February 2006 - Darwin-dev - lists.apple.com

Re: ioerror writing files to disk on PSNormalizer (via pstopdf)
by marcantoni＠femfum.com 03 Feb '06

03 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: SquirrelMail/1.4.3a > try this with the following command > > pstopdf ps_2_pdf_test.ps -o output.pdf > > > -- Dale Hi Dale, Your ps_2_pdf_test.ps file is a simply page description PostScript file and don't attempts to write any data inside the device disc. My question refers directly to the PostScript permissions to write data to disc. Exec the attached file via 'pstopdf' inside the OSX Terminal and see this result... FemFums-Computer:~ femfum$ pstopdf viam.ps -o ostres.pdf %%[ Error: ioerror; OffendingCommand: file ]%% Stack: (w) (ThisISaNewFileInDisk.txt) Understand my problem ? Thanks again, -- MaRC anToni Malagarriga i Picas T·(34)938721642 | (34)667517069 (][ www.femfum.com PostScript&PDF Tool Developers Desenvolupadors d'Eines PostScript i PDF _______________________________________________ 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 viam.ps.zip

1 0

ioerror writing files to disk on PSNormalizer (via pstopdf)
by marcantoni＠femfum.com 03 Feb '06

03 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: SquirrelMail/1.4.3a Hi, We relaunch a question posted some days ago on the Darwin-userlevel mailing list... Executing PS code via 'pstopdf' that writes a new file inside the disk, the interpreter fails and outputs an *ioerror*. I think related to the PS interpreter privileges in this context or perhaps a 'pstopdf' serious bug. But if I write this file in the virtual %ram% disc, the file writing is enabled with no problem. To demonstrate that, please exec this PS code inside PSNormalizer (via pstopdf)... (*) {dup (\n\n>>>) print flush print flush (<<<\n) print flush currentdevparams{== ==}forall} 256 string /IODevice resourceforall ...this code shows through the interpreter prompt the all available devices. Observing the PSNormalizer output result from the above PS call, we have three *writable* devices inside the PS context: %os% %disk0% and %ram% To demonstrate that I list below the result of this call inside an OSX 10.3 machine, via Terminal and pstopdf command... >>>%hostfont%<<< /FileSystem /Type false /Writeable true /Mounted 0 /SearchOrder true /HasNames true /Searchable false /Removable >>>%Distillery%<<< /Communications /Type >>>%os%<<< 0 /InitializeAction /FileSystem /Type true /Writeable 11246016 /Free true /Mounted 2 /SearchOrder true /HasNames false /Searchable 39066000 /LogicalSize false /Removable >>>%fontset%<<< /FileSystem /Type false /Writeable true /Mounted 10 /SearchOrder true /HasNames true /Searchable false /Removable >>>%disk0%<<< 0 /InitializeAction /FileSystem /Type true /Writeable 11246016 /Free true /Mounted 1 /SearchOrder true /HasNames true /Searchable 39066000 /LogicalSize false /Removable >>>%cid%<<< /FileSystem /Type false /Writeable true /Mounted 99 /SearchOrder true /HasNames true /Searchable false /Removable >>>%ram%<<< /FileSystem /Type true /Writeable 20000000 /Free true /Mounted 100 /SearchOrder 20000000 /PhysicalSize true /HasNames 1 /BlockSize false /Searchable 0 /InitializeAction 20000000 /LogicalSize false /Removable To summarize, if you exec this following code inside the PSNormalizer via pstopdf... (%ram%myFile) (w) file dup (AEIOU) writestring closefile (%ram%myFile) (r) file 64 string readstring pop == ... the prompt shows the (AEIOU) string with no problem The first line writes the (AEIOU) string iniside the file 'myFile' located in the %ram% device The second line reads from myFile the first 64 bytes and outputs this to the prompt (stdout) ... if you change the device origin of this code to %disk0% or %os% ... (%disk0%myFile) (w) file dup (AEIOU) writestring closefile (%disk0%myFile) (r) file 64 string readstring pop == ...the 'pstopdf' execution shows this weird error... %%[ Error: ioerror; OffendingCommand: file ]%% Stack: (w) (%disk0%myFile) ...more simply, can you exec... (myFile) (w) file (AEIOU) writestring ...relative to the default interpreter device storage, but you obtain the same *ioerror* AdobeAcrobatDistiller and Ghostscript works well when the IODevice resource reads... /Writeable true ...inside a concrete Device parameters dictionary. What happens in this context ? Have you a clear solution ? Any Apple/Darwin technician can help my? Many thanks, -- MaRC anToni Malagarriga i Picas T·(34)938721642 | (34)667517069 (][ www.femfum.com PostScript&PDF Tool Developers Desenvolupadors d'Eines PostScript i PDF _______________________________________________ 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: finding the offset of a partition
by Alexander von Below 03 Feb '06

03 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Am 03.02.2006 um 07:49 schrieb Jason Bobier: Bottom line: You have to get access to the partition map. Alex _______________________________________________ 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 can't seem to find a way to find the offset of a partition on a drive without reading the partition table myself. Is there an easy way to do this? i.e. given disk1s1 what is it's offset from the beginning of disk1? No, as Justin already suspected nothing inside a partition tells you anything about the location of that partition on the disk. And it would not make much sense either, as partitions are meant to be self- contained and portable. For example, you can easily copy a partition into a disk image without modifying anything within the partition. This email sent to site_archiver(a)lists.apple.com

