From bil at ccrma.Stanford.EDU Fri Aug 6 06:44:24 2010 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 6 Aug 2010 06:44:24 -0700 Subject: [CM] legolambda Message-ID: <20100806134334.M23142@ccrma.Stanford.EDU> (let* ((x 3) (arg '(x)) (body `((+ ,x x 1)))) ((apply lambda arg body) 12)) From taube at uiuc.edu Fri Aug 6 12:34:07 2010 From: taube at uiuc.edu (Heinrich Taube) Date: Fri, 6 Aug 2010 14:34:07 -0500 Subject: [CM] legolambda In-Reply-To: <20100806134334.M23142@ccrma.Stanford.EDU> References: <20100806134334.M23142@ccrma.Stanford.EDU> Message-ID: <6BAFBCCE-C3B2-4055-AF6B-18FAD4F54116@uiuc.edu> wow. applicable lambda builds a lambda that a lambda expr evaluates. sort of makes my head hurt! > (let* ((x 3) > (arg '(x)) > (body `((+ ,x x 1)))) > ((apply lambda arg body) 12)) > > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From bil at ccrma.Stanford.EDU Thu Aug 12 03:40:24 2010 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 12 Aug 2010 03:40:24 -0700 Subject: [CM] Snd 11.8 Message-ID: <20100812103935.M28494@ccrma.Stanford.EDU> Snd 11.8 --with-doubles is now the default in the configure script scan-channel and find-channel have changed slightly: they now return just the sample number where the match was found, or #f. *-button-color deprecated. index.cl and indexer.scm replaced by make-index.scm. s7: access to any symbol's dynamic binding via "special". Checked: gtk 2.90.5, sbcl 1.0.41, Ruby 1.9.1-p429, autoconf 2.67, cmucl 20b Thanks!: Dylan_, Mike Scholz, Rick Taube From plewto at gmail.com Mon Aug 16 03:09:46 2010 From: plewto at gmail.com (Steven Jones) Date: Mon, 16 Aug 2010 05:09:46 -0500 Subject: [CM] Problems with CM2 and SBCL Message-ID: I just installed Common Music 2 on Ubuto 10.04 and ran into problems with SBCL (version 1.0.29.11). Apparently sb-unix:unix-file-kind is not defined in the new version of SBCL. I ended up hacking cm.asd by replacing the line #+sbcl (eq :directory (sb-unix:unix-file-kind (namestring dir))) with #+sbcl (eq :directory (sb-impl::native-file-kind (namestring dir))) From dlphillips at woh.rr.com Tue Aug 17 03:35:52 2010 From: dlphillips at woh.rr.com (Dave Phillips) Date: Tue, 17 Aug 2010 06:35:52 -0400 Subject: [CM] article on athenaCL Message-ID: <4C6A6608.8000107@woh.rr.com> Greetings, Some of you might find this article interesting : http://www.linuxjournal.com/content/algorithmic-music-composition-linux-athenacl It's a unique system, very powerful. Best, dp From BBattey at dmu.ac.uk Wed Aug 25 12:54:36 2010 From: BBattey at dmu.ac.uk (Bret Battey) Date: Wed, 25 Aug 2010 20:54:36 +0100 Subject: [CM] Grace MIDI import / multiple message types References: <5EA1407A-3828-4227-8CFF-DD895D092BC1@uiuc.edu> Message-ID: <37DC19C2180900498561E9489A139796015B5854@ariadne.LEC-ADMIN.dmu.ac.uk> I'm in a situation where I am importing from a MIDI file. I will need to be able to access note messages and ctrl messages -- interleaved and in time order -- in order to perform my task. So (midifile-import midi-file 2 '(time key))) will get me the keynums and (midifile-import midi-file 2 '(time ctrl1 ctrl2))) will get me the controller messages But now I will have to code a sort mechanism to combine the two lists in time order. Is there a way to import the two message types directly into one list, enabling me to avoid having to code a sort? (midifile-import midi-file 2 '(time key ctrl1 ctrl2))) => Error: no midi message with values: (time key ctrl1 ctrl2) -------------- next part -------------- An HTML attachment was scrubbed... URL: From taube at uiuc.edu Wed Aug 25 14:06:00 2010 From: taube at uiuc.edu (Heinrich Taube) Date: Wed, 25 Aug 2010 16:06:00 -0500 Subject: [CM] Grace MIDI import / multiple message types In-Reply-To: <37DC19C2180900498561E9489A139796015B5854@ariadne.LEC-ADMIN.dmu.ac.uk> References: <5EA1407A-3828-4227-8CFF-DD895D092BC1@uiuc.edu> <37DC19C2180900498561E9489A139796015B5854@ariadne.LEC-ADMIN.dmu.ac.uk> Message-ID: <63A632D9-E51A-498F-A940-297A8A4489A4@uiuc.edu> bret first im sorry but witht