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

Re: wchar_t and printf not working
by Clark Cox 29 Mar '05

29 Mar '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:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=WfJ9nhyq5im4Xc+R/8uVv8B0bmeuxYPxq9Gk+nzLB4KfVWCGrKCeYybYQ25Yn0sKfkiP27dmq+hl49jbLqsyOMCCeKylXHGke1zBdjwOUsDCJ9Gl65bdFkPHC/+mRsZVvIpczySO3Zf8P8UuHXk5mC+HqLo6fQNJzMBHsIclRRY= On Mon, 28 Mar 2005 14:09:11 -0500, Michael B Allen <mba2000(a)ioplex.com> wrote: > On Mon, 28 Mar 2005 08:39:25 -0500 > Clark Cox <clarkcox3(a)gmail.com> wrote: > > > Actually, you will *never* see UTF-8 with more than 4 octets per > > codepoint. Period. That is the way that UTF-8 is defined. If you see a > > 5 or 6 octet character, then you are not reading UTF-8 data. > > This is incorrect. Please read the first sentence in section 2 of > RFC 2279 RFC 2279 has been obsoleted by RFC 3629 for years now: http://www.ietf.org/rfc/rfc3629.txt >From said RFC: [snip] 3. UTF-8 definition UTF-8 is defined by the Unicode Standard [UNICODE]. Descriptions and formulae can also be found in Annex D of ISO/IEC 10646-1 [ISO.10646] In UTF-8, characters from the U+0000..U+10FFFF range (the UTF-16 accessible range) are encoded using sequences of 1 to 4 octets. [snip] Also, note that the RFC is *not* authoritative. From the same RFC: [snip] NOTE -- The authoritative definition of UTF-8 is in [UNICODE]. This grammar is believed to describe the same thing Unicode describes, but does not claim to be authoritative. Implementors are urged to rely on the authoritative source, rather than on this ABNF. [snip] -- Clark S. Cox III clarkcox3(a)gmail.com http://www.livejournal.com/users/clarkcox3/ http://homepage.mac.com/clarkcox3/ _______________________________________________ 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: AppleOnboardAudio-247.2.2 (darwin 7.8)
by Benjamin Herrenschmidt 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Mon, 2005-03-28 at 15:28 -0800, Kevin Van Vechten wrote: > Hi Ben, > > I was able to get this to compile simply by defining the missing > enumerated types. Give this a try: > > diff -r -u AppleOnboardAudio-247.2.2.orig/AppleOnboardAudio/ > AppleOnboardAudioUserClient.h AppleOnboardAudio-247.2.2/ > AppleOnboardAudio/AppleOnboardAudioUserClient.h > --- AppleOnboardAudio-247.2.2.orig/AppleOnboardAudio/ > AppleOnboardAudioUserClient.h 2005-02-08 19:28:52.000000000 -0800 > +++ AppleOnboardAudio-247.2.2/AppleOnboardAudio/ > AppleOnboardAudioUserClient.h 2005-03-12 15:50:43.000000000 -0800 > @@ -22,7 +22,9 @@ > kCodec_TAS3004, > kCodec_CS8406, > kCodec_CS8420, > - kCodec_CS8416 > + kCodec_CS8416, > + kCodec_PCM3052, > + kCodec_NumberOfTypes > } ; Yah, I got it to build with the same trick, however, there are obviously entire parts of the driver that have been "stripped", like the Toonie driver (which is funny, as is basically does nothing but toggle a couple of GPIOs, no rocket science or critical IP in there :) Oh well, I'll wait for the next official release and see what it looks like. Ben. _______________________________________________ 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: AppleOnboardAudio-247.2.2 (darwin 7.8)
by Kevin Van Vechten 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Ben, On Mar 27, 2005, at 3:10 PM, Benjamin Herrenschmidt wrote: Hi ! It seems like various codec type definitions are missing. _______________________________________________ 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… I was able to get this to compile simply by defining the missing enumerated types. Give this a try: diff -r -u AppleOnboardAudio-247.2.2.orig/AppleOnboardAudio/ AppleOnboardAudioUserClient.h AppleOnboardAudio-247.2.2/ AppleOnboardAudio/AppleOnboardAudioUserClient.h --- AppleOnboardAudio-247.2.2.orig/AppleOnboardAudio/ AppleOnboardAudioUserClient.h 2005-02-08 19:28:52.000000000 -0800 +++ AppleOnboardAudio-247.2.2/AppleOnboardAudio/ AppleOnboardAudioUserClient.h 2005-03-12 15:50:43.000000000 -0800 @@ -22,7 +22,9 @@ kCodec_TAS3004, kCodec_CS8406, kCodec_CS8420, - kCodec_CS8416 + kCodec_CS8416, + kCodec_PCM3052, + kCodec_NumberOfTypes } ; // // HardwarePlugin State Just a note: AppleOnboardAudio-247.2.2 (darwin 7.8) doesn't build. Though Darwin 7.8 isn't referenced from the main publicsource page, so I suppose it might just be "unsupported" release :) Ben. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Where to find apple-gcc-1671 source? (wanted for distccd cross compiler)
by Kevin Van Vechten 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com <http://www.opensource.apple.com/darwinsource/November2004GCCUpdate/> Sorry for the oversight and inconvenience. - Kevin On Mar 26, 2005, at 1:36 AM, Adam Cohen wrote: Can anyone tell me how I can obtain the apple gcc release 1671 source (cvs or otherwise), corresonding to the November 2004 developer tools gcc update? I've been trying to build a ppc cross compiler for use with distccd, and although I've successfully built Apple's gcc-1495 on debian x86, I'd like to use the exact same version that I'm using on my development machines. I've tried using the Apple cvs repository for gcc3 located at: :pserver:anonymous@anoncvs.opensource.apple.com:/cvs/root but I couldn't find the apple-gcc-1671 branch. I also tried using the FSF anonymous cvs access at: :ext:anoncvs@savannah.gnu.org:/cvsroot/gcc but the earliest release I could find was apple-gcc-1740. So I take it somewhere between release 1495 and 1740, the cvs repository was transitioned from the Apple servers to the FSF servers, and it seems as though apple-gcc-671 is lost somewhere in cvs limbo. So if anyone has any idea where I can get a copy of this release, please let me know. Thanks, _______________________________________________ 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… These sources have been added to the Darwin web site as a .tar.gz download. The project is "gcc_os-1671" and is available at: Adam This email sent to site_archiver(a)lists.apple.com

