usablity of pure
Bjoern Lindig
I just felt like wanting to give a feedback: I rote an application, that
is ought to help me enter Data into a database and plot it. I used
pure-gplot and pure-sql3. It works amazingly well and was really easy
to programm. Even in bash, which I know quite well by now, this would
have taken me double the time then it did in pure. Maybe this is also
because the documentation is so good. Anyway, I'm convinced now, pure is
great!
Thank you all for all your work!
Best,
bjoern
Albert Graef
> Thank you all for all your work!
Glad that it works for you. :) Did you try pd-pure and pd-faust yet?
Albert
--
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email: Dr.G...@t-online.de, a...@muwiinfa.geschichte.uni-mainz.de
WWW: http://www.musikinformatik.uni-mainz.de/ag
Bjoern Lindig
> Glad that it works for you. :) Did you try pd-pure and pd-faust yet?
Yeah, I did. They work fine. I'm digging into it right now. Thanks for
reminding me.
BTW, I enjoy your examples very much! You have some very beautiful
midifiles there. I remember the old days, when I hear them (we were a
little oldschool those days, but the bouree is actually one reason why I
learned Guitar and Flute). I assume it is quite obvious when you start
searching, but where did you get them?
Your instrumentation is also very nice! I still need some time to
figure out how all of that is working, but I think it won't be long
until I switch over to pure/faust/pd.
Oh, that reminds me: You once told me that you are working on a
pure interface for SuperCollider. What happend? Are you still on it?
Best,
Bjoern
Albert Graef
> BTW, I enjoy your examples very much! You have some very beautiful
> midifiles there. I remember the old days, when I hear them (we were a
> little oldschool those days, but the bouree is actually one reason why I
> learned Guitar and Flute). I assume it is quite obvious when you start
> searching, but where did you get them?
Google is your friend. :) I think that I got the bouree.mid included
with pd-faust here: http://www.free-midi.org/song/jethro_tull-bouree.html
> Your instrumentation is also very nice! I still need some time to
> figure out how all of that is working, but I think it won't be long
> until I switch over to pure/faust/pd.
If you run into any problems, just ask. I can also send you a copy of my
pd-faust paper for LAC 2012 (I don't think that they have the papers
online yet), just let me know. And of course there will be live
streaming so you can watch my LAC presentation on Fri the 13th. :)
> Oh, that reminds me: You once told me that you are working on a
> pure interface for SuperCollider. What happend? Are you still on it?
Well, there's a minimal example in pure-liblo/examples/sc3test.pure. If
you can get that to work, it shouldn't be that hard to build a full Pure
API based on that. I also have a bunch of older Q-SC3 examples, like the
stuff on display at
http://q-lang.sourceforge.net/examples.html#Multimedia, but I didn't get
to porting them over to Pure yet.
I will eventually get back to this, but the next audio-related thing on
my TODO list is a pure-lv2 plugin so that we can start running
algorithmic composition stuff written in Pure right inside Ardour and
Qtractor, and then maybe something like pd-faust but for LV2. So it
might still take a little while until I have time to work on the
SuperCollider interface again, sorry. The day doesn't have enough hours,
so if you can port some of the SC-related Q stuff over to Pure, or do
your own interface, that would be very welcome!
Bjoern Lindig
> If you run into any problems, just ask. I can also send you a copy of
> my pd-faust paper for LAC 2012 (I don't think that they have the
> papers online yet), just let me know. And of course there will be
> live streaming so you can watch my LAC presentation on Fri the
> 13th. :)
Thanks for the reminder. I'll be online at 10:45 p.m., if calculation
is correct. I would love to read you paper in advance. I learned that
beeing prepared help a lot.
>
> Well, there's a minimal example in pure-liblo/examples/sc3test.pure.
> If you can get that to work, it shouldn't be that hard to build a
> full Pure API based on that. I also have a bunch of older Q-SC3
Yes, it works. Looks cool. Maybe I can do something here.
> examples, like the stuff on display at
> http://q-lang.sourceforge.net/examples.html#Multimedia, but I didn't
> get to porting them over to Pure yet.
Ok, lets see whether and when I can have a look at that. I'm a little
scared of Q, just because it is yet another foreign language (yafl).
But a port should not be too hard to do.
