set of rules for what I'm supposed to accomplish: 1. The wrapped application can't be easily retrieved and used outside the wrapper. 2. The wrapped application can't be required to be altered or recompiled. On Windows, this is done by loading the encrypted app into memory from disk within the wrapper application, decrypting it, then executing it, but it's not looking like that will work on OS X at this point, so I need to come up with another solution, I guess. This email sent to site_archiver(a)lists.apple.com

1 0

auto_ptr not behaving as per header comment
by Steve Checkoway 10 Aug '05

10 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I decided to give it a try: $ cat auto_ptr_test.cpp #include <memory> struct foo { }; struct bar : public foo { }; std::auto_ptr<bar> get_bar() { return std::auto_ptr<bar>(new bar); } However, if I comment out the std::auto_ptr<foo> ptr = get_bar(); line, it compiles. It looks like the third candidate: std::auto_ptr<_Tp>::auto_ptr(std::auto_ptr<_Tp>&) [with _Tp = foo] should match std::auto_ptr<foo>::auto_ptr(std::auto_ptr<foo>) but for some reason, it isn't. - Steve _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com I was taking a look at /usr/include/gcc/darwin/4.0/c++/memory and I noticed this comment: /** * @brief Automatic conversions * * These operations convert an %auto_ptr into and from an auto_ptr_ref * automatically as needed. This allows constructs such as * @code * auto_ptr<Derived> func_returning_auto_ptr(.....); * ... * auto_ptr<Base> ptr = func_returning_auto_ptr(.....); * @endcode */ int main() { std::auto_ptr<foo> ptr = get_bar(); std::auto_ptr<foo> ptr2(get_bar()); return 0; } $ g++ -Wall auto_ptr_test.cpp auto_ptr_test.cpp: In function 'int main()': auto_ptr_test.cpp:10: error: no matching function for call to 'std::auto_ptr<foo>::auto_ptr(std::auto_ptr<foo>)' /usr/include/gcc/darwin/4.0/c++/memory:349: note: candidates are: std::auto_ptr<_Tp>::auto_ptr(std::auto_ptr_ref<_Tp>) [with _Tp = foo] /usr/include/gcc/darwin/4.0/c++/memory:212: note: std::auto_ptr<_Tp>::auto_ptr(std::auto_ptr<_Tp1>&) [with _Tp1 = foo, _Tp = foo] /usr/include/gcc/darwin/4.0/c++/memory:199: note: std::auto_ptr<_Tp>::auto_ptr(std::auto_ptr<_Tp>&) [with _Tp = foo] auto_ptr_test.cpp:10: error: initializing temporary from result of 'std::auto_ptr<_Tp>::operator std::auto_ptr<_Tp1>() [with _Tp1 = foo, _Tp = bar]' Am I doing something incorrectly here or is there a bug in g++ or is the header file simply wrong? (Or a combination of the above?) smime.p7s

1 0

Re: Porting MIDI plug-in from CW to Xcode
by Pete Gontier 10 Aug '05

10 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Microsoft-Entourage/11.1.0.040913 _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Changing Mach-O app load behavior
by Rainer Brockerhoff 09 Aug '05

09 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com At 12:06 -0700 8/9/05, darwin-dev-request(a)lists.apple.com wrote: >From: Eric Shepherd <sheppy(a)syndicomm.com> >Date: Tue, 9 Aug 2005 11:41:40 -0400 >Message-ID: <1F76B17F-48E4-4DE1-82CC-9150F58CF6C6(a)syndicomm.com> > >I'm in the odd position of needing to intervene in the process of loading a Mach-O application's binary. I need to be able to load it into memory, do some preprocessing on it, then link and run it. I need to be able to do this without requiring any changes to an existing application; the goal is to be able to wrap an existing program to control its use during a demo period. > >I've looked at things like NSCreateObjectFileImageFromMemory, which sounds perfect, except it only works on bundles, which apparently doesn't include standard applications. Rebuilding an app with the -bundle flag to the linker makes it "almost" work, in that the app launches but the menu bar never appears, and keystrokes don't go to its windows. Eric, you could have a small loader program run which then would load the rest of your binary as a dylib, using NSCreateObjectFileImageFromMemory. I could go into details in private, if you need further hints. -- Rainer Brockerhoff <rainer(a)brockerhoff.net> Belo Horizonte, Brazil "The three chief virtues of a programmer are: Laziness, Impatience and Hubris" - Larry Wall Weblog: http://www.brockerhoff.net/bb/viewtopic.php _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Changing Mach-O app load behavior
by Eric Shepherd 09 Aug '05