1 0

Re: wchar_t and printf not working
by Brian Sammon 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com > > Actually, you will *never* see UTF-8 with more than 4 octets per > > codepoint. Period. That is the way that UTF-8 is defined. If you see a > > 5 or 6 octet character, then you are not reading UTF-8 data. > > This is incorrect. Please read the first sentence in section 2 of > RFC 2279: According to http://www.ietf.org/rfc/rfc3629.txt RFC 2279 has been obsoleted by RFC 3629. I haven't bothered to check if 3629 supports your argument. _______________________________________________ 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: wchar_t and printf not working
by Alexey Proskuryakov 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Mon, 28 Mar 2005 14:09:11 -0500 Michael B Allen <mba2000(a)ioplex.com> wrote: Actually, you will *never* see UTF-8 with more than 4 octets per codepoint. Period. That is the way that UTF-8 is defined. If you see a 5 or 6 octet character, then you are not reading UTF-8 data. This is incorrect. Please read the first sentence in section 2 of RFC 2279: - 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… Please read RFC 3251 <http://www.faqs.org/rfcs/rfc3251.html> "Electricity over IP", then! RFCs often have mistakes, and are complete nonsense sometimes. UTF-8 is defined in the Unicode standard. This email sent to site_archiver(a)lists.apple.com

1 0

Re: wchar_t and printf not working
by Michael B Allen 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Mon, 28 Mar 2005 08:39:25 -0500 Clark Cox <clarkcox3(a)gmail.com> wrote: > Actually, you will *never* see UTF-8 with more than 4 octets per > codepoint. Period. That is the way that UTF-8 is defined. If you see a > 5 or 6 octet character, then you are not reading UTF-8 data. This is incorrect. Please read the first sentence in section 2 of RFC 2279: 2. UTF-8 definition In UTF-8, characters are encoded using sequences of 1 to 6 octets. Mike [1] http://www.ietf.org/rfc/rfc2279.txt -- IRC - where men are men, women are men, and the boys are FBI agents. _______________________________________________ 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: wchar_t and printf not working
by Clark Cox 28 Mar '05

