There have been many changed since 6.12 and we are getting close to a
new release.  If anyone has an issue that must be resolved before 6.13
or other reasons now is the time to speak.  Draft release notes
below.  Again any corrections or improvements welcome.

==John ffitch
========================================================================


========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
# CSOUND VERSION 6.13 RELEASE NOTES

Not many new opcodes but there are a significant number of opcodes being
extended to use arrays in a variety of ways, widening the options for
users.  There have been many fixes to the core code as well as opcodes.
-- The Developers

## USER-LEVEL CHANGES

### New opcodes

- string2array is a variant of fillarray with he data coming from a
  string of space separated values.

- nstrstr returns the name string of an instrument number or an empty
  string if the number does not refer to a named instrument.

- ntof converts notename to frequency at i- and k-time.

### New Gen and Macros

- ampmidicurve is a new opcode that maps an input MIDI velocity number to an
  output gain factor with a maximum value of 1, modifying the output gain by
  a dynamic range and a shaping exponent.

### Orchestra

- The consistency of kr, sr and ksmps reworked especially when there
  is overriding.

- corrected default 0bdfs usage.

- Resolving the path for #include files reworked to be more liberal.

- reading and writing to multidimensional arrays was very wrong.  This
  is now correct.

- Better checking for unknown array types (issue #1124)

- In all array operations the size of an array is determined at init time and no 
  allocation happens at perf time.
  
- array arithmetic now respects --sample-accurate.

### Score

- The characters n and m could erroneously get ignored in scores.

- Resolving the path for #include files reworked to be more liberal.

- After an error the backtrace of files and macros incorrectly read
  the information for orchestra rather than scores; fixed.

- The end of an r (repeated) section was not always correct.

- Nested {} parts of a score could lead to errors.

- After an s statement a newline was required; no longer needed.

- The {} score loops have been reworked to allow macros and expressions
  in the loop count.
  
### Options

- The new option --use-system-sr st the sample rate to the hardware/system value.

### Modified Opcodes and Gens

- Sending failure in OSC is now a warning rather than an error.

- passign can now have an array as the target.

- version of bpf/bpfcos added to allow points defined via arrays.

- grain can now use tables of any size, which was only a power of 2.

- Changing colours in FL widgets now works (it previously did not redraw
  the colour).

- fillarray can read from a file of values overcoming the argument
  limit.

- sumarray now works for audio arrays as well as for scalar values.

- assignment of an audio value to an audio array now works.

- monitor was broken in the array form.

- gendyc now respects sample-accurate mode.

- mtof and ftom now have array versions.

- sc_lag and sc_lagud now use the first k- or a-rate input when no
  initial value is given.

- printarray now works for string arrays.

- changed2 now works for strings.

- diskgrain, syncgrain and syncloop now can do sample rate scaling.

- GEN01 correctly reads raw audio files when requested.

- ftaudio can now take two additional optional argument`s for the start
  and end of the table data being written to file.

- sensekey recoded in the 'key down' mode.

- loscilx can return an audio array.

- schedule opcode reports undefined instruments in all cases.

- event_i now accepts tagged instrument numbers.

- printarray treats %d correctly.

### Utilities

- hetro had a number of fixes and improvements.


### Frontends

- Belacsound:

- CsoundQt:

### General Usage

- There have been a number of improvements in the semantics for multicore;
  most of these are corrections with a few efficiency gains.

- There is a maximum number of arguments for an opcode which was
  neither explicit nor policed.  Attempts to use too many arguments
  now gives a syntax error.


## Bugs Fixed

- FLgetsnap fixed.

- directory fixed regarding file extensions.

- FLsetText reused a string incorrectly which led to incorrect values.

- fmb3 failed to initialise the lfo rate in some cases.

- ftaudio at i-rate was totally broken.

- following a reinit printks could be skipped; fixed

- printks fixed sot prints at correct times.

# SYSTEM LEVEL CHANGES

- Hash Table implementation modified to expand on load for better performance
  when map contains large number of entries

### System Changes

- plugin GEN functions can have a zero length, but the code must check
  for this and act accordingly.  This allows for deferred allocations.

- schedule reports undefined instr numbers/names and continues, rater than 
  causing an error.
  
- allow multiple calls to midi out controls.

### Translations

### API

- find_opcode_new and find_opcode_exact now exposed in API.

- After a reset a default message string callback handle is configured.

- New function csoundSystemSr added to the API to read hardware-imposed sample rate.

### Platform Specific

- WebAudio: libsndfile now compiled with FLAC and OGG support.