09 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Eric Shepherd Owner Syndicomm http://www.syndicomm.com On Aug 9, 2005, at 5:12 PM, Finlay Dobbie wrote: On 09/08/05, Eric Shepherd <sheppy(a)syndicomm.com> wrote: Folks are contacting me basically saying this is going to be a huge amount of work, which I pretty much figured. Anybody know why I can't get dyld to build? I'm kind of curious why it's putting up such a fight. If you're trying to build bits of Darwin, you should really use darwinbuild <http://www.opendarwin.org/projects/darwinbuild/>. It can be a tricky business. -- Finlay _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Ah-ha! I had been trying to find that but couldn't find a link to it anywhere. Thanks! This email sent to site_archiver(a)lists.apple.com

1 0

Re: Changing Mach-O app load behavior
by Finlay Dobbie 09 Aug '05

09 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=WLpPjMdGIe6h0vetm+D5maC33roSW4kdqsCUPw4t551ptr+QfMoHTADVZOaoYJfC+zIqYlsC5m/yESLu/mcx5xdO5JGj9UZ/f5NS0kIGNcLNmx8qNYdDsDWQPbQ/3RhsQ03Gj+dZPGM/HOwROPhdcoC6S2q7903ebVWOyMGsuvI= On 09/08/05, Eric Shepherd <sheppy(a)syndicomm.com> wrote: > Folks are contacting me basically saying this is going to be a huge > amount of work, which I pretty much figured. Anybody know why I > can't get dyld to build? I'm kind of curious why it's putting up > such a fight. If you're trying to build bits of Darwin, you should really use darwinbuild <http://www.opendarwin.org/projects/darwinbuild/>. It can be a tricky business. -- Finlay _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Changing Mach-O app load behavior
by Eric Shepherd 09 Aug '05

09 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Eric Shepherd Owner Syndicomm http://www.syndicomm.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Folks are contacting me basically saying this is going to be a huge amount of work, which I pretty much figured. Anybody know why I can't get dyld to build? I'm kind of curious why it's putting up such a fight. The assembler doesn't work automatically from Xcode; I have to manually re-issue the commands to build the two assembly files, changing the "-x" assembler to "-x assembler-with-cpp". And after that, it won't link, with a series of errors about multiply-defined functions. This email sent to site_archiver(a)lists.apple.com

1 0

Re: finding local shares
by Jason Townsend 09 Aug '05

09 Aug '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Philip, This might be expedient, but it's not without peril. :-( -Jason _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Aug 8, 2005, at 6:56 PM, Jordan K. Hubbard wrote: On Aug 8, 2005, at 4:58 PM, Jason Townsend wrote: On Aug 4, 2005, at 3:11 PM, Philip D. Wasson wrote: I'm trying to figure out how to generate a list of local shares (exports, whatever). More precisely, given the share name someone would see from a different machine, and knowing the share is from the machine I'm running on, how do I find the path to that directory? I thought I had it figured out, using a combination of property lists and NetInfo (./config/SharePoints, ./users/<username>/ sharedDir), but now I find that in 10.3 and later I can't use NetInfo. Is that information now accessed via Open Directory? I don't suppose anyone has done something like this already and could point me at some sample code? (I guess anything that reads something that is in NetInfo the new way would be extremely helpful.) TIA There are two options... the most expedient one would be to grab the headers for the NetInfo APIs from the appropriate Darwin project and just continue using them. This can be done by grabbing the appropriate version of the Libinfo code from http:// darwinsource.opendarwin.org or http://www.opensource.apple.com/ darwinsource/. The netinfo headers were removed because the APIs are now fully deprecated and will go away completely in future releases of the OS. For now, it's may be a reasonable stop-gap solution, just be aware that it's going to break at some point. Right... in my haste to respond I didn't fully explain the caveats about this option. This email sent to site_archiver(a)lists.apple.com

1 0