>
> I will eventually get back to this, but the next audio-related thing
> on my TODO list is a pure-lv2 plugin so that we can start running
> algorithmic composition stuff written in Pure right inside Ardour and
> Qtractor, and then maybe something like pd-faust but for LV2. So it
> might still take a little while until I have time to work on the
> SuperCollider interface again, sorry. The day doesn't have enough
> hours, so if you can port some of the SC-related Q stuff over to
> Pure, or do your own interface, that would be very welcome!
The lv2 stuff sounds amazing! I'll do my best to give you some backup.
Best,
Bjoern
Bjoern Lindig
Oh, Q is not ported to x86-64 yet. Is there a chance to build it
anyway? This error seems to be related to libtools. I have installed
libltdl 2.4.2-4, gcc is version 4.6.3.
(cd .libs && gcc -march=x86-64 -mtune=generic -O2 -pipe -c
-fno-builtin "qS.c") rm -f .libs/qS.c .libs/q.nm .libs/q.nmS .libs/q.nmT
gcc -DYEAR=\"2008\" -DSYSINFO=\"x86_64-unknown-linux-gnu\"
-DQPATH=\".:/usr/share/q/lib:/usr/lib/q\" -DQEXEC=\"/usr/bin/q\"
-DLIBTOOL=\"/usr/lib/q/libtool\" -DCC=\"gcc\" -march=x86-64
-mtune=generic -O2 -pipe .libs/qS.o -Wl,--hash-style=gnu
-Wl,--as-needed -O3 -o .libs/q q-q.o q-qm.o q-qmcode.o q-qmfuns.o
q-qmmatch.o q-qmprint.o q-qmparse.o q-qmlex.o q-qbase.o q-sys.o
q-random.o q-uchar.o q-getopt.o q-getopt1.o
-Wl,--export-dynamic ../libq/.libs/libq.so -lfl -lgmp -lltdl -lpthread
-lreadline -lncursesw -lcrypt -lrt -lutil -lnsl -lm q-q.o: In function
`main': q.c:(.text.startup+0x47): undefined reference to
`lt__PROGRAM__LTX_preloaded_symbols' collect2: ld returned 1 exit
status make[3]: *** [q] Error 1
Best,
Bjoern
Albert Graef
> Thanks for the reminder. I'll be online at 10:45 p.m., if calculation
> is correct. I would love to read you paper in advance. I learned that
> beeing prepared help a lot.
You have mail.
> Ok, lets see whether and when I can have a look at that. I'm a little
> scared of Q, just because it is yet another foreign language (yafl).
> But a port should not be too hard to do.
Well, that was just a suggestion. Actually, I'd be more interested in
stuff that you can come up with on your own. ;-)
Albert Graef
> Oh, Q is not ported to x86-64 yet. Is there a chance to build it
> anyway?
Nope. It's 32 bit only. I never bothered porting it to 64 bit, since I
made Pure instead.
John Cowan
> Nope. It's 32 bit only. I never bothered porting it to 64 bit, since I
> made Pure instead.
Q can be built on a 64-bit userland by adding -m32 to the CFLAGS and
LDFLAGS. Admittedly I haven't done this recently.
--
Henry S. Thompson said, / "Syntactic, structural, John Cowan
Value constraints we / Express on the fly." co...@ccil.org
Simon St. Laurent: "Your / Incomprehensible http://www.ccil.org/~cowan
Abracadabralike / schemas must die!"
Albert Graef
> Albert Graef scripsit:
>
>> Nope. It's 32 bit only. I never bothered porting it to 64 bit, since I
>> made Pure instead.
>
> Q can be built on a 64-bit userland by adding -m32 to the CFLAGS and
> LDFLAGS. Admittedly I haven't done this recently.
I must admit that I never tried this myself, but it might even work. The
Q sources haven't been updated for quite some time, though, so you'll
probably have to work around some bit rot. ;-) Also, the Q-Midi module
uses Grame's MidiShare (rather than PortMidi which is what pure-midi
uses). Alas, MidiShare probably won't run on 64 bit systems at all
because it includes a Linux kernel module which was never ported to
x86_64 either, AFAIK.
Ryan Schmidt
> On 03/08/2012 02:35 AM, John Cowan wrote:
>> Albert Graef scripsit:
>>
>>> Nope. It's 32 bit only. I never bothered porting it to 64 bit, since I
>>> made Pure instead.
>>
>> Q can be built on a 64-bit userland by adding -m32 to the CFLAGS and
>> LDFLAGS. Admittedly I haven't done this recently.
>
> I must admit that I never tried this myself, but it might even work.
FWIW, most Mac users who have been using Q in the past 2.5 years have been running Q 64-bit.