1 0

Re: pb compiler
by Locatelli Jérôme 02 Feb '06

02 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thank you Le 2 févr. 06, à 16:54, Peter O'Gorman a écrit : -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Shantonu Sen wrote: | Probably because you're using -all_load | | Shantonu | | On Feb 2, 2006, at 5:22 AM, Locatelli Jérôme wrote: | |> Hello, |> |> I try to compile an application on MacOS 10.4.4 with g++, using |> GNU libtool 1.5.10. |> I've got the following message. I can't correct it by removing |> "all_load" in the libtool script. |> |> Does it exist any other solutions?? Thank. | GNU libtool had the -all_load flag removed for darwin with this patch: 2004-08-02 Geoffrey Keating <geoffk(a)apple.com> ~ * libtool.m4 [darwin]: Don't use -all_load. Which came before this one: 2004-08-07 Peter O'Gorman <peter(a)pogma.com> ~ GNU libtool-1.5.8 is released. Peter -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (Darwin) iQCVAwUBQ+IrO7iDAg3OZTLPAQLSUQP/ZTerHWTwjn2DVC0Tcm9LCCFn3ISOZsf1 BDe/0pqfwEDP0nbOAmphkpLD4IYncdCwJfV7MovQP9tFUZnOkFK0ZJ/COukVp57P 5tqYY06+71j/373F1SJ190F3aODmgbcO27r7eqOgSSitmzrUnKms8kXJSXb3HgVN NmPUWLxRsB8= =ik5U -----END PGP SIGNATURE----- --------------------------------------------------- Locatelli Jérôme Institut Laue-Langevin Service Contrôle des instruments 6 rue Jules Horowitz, BP 156 38042 Grenoble cedex 9 http://www.ill.fr/ Tel: 33 (0)4 76 20 76 27 Fax: 33 (0)4 76 20 77 00 --------------------------------------------------- _______________________________________________ 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… OK , I found a wrong libtool.m4 in /usr/share/aclocal/ , I replace it by the good one from /sw/share/aclocal/ So if you're getting -all_load in your link line, and you're not adding it yourself, you're not using libtool-1.5.10. Ensure that the libtool.m4 that is put in your configure script does not have the -all_load flag. This email sent to site_archiver(a)lists.apple.com

1 0

Re: finding the offset of a partition
by Justin C. Walker 02 Feb '06

02 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Feb 2, 2006, at 22:49 , Jason Bobier wrote: Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large Institute for the Absorption of Federal Funds -------- If you're not confused, You're not paying attention -------- _______________________________________________ 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 can't seem to find a way to find the offset of a partition on a drive without reading the partition table myself. Is there an easy way to do this? i.e. given disk1s1 what is it's offset from the beginning of disk1? AFAIK, there is nothing intrinsic about partitions. It's the partition table that defines them. That's what the system uses to find them, when partitions on a disk are mounted. Why do you think it's otherwise possible? This email sent to site_archiver(a)lists.apple.com

1 0

finding the offset of a partition
by Jason Bobier 02 Feb '06

02 Feb '06

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… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Track leaked Filedescriptors
by Totte Alm 02 Feb '06

02 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hello folks! Correction! Case closed for me anyways. Thanks for all support here in this. Stuck between a rock and a damn hard place. // Totte - Code 66 _______________________________________________ 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… OK, bug filed, Bug ID# 4425932 Reproducible on 10.4 -> 10.4.3. This seems to be fixed on 10.4.4. The XServe has been up for 3 days now, before we had < 12 hour uptime, and I cannot rep this on my 10.4.4 machine, allthough its a piece of cake on 10.4.3 and below. If anyone wanna persue what changed between 10.4.3 and 10.4.4 that made this horrible behavior to go away, I rally wanna know where the changes were made, so I can better understand what really happend. // Totte I know the setup is a little demanding but Iried to explain as good as possible. Thanks for taking this seriously. This is a customer mission critical system that this started to happen on when we went from 10.3 ->Z 10.4 and at the same time updated the DBServer. We cannot step backwards, as we have "updated" the databases for the new format, and I guess backing down would just be a horrible by itself. This email sent to site_archiver(a)lists.apple.com

1 0

ASL - the long-winded answer
by Marc Majka 02 Feb '06

02 Feb '06

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… This email sent to site_archiver(a)lists.apple.com

1 0

Re: launchd and unix sockets
by Dave Zarzycki 02 Feb '06

02 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com davez On Feb 2, 2006, at 12:57 PM, Graham J Lee wrote: Cheers, Graham. -- Graham Lee UNIX Systems Manager, Oxford Physics Practical Course http://users.ox.ac.uk/~wadh1342 _______________________________________________ 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/zarzycki%40apple.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… Well, I can only speak for launchd. In this case, the sources for Intel are simply newer than the PowerPC sources. But the next software update of launchd should bring the PowerPC side up to date. On 2/2/06 20:03, Dave Zarzycki wrote: Yes an no. Not in earlier Tiger builds, but I believe we fixed this for 10.4.4 for Intel, I'm not sure if this change showed up in 10.4.4 for PowerPC. That, and the latest Darwin release, both imply that the sources for 10.4.x/`arch` are coming from different, um, sources; would that be the case? Will there be a unified source release at any point? This email sent to zarzycki(a)apple.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: launchd and unix sockets
by Graham J Lee 02 Feb '06

02 Feb '06

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Thunderbird 1.5 (Macintosh/20051201) Cheers, Graham. -- Graham Lee UNIX Systems Manager, Oxford Physics Practical Course http://users.ox.ac.uk/~wadh1342 _______________________________________________ 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 2/2/06 20:03, Dave Zarzycki wrote: Yes an no. Not in earlier Tiger builds, but I believe we fixed this for 10.4.4 for Intel, I'm not sure if this change showed up in 10.4.4 for PowerPC. That, and the latest Darwin release, both imply that the sources for 10.4.x/`arch` are coming from different, um, sources; would that be the case? Will there be a unified source release at any point? This email sent to site_archiver(a)lists.apple.com

1 0