[L2Ork-dev] disis_munger~ in last purr-data release?

Fri Mar 2 11:22:41 EST 2018

Thank you for the clarification, Albert. What I am confused by is that the
old version of the build system was capable of doing this and working
around this problem as it could generate the final binary without any
interruptions. Further, the flext build system has not changed in ages and
I doubt it ever ever will so if we create a workaround for it it should
work for the foreseeable time in the future. What is not clear to me is
whether the autobuild systems are susceptible to the interruptions flext
build system generates unlike the automated script I put together.

-- 
Ivica Ico Bukvic, D.M.A.
Creative Technologies in Music
Director -- DISIS, L2Ork, CTM
ICAT Senior Fellow
Virginia Tech
School of Performing Arts – 0141
Blacksburg, VA 24061
(540) 231-6139
ico at vt.edu
www.performingarts.vt.edu
disis.icat.vt.edu
l2ork.icat.vt.edu
ico.bukvic.net

On Mar 2, 2018 11:04, "Albert Graef" <aggraef at gmail.com> wrote:

> Ico, I think that what Jonathan tried to say (maybe he should have been
> more blunt about it), is that the existing flext build system just can't be
> used any more without breaking the continuous integration stuff he has in
> place. The CI is really important. It does all builds automatically at all
> times so that we can *always* be sure that, whatever goes in, also builds
> on all the supported platforms. It took a while to get it there, so that's
> not something to be toyed with.
>
> If we want flext in there, its build system needs to be fixed up so that
> it builds on first attempt, without any horrible kludges and manual
> intervention. Simple as that. So someone needs to put the effort in. Not me
> either. :)
>
> Albert
>
>
> On Fri, Mar 2, 2018 at 3:32 PM, Ivica Ico Bukvic <ico at vt.edu> wrote:
>
>> Thank you, Jonathan. Would it be possible to simply borrow then 1.0 build
>> scripts for flext or are there other issues pertaining to Windows and/or
>> OSX?
>>
>>
>>
>> On 3/2/2018 12:18 AM, Jonathan Wilkes wrote:
>>
>>> On Thu, Mar 1, 2018 at 11:24 PM, Ivica Ico Bukvic <ico at vt.edu> wrote:
>>>
>>>> I understand that flext not building easily or at all using the build
>>>> environment Jonathan set up is an issue. I am, however, reluctant on
>>>> letting
>>>> go of flext as there are other externals it offers like pyext that can
>>>> be
>>>> particularly useful. Is there anyone in the community who may be
>>>> willing to
>>>> help with trying to build flext using Jonathan's build tools? Jonathan,
>>>> is
>>>> there a page that describes what these tools are exactly and what were
>>>> the
>>>> issues that prevented flext from building, so that we can take it from
>>>> there?
>>>>
>>> The build tools are exactly the same as the ones in Pd-l2ork 1.0-- a
>>> set of recursive
>>> makefiles which call external library makefiles as specified in
>>> externals/Makefile.
>>>
>>> I vaguely remember it not building under msys2 which is the environment
>>> used
>>> for building Purr Data under Windows. I do remember noticing
>>> that flext uses a set of custom build scripts and has its own package
>>> file
>>> format.
>>>
>>> Then I noticed that the main build script for flext isn't idempotent.
>>> You have to run
>>> it once, manually edit a file that gets created, then run it again.
>>>
>>> Finally, I noticed that the recipe in externals/Makefile for building
>>> flext under
>>> Pd extended actually ran the build script THREE times instead of two.
>>>
>>> With that in mind, my strong suggestion is for someone to improve flext's
>>> build system upstream and make it use one of the several standard build
>>> systems instead of a set of custom scripts which are difficult to reason
>>> about.
>>> In this case we could easily ship it with Purr Data and easily maintain
>>> it.
>>>
>>> If that isn't possible, I'll happily accept merge requests for
>>> building/shipping
>>> flext as long as the submitter agrees to maintain it going forward
>>> instead of me.
>>>
>>> -Jonathan
>>>
>>> Thank you.
>>>>
>>>> Best,
>>>>
>>>> Ico
>>>>
>>>>
>>>>
>>>> On 2/28/2018 6:24 PM, Jonathan Wilkes wrote:
>>>>
>>>>> Hm... what is the point of variables like grate_var_connected? It
>>>>> appears to get set but not used, unless I'm missing some C++/Flext
>>>>> magic...
>>>>>