This is because most Intel Macs (all but the first generation released in 2005) are 64-bit. As of Snow Leopard (released in 2009) the default compiler architecture is x86_64 (on 64-bit Macs). So most people compiling Q on a Mac in the past 2.5 years would get a 64-bit version.
Q is available in MacPorts. I have received no bug reports about any problems relating to it being 64-bit (or, I must admit, about any other problem; no tickets have ever been filed against Q in MacPorts).
So either Mac users have been your guinea pigs over the past few years and have shown that Q works 64-bit, or nobody is using Q anymore. :)
John Cowan
> So either Mac users have been your guinea pigs over the past few years
> and have shown that Q works 64-bit, or nobody is using Q anymore. :)
It definitely does not. I tried recompiling Q after my (then) employer
switched from 32-bit userland to 64-bit userland on a 64-bit Linux kernel.
It may have compiled, but it definitely didn't run, so I judiciously
added -m32 and all was well.
I suggest trying to get Q removed from MacPorts (and Fink, if it's there)
altogether. It's not supported upstream (i.e. by Albert) any more.
And the same advice to those who have the ear of other repository-keepers,
like various Linux distros.
--
So they play that [tune] on John Cowan
their fascist banjos, eh? co...@ccil.org
--Great-Souled Sam http://www.ccil.org/~cowan
Ryan Schmidt
On Mar 8, 2012, at 12:56, John Cowan wrote:
> Ryan Schmidt scripsit:
>
>> So either Mac users have been your guinea pigs over the past few years
>> and have shown that Q works 64-bit, or nobody is using Q anymore. :)
>
> It definitely does not. I tried recompiling Q after my (then) employer
> switched from 32-bit userland to 64-bit userland on a 64-bit Linux kernel.
> It may have compiled, but it definitely didn't run, so I judiciously
> added -m32 and all was well.
That's certainly possible; I don't know how to use Q so I haven't tried.
If Q only works 32-bit we could modify Q in MacPorts to force it to compile 32-bit.
> I suggest trying to get Q removed from MacPorts (and Fink, if it's there)
> altogether. It's not supported upstream (i.e. by Albert) any more.
> And the same advice to those who have the ear of other repository-keepers,
> like various Linux distros.
I am happy to remove Q from MacPorts, if the software is dead and that is what Albert desires. But, the Q web site makes no statement about Q being deprecated, unsupported, obsolete, discontinued, or similar. What it says is "Q is still being maintained, so please continue to report bugs and send patches." As such I would assume Q should continue to remain available in MacPorts and other package managers.
Albert Graef
> I suggest trying to get Q removed from MacPorts (and Fink, if it's there)
> altogether. It's not supported upstream (i.e. by Albert) any more.
> And the same advice to those who have the ear of other repository-keepers,
> like various Linux distros.
I think I have to agree with this. I just don't have the time to support
two such projects at the same time, and Pure has been dubbed as Q's
successor for quite some time now. There's still a fairly complete
collection of Windows packages on the Q website for those who still want
to run it, and BSD/Linux/OSX users can always help themselves with the
sources if they must.
Fink never had Q. The Fedora and SUSE Q packages have long bitten the
dust as well. Concerning MacPorts, if the 32 bit version still compiles,
it should also work. But John is right, it definitely won't work on 64
bit; we've had various reports about this on Q ml back in the day
(mostly Linux-related, but I don't see how it could be different on 64
bit OSX).
Fortunately, most of the interesting Q stuff has been ported to Pure
already, or we have better replacements (such as Peter's SQLite module).
It's just some of the multimedia examples that haven't been ported yet,
and the SmokeQt interface of course (but we have pure-gtk and
pure-tk+Gnocl instead). I still have a SmokeQt4 module on my TODO list,
but this is probably going to be quite amount of work, and I'm not sure
if it's really worth the effort. It might be a better idea, as Eddie
suggested, to make pure-gtk work with GTK3 once that's reasonably stable.
Albert
Albert Graef
> That's certainly possible; I don't know how to use Q so I haven't tried.
Try 'make test' after compiling. You can also just run the 'q' command.
IIRC, it's printing lots of errors on 64 bit. If it's working it should
just print the sign-on message and display its prompt.
> If Q only works 32-bit we could modify Q in MacPorts to force it to compile 32-bit.
That would be an option to keep it available for those who still want to
use it. But feel free to just remove it.