the start of school i simply havnt had any free cycles this week to get the meta messages. i think maybe what ill should do is allow specs like ((time key) (time clt1 ctl2) ...) that will parse out differnet "value specs" all in one pass. will that suffice? again really sorry for the delay! i should have from free time starting friday morning, i dont think it will take me long to add this On Aug 25, 2010, at 2:54 PM, Bret Battey wrote: > I'm in a situation where I am importing from a MIDI file. I will > need to be able to access note messages and ctrl messages -- > interleaved and in time order -- in order to perform my task. > > So > > (midifile-import midi-file 2 '(time key))) > > will get me the keynums > > and > > (midifile-import midi-file 2 '(time ctrl1 ctrl2))) > > will get me the controller messages > > But now I will have to code a sort mechanism to combine the two > lists in time order. > > Is there a way to import the two message types directly into one > list, enabling me to avoid having to code a sort? > > (midifile-import midi-file 2 '(time key ctrl1 ctrl2))) => Error: no > midi message with values: (time key ctrl1 ctrl2) > > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From BBattey at dmu.ac.uk Thu Aug 26 04:34:14 2010 From: BBattey at dmu.ac.uk (Bret Battey) Date: Thu, 26 Aug 2010 12:34:14 +0100 Subject: [CM] Grace MIDI import / multiple message types References: <5EA1407A-3828-4227-8CFF-DD895D092BC1@uiuc.edu> <37DC19C2180900498561E9489A139796015B5854@ariadne.LEC-ADMIN.dmu.ac.uk> <63A632D9-E51A-498F-A940-297A8A4489A4@uiuc.edu> Message-ID: <37DC19C2180900498561E9489A139796015B5855@ariadne.LEC-ADMIN.dmu.ac.uk> >> i think maybe what ill should do is allow specs like >> ((time key) (time clt1 ctl2) ...) Yes -- I think that would work. The trick seems to be returning a list that allows one to tell the different message types apart. So, in this example, we might get something like (?): ( ((0.0 60) ()) ;; a note at time 0 ((1.2 62) ()) ;; a note at time 1.2 (() (1.5 1 64)) ;; controller 1, value 64, at time 1.5 ((1.8 64) (1.8 1 68)) ;; note at time 1.8, controller at time 1.8 ;; (or one could have two distinct entries for each event) ) This format could also be helpful because one could order the spec based on the time-priority logic one needs. That is, if I have events simultaneous in time but logic-wise need to process (say) controller values 1st, I would just place the controller first in the spec. It is interesting that the Grace MIDI file import can import the meta messages. Does that mean there is an undocumented keyword already that allows one to import meta messages via midifile-import? ('meta' doesn't do it ;) Thanks, -=Bret -------------- next part -------------- An HTML attachment was scrubbed... URL: From taube at uiuc.edu Thu Aug 26 05:06:30 2010 From: taube at uiuc.edu (Heinrich Taube) Date: Thu, 26 Aug 2010 07:06:30 -0500 Subject: [CM] Grace MIDI import / multiple message types In-Reply-To: <37DC19C2180900498561E9489A139796015B5855@ariadne.LEC-ADMIN.dmu.ac.uk> References: <5EA1407A-3828-4227-8CFF-DD895D092BC1@uiuc.edu> <37DC19C2180900498561E9489A139796015B5854@ariadne.LEC-ADMIN.dmu.ac.uk> <63A632D9-E51A-498F-A940-297A8A4489A4@uiuc.edu> <37DC19C2180900498561E9489A139796015B5855@ariadne.LEC-ADMIN.dmu.ac.uk> Message-ID: yes telling message apart can be a problem and there is no object system to make it easy. what you suggest is an interesting approach. -- another possiblility : i could allow you to import the "opcode" of the message as a tag: (op time cltl1 ctl2) (op time key) -- another possibility: call the current function 'parse-midifile' and then implment a 'import-midifile' that bring everything in, including time and opcode On Aug 26, 2010, at 6:34 AM, Bret Battey wrote: > >> i think maybe what ill should do is allow specs like > >> ((time key) (time clt1 ctl2) ...) > > Yes -- I think that would work. The trick seems to be returning a > list that allows one to tell the different message types apart. So, > in this example, we