From mekentosj at gmail.com Mon Sep 11 10:38:44 2006 From: mekentosj at gmail.com (Alexander Griekspoor) Date: Mon, 11 Sep 2006 16:38:44 +0200 Subject: [Biococoa-dev] Xgrid & Biococoa mentioned Message-ID: <13ADE4EE-7C5B-4A79-8CED-1287D55669C1@gmail.com> Hi guys, Perhaps an interesting read: http://www.macresearch.org/ interview_with_alexander_griekspoor_from_mekentosj Charles, nice Drew asked about our widget, hope it again brings some people to the project. Koen, I do realize that you and the other still wait for my promised report on the WWDC. My thesis will hopefully be sent to the printer by the end of the week, then I should have time to breath again (and write reports). Cheers, Alex ********************************************************* ** Alexander Griekspoor ** ********************************************************* The Netherlands Cancer Institute Department of Tumorbiology (H4) Plesmanlaan 121, 1066 CX, Amsterdam Tel: + 31 20 - 512 2023 Fax: + 31 20 - 512 2029 E-mail: a.griekspoor at nki.nl AIM: mekentosj at mac.com Web: http://www.mekentosj.com EnzymeX - To cut or not to cut 2006 Winner of the Apple Design Awards Best Mac OS X Scientific Solution http://www.mekentosj.com/enzymex ********************************************************* -------------- next part -------------- An HTML attachment was scrubbed... URL: From sweetcocoa at mac.com Mon Sep 11 12:46:02 2006 From: sweetcocoa at mac.com (Peter Schols) Date: Mon, 11 Sep 2006 18:46:02 +0200 Subject: [Biococoa-dev] Xgrid & Biococoa mentioned In-Reply-To: <13ADE4EE-7C5B-4A79-8CED-1287D55669C1@gmail.com> References: <13ADE4EE-7C5B-4A79-8CED-1287D55669C1@gmail.com> Message-ID: <90502752-3E5E-4E34-9C37-3C4BDF3552B4@mac.com> Great interview, congratulations Alex! (and good luck with the final work on your PhD, please send an invitation to the BC list) Peter On 11 Sep 2006, at 16:38, Alexander Griekspoor wrote: > Hi guys, > Perhaps an interesting read: > http://www.macresearch.org/ > interview_with_alexander_griekspoor_from_mekentosj > Charles, nice Drew asked about our widget, hope it again brings > some people to the project. > Koen, I do realize that you and the other still wait for my > promised report on the WWDC. My thesis will hopefully be sent to > the printer by the end of the week, then I should have time to > breath again (and write reports). > Cheers, > Alex > > ********************************************************* > ** Alexander Griekspoor ** > ********************************************************* > The Netherlands Cancer Institute > Department of Tumorbiology (H4) > Plesmanlaan 121, 1066 CX, Amsterdam > Tel: + 31 20 - 512 2023 > Fax: + 31 20 - 512 2029 > E-mail: a.griekspoor at nki.nl > AIM: mekentosj at mac.com > Web: http://www.mekentosj.com > > EnzymeX - To cut or not to cut > 2006 Winner of the Apple Design Awards > Best Mac OS X Scientific Solution > http://www.mekentosj.com/enzymex > > ********************************************************* > > > _______________________________________________ > Biococoa-dev mailing list > Biococoa-dev at bioinformatics.org > https://bioinformatics.org/mailman/listinfo/biococoa-dev From schristley at mac.com Thu Sep 21 17:49:49 2006 From: schristley at mac.com (Scott Christley) Date: Thu, 21 Sep 2006 17:49:49 -0400 Subject: [Biococoa-dev] Greetings Message-ID: Hello all, I first became aware of BioCocoa at the 2005 WWDC, at the time I was just starting to get involved in biology simulations and bioinformatics with my PhD; now that I'm deep in it, I would like to contribute code I've developed. I wasn't able to make it to the recent WWDC, but congratulations to the EnzymeX guys! One thing I am interested in is GNUstep support. From looking at the archives, I may be opening a can of worms but onward I go. Why I am interested in GNUstep is because I use both platforms. Specifically my Mac tends to be my development platform and for visualization, but when I kick off a bunch of simulations or do some heavy computation, I pass it over to the computer cluster which tends to be 64bit Opteron processors running Linux, so I use GNUstep there. My understanding from reading the mailing list archives is that optimizations to use CoreFoundation is what killed GNUstep compatibility. Are people still interested in this, if I were to tackle this issue, supporting both platforms in a clean way and maintaining the compatibility? From looking at the code, I think many