28 Mar '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:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=oRmIVb3bJBb2Co4JdHj0OSIoH2g7iM3huiEIv7xnKwl6mXkpZ3uBZLftY2RBd0AVEYqoMjbRHcTfC9+DwJFuRsUi19QLa7RVw9BdLKX3jmS1+nxqRAEYTOEZlH9gMzPNgYoBiHBxK+iF6FJCZgdXrLw3axC2jJpZvz9XN4oQq8A= On Mon, 28 Mar 2005 03:51:50 -0500, Michael B Allen <mba2000(a)ioplex.com> wrote: > On Mon, 28 Mar 2005 11:36:10 +0400 > Alexey Proskuryakov <ap-carbon(a)rambler.ru> wrote: > > > > Each character may occupy between 1 and 6 bytes [1]. > > > > More precisely, between 1 and 4: > > <http://www.unicode.org/faq/utf_bom.html#30>. > > At risk of being pedantic this is just talking about how to convert a > UTF-16 character into a UTF-8 one. Because UTF-16 with a surrogate can > only represent 21 bits of the Unicode code space only 4 bytes is necessary > to encode any character in UTF-8. Unicode *only hase* 21 bits of code space, even UTF-32 only uses 21 bits. > But UTF-8 can encode the full 31 bit code space which needs at most 6 bytes to be > represented in UTF-8. But unless you're doing Klingon you'll never actually see more > than 4. Actually, you will *never* see UTF-8 with more than 4 octets per codepoint. Period. That is the way that UTF-8 is defined. If you see a 5 or 6 octet character, then you are not reading UTF-8 data. -- Clark S. Cox III clarkcox3(a)gmail.com http://www.livejournal.com/users/clarkcox3/ http://homepage.mac.com/clarkcox3/ _______________________________________________ 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: wchar_t and printf not working
by Alexey Proskuryakov 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Microsoft-Entourage/11.1.0.040913 On 28.03.2005 12:51, "Michael B Allen" <mba2000(a)ioplex.com> wrote: >> More precisely, between 1 and 4: >> <http://www.unicode.org/faq/utf_bom.html#30>. > > At risk of being pedantic this is just talking about how to convert a > UTF-16 character into a UTF-8 one. That's because the only imaginable way to come up with something that looks similar to a 6-byte UTF-8 used to be an incorrect conversion from UTF-16 :) > Because UTF-16 with a surrogate can > only represent 21 bits of the Unicode code space only 4 bytes is necessary > to encode any character in UTF-8. But UTF-8 can encode the full 31 bit > code space which needs at most 6 bytes to be represented in UTF-8. But > unless you're doing Klingon you'll never actually see more than 4. Actually, the Unicode code space is currently not 32 or 31 bits, it's 0...10FFFF (of which most codes are unassigned, but that's another topic): <http://www.unicode.org/faq/utf_bom.html#9>. The FAQ also explicitly states that UTF-8 needs 4 bytes as a maximum. AFAIK, there's no official guarantee that a future version of Unicode won't extend the code space yet, but such a guarantee is being considered. Aren't we too offtopic now? And after all, I'm mostly citing a single page of the Unicode FAQ, which anyone interested can and should read. - 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… This email sent to site_archiver(a)lists.apple.com

1 0

Re: wchar_t and printf not working
by Michael B Allen 28 Mar '05

28 Mar '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Mon, 28 Mar 2005 11:36:10 +0400 Alexey Proskuryakov <ap-carbon(a)rambler.ru> wrote: > > Each character may occupy between 1 and 6 bytes [1]. > > More precisely, between 1 and 4: > <http://www.unicode.org/faq/utf_bom.html#30>. At risk of being pedantic this is just talking about how to convert a UTF-16 character into a UTF-8 one. Because UTF-16 with a surrogate can only represent 21 bits of the Unicode code space only 4 bytes is necessary to encode any character in UTF-8. But UTF-8 can encode the full 31 bit code space which needs at most 6 bytes to be represented in UTF-8. But unless you're doing Klingon you'll never actually see more than 4. Mike -- IRC - where men are men, women are men, and the boys are FBI agents. _______________________________________________ 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