>>>>> Also, what's the point of tempgrate vs. using a function local tmp
>>>>> variable?
>>>>>
>>>>> If none of those are needed it looks like I can just bind some
>>>>> ancillary float inlets to relevant struct members.
>>>>>
>>>>> -Jonathan
>>>>>
>>>>> On Wed, Feb 28, 2018 at 5:27 PM, Jonathan Wilkes <
>>>>> jon.w.wilkes at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> If someone can make a wrapper for the STK C++ library calls I can just
>>>>>> try porting it to C.
>>>>>>
>>>>>> -Jonathan
>>>>>>
>>>>>> On Wed, Feb 28, 2018 at 3:14 PM, Alberto Zin <alberto.zin at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Compiling disis_munger~ is not for the faint-hearted, (it requires
>>>>>>> flext
>>>>>>> build system in place), so I hope it will be built automatically
>>>>>>> again..
>>>>>>>
>>>>>>> Alberto
>>>>>>>
>>>>>>> 2018-02-28 14:38 GMT+01:00 Albert Graef <aggraef at gmail.com>:
>>>>>>>
>>>>>>>> It's in https://git.purrdata.net/jwilkes/purr-data/commit/e72deb8d,
>>>>>>>> and
>>>>>>>> it
>>>>>>>> says right in the commit message that disis_munger was excluded, but
>>>>>>>> not
>>>>>>>> why. :)
>>>>>>>>
>>>>>>>> On Wed, Feb 28, 2018 at 2:35 PM, Albert Graef <aggraef at gmail.com>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> Yep, it's not being built. Which might be due to the fact that it
>>>>>>>>> still
>>>>>>>>> sits in its own subdir in l2ork_addons/, whereas the other DISIS
>>>>>>>>> externals
>>>>>>>>> were moved to externals/disis/ a while back. Jonathan, maybe you
>>>>>>>>> missed that
>>>>>>>>> one? Or maybe it was just never being built? Let me try to find
>>>>>>>>> that
>>>>>>>>> commit...
>>>>>>>>>
>>>>>>>>> On Wed, Feb 28, 2018 at 12:24 PM, Alberto Zin <
>>>>>>>>> alberto.zin at gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> Hello, I use Albert's repository for getting and update purr-data
>>>>>>>>>> on
>>>>>>>>>> Linux mint 18.03 (based on Ubuntu 16.04).  is it just me or
>>>>>>>>>> disis_munger~.
>>>>>>>>>> pd_linux is missing/not compiled?
>>>>>>>>>> Thanks, Alberto.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> L2Ork-dev mailing list
>>>>>>>>>> L2Ork-dev at disis.music.vt.edu
>>>>>>>>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Dr. Albert Gr"af
>>>>>>>>> Computer Music Research Group, JGU Mainz, Germany
>>>>>>>>> Email:  aggraef at gmail.com
>>>>>>>>> WWW:    https://plus.google.com/+AlbertGraef
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Dr. Albert Gr"af
>>>>>>>> Computer Music Research Group, JGU Mainz, Germany
>>>>>>>> Email:  aggraef at gmail.com
>>>>>>>> WWW:    https://plus.google.com/+AlbertGraef
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> L2Ork-dev mailing list
>>>>>>>> L2Ork-dev at disis.music.vt.edu
>>>>>>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> https://soundcloud.com/albertoz
>>>>>>> https://www.flickr.com/photos/94787902@N08/
>>>>>>> http://sites.google.com/site/albertozin/
>>>>>>>
>>>>>>> 47 41 4d 45 20 4f 56 45 52
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> L2Ork-dev mailing list
>>>>>>> L2Ork-dev at disis.music.vt.edu
>>>>>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>>>>>
>>>>>> _______________________________________________
>>>>> L2Ork-dev mailing list
>>>>> L2Ork-dev at disis.music.vt.edu
>>>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>>>
>>>>
>>>> _______________________________________________
>>>> L2Ork-dev mailing list
>>>> L2Ork-dev at disis.music.vt.edu
>>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>>
>>> _______________________________________________
>>> L2Ork-dev mailing list
>>> L2Ork-dev at disis.music.vt.edu
>>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>>
>>
>> _______________________________________________
>> L2Ork-dev mailing list
>> L2Ork-dev at disis.music.vt.edu
>> https://disis.music.vt.edu/listinfo/l2ork-dev
>>
>
>
>
> --
> Dr. Albert Gr"af
> Computer Music Research Group, JGU Mainz, Germany
> Email:  aggraef at gmail.com
> WWW:    https://plus.google.com/+AlbertGraef
>
> _______________________________________________
> L2Ork-dev mailing list
> L2Ork-dev at disis.music.vt.edu
> https://disis.music.vt.edu/listinfo/l2ork-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://disis.music.vt.edu/pipermail/l2ork-dev/attachments/20180302/6873cee8/attachment.html>