of the differences can be handled with careful macros, keeping CoreFoundation speed on Mac but OpenStep code for GNUstep. As for BioCocoa itself, with some of the work I've been doing I've been writing various parsers as well as playing around with some ideas, and I've thought well if I do all this work it would be nice to make it available. I could just start my own project, but maybe it would be good to add this into BioCocoa. Anyways, these are the ideas I'm thinking: * Gene expression data, currently Affymextric data sets that can be gotten from GEO. Parser to load in the data, form links with ontology, genes, etc. * Ontologies, I have a plain parser for OBO which seems to be becoming a standard format. There is also the OWL variants which is semi-handled by XML parsing. * Web Services, I've mainly been playing around with web services provided by NCBI, EBI, and KEGG. This all falls under the general idea of integration of multiple data sources for doing bioinformatic analysis and simulations. I have some other ideas as well but this is a good start. Let me know what you think. cheers Scott From biococoa at bioworxx.com Fri Sep 22 05:01:30 2006 From: biococoa at bioworxx.com (Philipp Seibel) Date: Fri, 22 Sep 2006 11:01:30 +0200 Subject: [Biococoa-dev] Greetings In-Reply-To: References: Message-ID: <10D4D5CE-4454-4D52-A837-484DEE870F78@bioworxx.com> Hey Scott, > > I first became aware of BioCocoa at the 2005 WWDC, at the time I > was just starting to get involved in biology simulations and > bioinformatics with my PhD; now that I'm deep in it, I would like > to contribute code I've developed. I wasn't able to make it to the > recent WWDC, but congratulations to the EnzymeX guys! Nice to hear, contributions are welcome of course:-). > One thing I am interested in is GNUstep support. From looking at > the archives, I may be opening a can of worms but onward I go. Why > I am interested in GNUstep is because I use both platforms. > Specifically my Mac tends to be my development platform and for > visualization, but when I kick off a bunch of simulations or do > some heavy computation, I pass it over to the computer cluster > which tends to be 64bit Opteron processors running Linux, so I use > GNUstep there. Sounds interesting, in my opinion it would make sense to make as much as possible GNUstep compatible. > > My understanding from reading the mailing list archives is that > optimizations to use CoreFoundation is what killed GNUstep > compatibility. Are people still interested in this, if I were to > tackle this issue, supporting both platforms in a clean way and > maintaining the compatibility? From looking at the code, I think > many of the differences can be handled with careful macros, keeping > CoreFoundation speed on Mac but OpenStep code for GNUstep. > > As for BioCocoa itself, with some of the work I've been doing I've > been writing various parsers as well as playing around with some > ideas, and I've thought well if I do all this work it would be nice > to make it available. I could just start my own project, but maybe > it would be good to add this into BioCocoa. Anyways, these are the > ideas I'm thinking: > > * Gene expression data, currently Affymextric data sets that can be > gotten from GEO. Parser to load in the data, form links with > ontology, genes, etc. > > * Ontologies, I have a plain parser for OBO which seems to be > becoming a standard format. There is also the OWL variants which > is semi-handled by XML parsing. > > * Web Services, I've mainly been playing around with web services > provided by NCBI, EBI, and KEGG. > So i'm thinking of building a RNA Framework for BioCocoa so i think it would make sense if you could make a Microarray framework containing your code. So organizing code for different bioinformatic areas in different frameworks would be a good solution, i think. Like eg. the apple quartz.framework. > > This all falls under the general idea of integration of multiple > data sources for doing bioinformatic analysis and simulations. I > have some other ideas as well but this is a good start. > > Let me know what you think. I think it sounds promising :-). Btw. do you have support for MAGE-ML? Cheers, Phil From schristley at mac.com Fri Sep 22 10:57:55 2006 From: schristley at mac.com (Scott Christley) Date: Fri, 22 Sep 2006 10:57:55 -0400 Subject: [Biococoa-dev] Greetings In-Reply-To: <10D4D5CE-4454-4D52-A837-484DEE870F78@bioworxx.com> References: <10D4D5CE-4454-4D52-A837-484DEE870F78@bioworxx.com> Message-ID: <632DBD44-13EA-40A7-BAD1-5544828B283F@mac.com> Hello Philipp, On Sep 22, 2006, at 5:01 AM, Philipp Seibel wrote: > > So i'm thinking of building a RNA Framework for BioCocoa so i think > it would make sense if you could make a Microarray framework > containing your code. So organizing code for different > bioinformatic areas in different frameworks would be a good > solution, i think. Like eg. the apple quartz.framework. Do you mean like those umbrella frameworks, a framework with a bunch of frameworks within it? I tried looking into how to make those but it seems Xcode doesn't directly support building them. I guess there are some developers out there who figured out how to build them but the details are sketchy. Regardless, I like the idea and I agree that is better than either having a bunch of individual frameworks or one huge framework with everything in it. > I think it sounds promising :-). Btw. do you have support for MAGE-ML? I haven't worked with MAGE-ML directly, but because it is an XML format, parsing is straight-forward. It then becomes a matter pulling the desired info out of the XML. Scott From kvddrift at earthlink.net Fri Sep 22 22:07:12 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Fri, 22 Sep 2006 22:07:12 -0400 Subject: [Biococoa-dev] Greetings In-Reply-To: References: Message-ID: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> On Sep 21, 2006, at 5:49 PM, Scott Christley wrote: > Hello all, > > I first became aware of BioCocoa at the 2005 WWDC, at the time I > was just starting to get involved in biology simulations and > bioinformatics with my PhD; now that I'm deep in it, I would like > to contribute code I've developed. I wasn't able to make it to the > recent WWDC, but congratulations to the EnzymeX guys! Hi Scott, Welcome to BioCocoa - it's good to see new names on the mailing list! > One thing I am interested in is GNUstep support. From looking at > the archives, I may be opening a can of worms but onward I go. Why > I am interested in GNUstep is because I use both platforms. > Specifically my Mac tends to be my development platform and for > visualization, but when I kick off a bunch of simulations or do > some heavy computation, I pass it over to the computer cluster > which tends to be 64bit Opteron processors running Linux, so I use > GNUstep there. > > My understanding from reading the mailing list archives is that > optimizations to use CoreFoundation is what killed GNUstep > compatibility. Are people still interested in this, if I were to > tackle this issue, supporting both platforms in a clean way and > maintaining the compatibility? From looking at the code, I think > many of the differences can be handled with careful macros, keeping > CoreFoundation speed on Mac but OpenStep code for GNUstep. IIRC, we decided to no longer support GNUStep as of version 2.0, but I don't recall the reason for this. I am not familiar with GNUStep, but from what you are writing, we need to start using macros for various functions? > As for BioCocoa itself, with some of the work I've been doing I've > been writing various parsers as well as playing around with some > ideas, and I've thought well if I do all this work it would be nice > to make it available. I could just start my own project, but maybe > it would be good to add this into BioCocoa. I think it would be a great idea to have more functionality in the framework. cheers, - Koen. From schristley at mac.com Sun Sep 24 17:45:28 2006 From: schristley at mac.com (Scott Christley) Date: Sun, 24 Sep 2006 17:45:28 -0400 Subject: [Biococoa-dev] Greetings In-Reply-To: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> Message-ID: On Sep 22, 2006, at 10:07 PM, Koen van der Drift wrote: > I am not familiar with GNUStep, but from what you are writing, we > need to start using macros for various functions? Yes, it looks to be straight forward as most use of CoreFoundation appears to be working with arrays. So I've created an internal header file with macros like this. // Variable declarations #define DECLARE_INDEX(variable) CFIndex variable // Range operations #define MAKE_RANGE(location, length) \ CFRangeMake( (location), (length) ) // Array operations #define ARRAY_GET_VALUE_AT_INDEX(array, index) \ CFArrayGetValueAtIndex( (CFMutableArrayRef) (array), (index) ) #define ARRAY_APPEND_VALUE(array, object) \ CFArrayAppendValue ( (CFMutableArrayRef) (array), (object) ) #define ARRAY_INSERT_VALUE_AT_INDEX(array, index, object) \ CFArrayInsertValueAtIndex ( (CFMutableArrayRef) (array), (index), (object) ) #define ARRAY_RANGE_CONTAINS_VALUE(array, range, object) \ CFArrayContainsValue ( (CFArrayRef) (array), (range), (object) ) #define ARRAY_GET_COUNT(array) \ CFArrayGetCount ( (CFArrayRef) (array) ) // Set operations #define SET_CONTAINS_VALUE(set, object) \ CFSetContainsValue( (CFSetRef) (set), (object)); cheers Scott -------------- next part -------------- An HTML attachment was scrubbed... URL: From kvddrift at earthlink.net Sun Sep 24 18:18:34 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Sun, 24 Sep 2006 18:18:34 -0400 Subject: [Biococoa-dev] Re: GNUstep patch In-Reply-To: <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> Message-ID: <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> On Sep 24, 2006, at 5:54 PM, Scott Christley wrote: > I didn't want to send this patch to everyone on the mailing list, > so hopefully you can review and apply if it looks good. > > One file has been added to the Xcode project, BCInternal.h. > Everything else should be makefiles for GNUstep, unfortunately the > way the make system works under GNUstep, need a makefile in each > directory that has source files. This isn't a complete port, just > BCFoundation, but I will work on the other stuff soon. Scott, Thanks for sending the patch files. Before we apply them, I would acctually like to get some more input from the mailinglist and see what opinion others have. This is a pretty big change in the project files, and I don't want to just go ahead and apply it. Furthermore, is GNUStep aware of all the changes in recent OS X version, eg does it know about all the new classes in BCFoundation and BCAppkit? If there are any classes that GNUStep doesn't know about, we would be forced to start writing separate code, and at this time I'm not sure if that would be a recommended path forward. Please note that I am not trying to stay away from GNUStep, I just want to make sure that it doesn't interfere with the current framework and future developments/. cheers, - Koen. From schristley at mac.com Sun Sep 24 18:43:10 2006 From: schristley at mac.com (Scott Christley) Date: Sun, 24 Sep 2006 18:43:10 -0400 Subject: [Biococoa-dev] Re: GNUstep patch In-Reply-To: <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> Message-ID: I noticed that there were a number of source code files in the directories, but they were not in the Xcode project, for example BCFoundation/BCSequence/BCSequenceDNA.{h,m} Am I correct that these are the old files and version 2.0 are just the files? BCFoundation/BCSequence/BCAnnotation.{h,m} BCFoundation/BCSequence/BCSequence.{h,m} BCFoundation/BCSequence/BCSequenceArray.{h,m} BCFoundation/BCSequence/BCSequenceCodon.{h,m} Essentially I compile the source files that I saw in the Xcode project. These all compile fine, but I haven't tried to run a test program. I was thinking that I would try to get the tests running on Mac then see how they run on GNUstep. I'm not sure what changes there are that might not be supported by GNUstep. Scott On Sep 24, 2006, at 6:18 PM, Koen van der Drift wrote: > > On Sep 24, 2006, at 5:54 PM, Scott Christley wrote: > >> I didn't want to send this patch to everyone on the mailing list, >> so hopefully you can review and apply if it looks good. >> >> One file has been added to the Xcode project, BCInternal.h. >> Everything else should be makefiles for GNUstep, unfortunately the >> way the make system works under GNUstep, need a makefile in each >> directory that has source files. This isn't a complete port, just >> BCFoundation, but I will work on the other stuff soon. > > > Scott, > > Thanks for sending the patch files. Before we apply them, I would > acctually like to get some more input from the mailinglist and see > what opinion others have. This is a pretty big change in the > project files, and I don't want to just go ahead and apply it. > Furthermore, is GNUStep aware of all the changes in recent OS X > version, eg does it know about all the new classes in BCFoundation > and BCAppkit? If there are any classes that GNUStep doesn't know > about, we would be forced to start writing separate code, and at > this time I'm not sure if that would be a recommended path forward. > > Please note that I am not trying to stay away from GNUStep, I just > want to make sure that it doesn't interfere with the current > framework and future developments/. > > cheers, > > - Koen. From kvddrift at earthlink.net Sun Sep 24 19:07:05 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Sun, 24 Sep 2006 19:07:05 -0400 Subject: [Biococoa-dev] Re: GNUstep patch In-Reply-To: References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> Message-ID: <08637520-B096-435B-A372-A862DC2AE58F@earthlink.net> On Sep 24, 2006, at 6:43 PM, Scott Christley wrote: > > I noticed that there were a number of source code files in the > directories, but they were not in the Xcode project, for example > > BCFoundation/BCSequence/BCSequenceDNA.