might get something like (?): > > ( > ((0.0 60) ()) ;; a note at time 0 > ((1.2 62) ()) ;; a note at time 1.2 > (() (1.5 1 64)) ;; controller 1, value 64, at time 1.5 > ((1.8 64) (1.8 1 68)) ;; note at time 1.8, controller at time 1.8 > ;; (or one could have two distinct entries > for each event) > ) > > This format could also be helpful because one could order the spec > based on the time-priority logic one needs. That is, if I have > events simultaneous in time but logic-wise need to process (say) > controller values 1st, I would just place the controller first in > the spec. > > It is interesting that the Grace MIDI file import can import the > meta messages. Does that mean there is an undocumented keyword > already that allows one to import meta messages via midifile-import? > ('meta' doesn't do it ;) > > Thanks, > -=Bret > From gkjoyce at gmail.com Tue Aug 31 16:53:04 2010 From: gkjoyce at gmail.com (Gregory Joyce) Date: Tue, 31 Aug 2010 16:53:04 -0700 Subject: [CM] Compilation problems Message-ID: So using the version off of SVN I got this error on my first "make" src/CmSupport.cpp: In function ?void cm_print_markov_table(s7_cell*, s7_cell*, int, int)?: src/CmSupport.cpp:86:63: error: too few arguments to function ?s7_cell* s7_object_to_string(s7_scheme*, s7_cell*, bool)? /usr/include/s7.h:137:12: note: declared here make[1]: *** [obj/cm/Release/CmSupport.o] Error 1 make: *** [cm] Error 2 I put ",true" into that statement and that got rid of that error. I have no idea whether that should be "true" or not. When Grace tries to build I get this: In file included from src/Main.cpp:37:0: src/Skin.h: In member function ?virtual void WindowsSkin::drawPopupMenuBackground(juce::Graphics&, int, int)?: src/Skin.h:52:47: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:52:47: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:53:49: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:53:49: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h: In member function ?virtual void WindowsSkin::drawPopupMenuItem(juce::Graphics&, int, int, bool, bool, bool, bool, bool, const juce::String&, const juce::String&, juce::Image*, const juce::Colour*)?: src/Skin.h:69:45: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:69:45: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:84:66: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:84:66: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:85:43: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:85:43: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:89:49: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:89:49: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h: In member function ?virtual void WindowsSkin::drawMenuBarItem(juce::Graphics&, int, int, int, const juce::String&, bool, bool, bool, juce::MenuBarComponent&)?: src/Skin.h:204:51: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:204:51: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:207:51: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:207:51: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:214:62: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:214:62: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:215:39: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:215:39: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h: In member function ?virtual void WindowsSkin::drawMenuBarBackground(juce::Graphics&, int, int, bool, juce::MenuBarComponent&)?: src/Skin.h:230:47: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:230:47: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:232:49: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:232:49: error: for a function-style cast, remove the redundant ?::Colour? src/Skin.h:235:49: error: cannot call constructor ?juce::Colour::Colour? directly src/Skin.h:235:49: error: for a function-style cast, remove the redundant ?::Colour? make[1]: *** [obj/grace/Release/Main.o] Error 1 make: *** [grace] Error 2 Which I have no idea about. Any thoughts? Thanks Greg