> But, the Q web site makes no statement about Q being deprecated, unsupported, obsolete, discontinued, or similar. What it says is "Q is still being maintained, so please continue to report bugs and send patches."
Well, that's old news from 2008 when Pure still was in its infancy. I
should probably update this some time. ;-)
Albert
John Cowan
> Well, that's old news from 2008 when Pure still was in its infancy. I
> should probably update this some time. ;-)
I'd just replace the "Welcome!" banner with "Q is no longer supported, see
Pure" in the same size of type.
--
He made the Legislature meet at one-horse John Cowan
tank-towns out in the alfalfa belt, so that co...@ccil.org
hardly nobody could get there and most of http://www.ccil.org/~cowan
the leaders would stay home and let him go --H.L. Mencken's
to work and do things as he pleased. Declaration of Independence
Albert Graef
> I'd just replace the "Welcome!" banner with "Q is no longer supported, see
> Pure" in the same size of type.
Yeah, I should really do this. Seems that people are still interested
after all this time, considering that Q has had 70 downloads this week
alone. ;-)
Albert Graef
> On 03/08/2012 10:22 PM, John Cowan wrote:
>> I'd just replace the "Welcome!" banner with "Q is no longer supported,
>> see Pure" in the same size of type.
>
> Yeah, I should really do this. Seems that people are still interested
> after all this time, considering that Q has had 70 downloads this week
> alone. ;-)
Done.
Ryan Schmidt
On Mar 8, 2012, at 15:03, Albert Graef wrote:
> On 03/08/2012 09:04 PM, Ryan Schmidt wrote:
>> That's certainly possible; I don't know how to use Q so I haven't tried.
>
> Try 'make test' after compiling.
Thanks; it says:
def: error loading module
def: error loading module
Warning: 312 unresolved external symbols
! File def, line 77: Value mismatch in definition
! File def, line 268: Value mismatch in definition
> You can also just run the 'q' command.
It says:
def: error loading module
Warning: 141 unresolved external symbols
____
/ __ \ Q interpreter version 7.11 (i386-apple-darwin10.8.0)
/ /_/ / Copyright (c) 1991-2008 by Albert Graef
\___\_\ http://q-lang.sourceforge.net
This software is distributed under the terms of the GNU General Public
License version 2 or later; type `copying' for details.
! File def, line 77: Value mismatch in definition
==>
> IIRC, it's printing lots of errors on 64 bit. If it's working it should just print the sign-on message and display its prompt.
>
>> If Q only works 32-bit we could modify Q in MacPorts to force it to compile 32-bit.
>
> That would be an option to keep it available for those who still want to use it. But feel free to just remove it.
When compiled for i386 the tests pass:
Testing Q 7.11 (i386-apple-darwin10.8.0)
TEST: local variables......... passed
TEST: arithmetic.............. passed
TEST: bignums................. passed
TEST: rational numbers........ passed
TEST: complex numbers......... passed
TEST: logical................. passed
TEST: conditionals............ passed
TEST: strings................. passed
TEST: lists................... passed
TEST: tuples.................. passed
TEST: streams................. passed
TEST: lambda.................. passed
TEST: list comprehensions..... passed
TEST: quicksort............... passed
TEST: random numbers.......... passed
TEST: prime test.............. passed
TEST: quadratic residues...... passed
TEST: factorization........... passed
TEST: regular expressions..... passed
preparing POSIX threads tests... done
TEST: thread creation......... passed
TEST: thread cancellation..... passed
TEST: mutexes................. passed
TEST: conditions.............. passed
TEST: semaphores.............. passed
*** ALL TESTS HAVE BEEN PASSED ***
And "q" runs without complaint:
____
/ __ \ Q interpreter version 7.11 (i386-apple-darwin10.8.0)
/ /_/ / Copyright (c) 1991-2008 by Albert Graef
\___\_\ http://q-lang.sourceforge.net
This software is distributed under the terms of the GNU General Public
License version 2 or later; type `copying' for details.
==>
>> But, the Q web site makes no statement about Q being deprecated, unsupported, obsolete, discontinued, or similar. What it says is "Q is still being maintained, so please continue to report bugs and send patches."
>
> Well, that's old news from 2008 when Pure still was in its infancy. I should probably update this some time. ;-)
I've updated the q in MacPorts to force a 32-bit compile, and to display a note recommending Pure instead.
Albert Graef
> I've updated the q in MacPorts to force a 32-bit compile, and to display a note recommending Pure instead.
Thanks!