{h,m} Those files are all deprecated, we now only use BCSequence, and no subclasses. To id a sequence, we use BCSymbolSets (similar to alphabets in BioJava and BioPerl). > Essentially I compile the source files that I saw in the Xcode > project. These all compile fine, but I haven't tried to run a test > program. I was thinking that I would try to get the tests running > on Mac then see how they run on GNUstep. > > I'm not sure what changes there are that might not be supported by > GNUstep. I suggest that you make sure that everything build and runs fine in GNUStep before we apply any patches to the framework. Besides the tests, there are also some example programs in the project. - Koen. From schristley at mac.com Sun Sep 24 20:46:34 2006 From: schristley at mac.com (Scott Christley) Date: Sun, 24 Sep 2006 20:46:34 -0400 Subject: [Biococoa-dev] test question In-Reply-To: <08637520-B096-435B-A372-A862DC2AE58F@earthlink.net> References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> <08637520-B096-435B-A372-A862DC2AE58F@earthlink.net> Message-ID: Hello, So I'm looking at the tests, this test case fails: //initializer with string //using default symbol set //string composed only of unknown symbols - (void)testInitStringAllUnknown { NSString *initial,*expected,*obtained; BCSequence *sequence; initial= @"123458()%$#@!#"; expected=@""; sequence=[BCSequence sequenceWithString:initial symbolSet: [BCSymbolSet dnaSymbolSet]]; //sequence=[BCSequence sequenceWithString:initial usingType:BCSequenceTypeDNA]; obtained=[sequence sequenceString]; //all the errors are concatenated NSMutableString *error=[NSMutableString stringWithString:@""]; if (sequence==nil) [error appendString:@"Sequence is nil\n"]; if (![obtained isEqual:expected]) [error appendFormat:@"Sequence should be %@, but is %@\n", expected,obtained]; if ([sequence sequenceType]!=BCSequenceTypeDNA) [error appendFormat:@"Sequence should be DNA but is %d\n", [sequence sequenceType]]; if ([sequence symbolSet]!=[BCSymbolSet dnaSymbolSet]) [error appendFormat:@"Symbol set should be %@ but is %@\n", [BCSymbolSet dnaSymbolSet],[sequence symbolSet]]; if ([[sequence symbolSet] sequenceType]!=[sequence sequenceType]) [error appendFormat:@"The symbol set is of type %d but should be of type %d, the same type as the sequence\n", [[sequence symbolSet] sequenceType],[sequence sequenceType]]; //if error!=@"", the test failed STAssertTrue ( [error isEqualToString:@""],error); } Now it fails because obtained is nil instead of the empty string @"" I traced this down (pain to debug with the test framework but anyways) to the sequenceString method in BCSequence.m - (NSString*)sequenceString { unsigned int length = [self length]; if ( length ) return [self subSequenceStringInRange: NSMakeRange( 0, length ) ]; else return nil; } So I think it should return the empty string if length == 0, not nil. Sound right? That fixes the test but wanted to check if that is desired behavior. thanks Scott -------------- next part -------------- An HTML attachment was scrubbed... URL: From kvddrift at earthlink.net Sun Sep 24 21:01:20 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Sun, 24 Sep 2006 21:01:20 -0400 Subject: [Biococoa-dev] test question In-Reply-To: References: <3D523C2D-4D3D-4CBA-A050-EE0883445BB5@earthlink.net> <1C26CF48-7C05-412C-B304-D1E612E0027A@mac.com> <7E4E3B82-561A-4ADF-B490-AF5DB2078E84@earthlink.net> <08637520-B096-435B-A372-A862DC2AE58F@earthlink.net> Message-ID: On Sep 24, 2006, at 8:46 PM, Scott Christley wrote: > I traced this down (pain to debug with the test framework but > anyways) to the sequenceString method in BCSequence.m > > > > - (NSString*)sequenceString > { > unsigned int length = [self length]; > > if ( length ) > return [self subSequenceStringInRange: NSMakeRange( 0, length ) ]; > else > return nil; > } > > > > So I think it should return the empty string if length == 0, not > nil. Sound right? That fixes the test but wanted to check if that > is desired behavior. > Yep - that's correct. You can commit the change - do you already have access? Otherwise I will do it. Thanks for finding the bug. - Koen. From kvddrift at earthlink.net Mon Sep 25 21:18:38 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Mon, 25 Sep 2006 21:18:38 -0400 Subject: [Biococoa-dev] SenTesting Message-ID: Hi, Thanks to the effort of Scott, the Tests in the framework now compile and build without errors. However, there were some issues, besides fixing the source code. It turns out on my system I had two copies of SenTestingKit.framework, one in /Library/Frameworks, and one in / System/Library/Frameworks. The first one turns out to be a left over from my 10.3 installation, when the framework had to be added manually. The second one is included with 10.4. Just removing the framework from /Library/Frameworks isn't enough, because we want to be backward compatible with 10.3.9. I have tried several settings, to eventually get it to work, but I don't know if the project will still build under 10.3. Anyone has an idea how to set the project and target settings, so that the SenTestingKit can be used both in 10.3 and 10.4 ? Because the errors in the tet target are non-critical, I have committed the changes from Scott, so you can try out various settings. thanks. From charles.parnot at gmail.com Tue Sep 26 01:27:44 2006 From: charles.parnot at gmail.com (Charles Parnot) Date: Mon, 25 Sep 2006 22:27:44 -0700 Subject: [Biococoa-dev] SenTesting In-Reply-To: References: Message-ID: <852B63CA-4D1E-4449-8D23-D5AA9247E11C@gmail.com> The reason to build in 10.3 would be to test the framework in 10.3. If the tests do not compile as is, when the framework is in /Library/ Frameworks and you are running 10.3, then it seems the only option is to have a separate test target for 10.3 (yet another target!!!). An alternative is to say that testing with 10.4 should be enough, if the compatibility for 10.3 is maintained by sticking to 10.3-only symbols, and we don't start fiddling with #ifdef. In other words, the code to be compiled is the same in 10.3 and 10.4, and we can assume it will behave the same (of course, this is never 100% guaranteed if there are version-specific bugs in the Apple's frameworks that affect us). Bottom line: trying to maintain 10.3 compatible tests would probably be a pain in the butt. I would be curious however if there is some simple mechanism for that. charles On Sep 25, 2006, at 6:18 PM, Koen van der Drift wrote: > Hi, > > Thanks to the effort of Scott, the Tests in the framework now > compile and build without errors. However, there were some issues, > besides fixing the source code. > > It turns out on my system I had two copies of > SenTestingKit.framework, one in /Library/Frameworks, and one in / > System/Library/Frameworks. The first one turns out to be a left > over from my 10.3 installation, when the framework had to be added > manually. The second one is included with 10.4. Just removing the > framework from /Library/Frameworks isn't enough, because we want to > be backward compatible with 10.3.9. I have tried several settings, > to eventually get it to work, but I don't know if the project will > still build under 10.3. > > Anyone has an idea how to set the project and target settings, so > that the SenTestingKit can be used both in 10.3 and 10.4 ? > > > Because the errors in the tet target are non-critical, I have > committed the changes from Scott, so you can try out various settings. > > > thanks. > _______________________________________________ > Biococoa-dev mailing list > Biococoa-dev at bioinformatics.org > https://bioinformatics.org/mailman/listinfo/biococoa-dev -- Xgrid-at-Stanford Help science move fast forward: http://cmgm.stanford.edu/~cparnot/xgrid-stanford Charles Parnot charles.parnot at gmail.com From kvddrift at earthlink.net Tue Sep 26 08:53:26 2006 From: kvddrift at earthlink.net (Koen van der Drift) Date: Tue, 26 Sep 2006 08:53:26 -0400 (GMT-04:00) Subject: [Biococoa-dev] SenTesting Message-ID: <33000747.1159275207101.JavaMail.root@elwamui-hybrid.atl.sa.earthlink.net> >Bottom line: trying to maintain 10.3 compatible tests would probably >be a pain in the butt. > I agree Charles. I will adjust the settings of the target so it will build in 10.4. - Koen.