- iOS

- Android

- Windows
 - in both orchestra and score the path tracking of #include expects a \ separator.

- MacOS

- GNU/Linux

- Haiku port now available.

- Bela
 - allow analog in and out with different channel numbers.

==END==

-----------------------------------------------------------------------
The following may need an entry above
------------------------------------------------------------------------
commit a0d0caf2f642a788a47b5dc138380d086fac058a
Merge: 35873dd434 175a790712
Author: John ffitch 
Date:   Wed Jun 26 15:05:03 2019 +0100

commit 0b4c6edb9f8631b2a1d207b25432326802518a04
Author: Felipe Sateler 
Date:   Sun Feb 3 18:54:08 2019 -0300

    perfThread: wait the recording thread before waiting for the performance thread

    If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
    thread but it has not been stopped yet.

Happy to test it.
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> 
> Hi,
> 
> Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> I will make a release candidate soon.
> 
> tarmo
> 
> Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> There have been many changed since 6.12 and we are getting close to a
> new release.  If anyone has an issue that must be resolved before 6.13
> or other reasons now is the time to speak.  Draft release notes
> below.  Again any corrections or improvements welcome.
> 
> ==John ffitch
> ========================================================================
> 
> 
> ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> # CSOUND VERSION 6.13 RELEASE NOTES
> 
> Not many new opcodes but there are a significant number of opcodes being
> extended to use arrays in a variety of ways, widening the options for
> users.  There have been many fixes to the core code as well as opcodes.
> -- The Developers
> 
> ## USER-LEVEL CHANGES
> 
> ### New opcodes
> 
> - string2array is a variant of fillarray with he data coming from a
>   string of space separated values.
> 
> - nstrstr returns the name string of an instrument number or an empty
>   string if the number does not refer to a named instrument.
> 
> - ntof converts notename to frequency at i- and k-time.
> 
> ### New Gen and Macros
> 
> - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
>   output gain factor with a maximum value of 1, modifying the output gain by
>   a dynamic range and a shaping exponent.
> 
> ### Orchestra
> 
> - The consistency of kr, sr and ksmps reworked especially when there
>   is overriding.
> 
> - corrected default 0bdfs usage.
> 
> - Resolving the path for #include files reworked to be more liberal.
> 
> - reading and writing to multidimensional arrays was very wrong.  This
>   is now correct.
> 
> - Better checking for unknown array types (issue #1124)
> 
> - In all array operations the size of an array is determined at init time and no 
>   allocation happens at perf time.
> 
> - array arithmetic now respects --sample-accurate.
> 
> ### Score
> 
> - The characters n and m could erroneously get ignored in scores.
> 
> - Resolving the path for #include files reworked to be more liberal.
> 
> - After an error the backtrace of files and macros incorrectly read
>   the information for orchestra rather than scores; fixed.
> 
> - The end of an r (repeated) section was not always correct.
> 
> - Nested {} parts of a score could lead to errors.
> 
> - After an s statement a newline was required; no longer needed.
> 
> - The {} score loops have been reworked to allow macros and expressions
>   in the loop count.
> 
> ### Options
> 
> - The new option --use-system-sr st the sample rate to the hardware/system value.
> 
> ### Modified Opcodes and Gens
> 
> - Sending failure in OSC is now a warning rather than an error.
> 
> - passign can now have an array as the target.
> 
> - version of bpf/bpfcos added to allow points defined via arrays.
> 
> - grain can now use tables of any size, which was only a power of 2.
> 
> - Changing colours in FL widgets now works (it previously did not redraw
>   the colour).
> 
> - fillarray can read from a file of values overcoming the argument
>   limit.
> 
> - sumarray now works for audio arrays as well as for scalar values.
> 
> - assignment of an audio value to an audio array now works.
> 
> - monitor was broken in the array form.
> 
> - gendyc now respects sample-accurate mode.
> 
> - mtof and ftom now have array versions.
> 
> - sc_lag and sc_lagud now use the first k- or a-rate input when no
>   initial value is given.
> 
> - printarray now works for string arrays.
> 
> - changed2 now works for strings.
> 
> - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> 
> - GEN01 correctly reads raw audio files when requested.
> 
> - ftaudio can now take two additional optional argument`s for the start
>   and end of the table data being written to file.
> 
> - sensekey recoded in the 'key down' mode.
> 
> - loscilx can return an audio array.
> 
> - schedule opcode reports undefined instruments in all cases.
> 
> - event_i now accepts tagged instrument numbers.
> 
> - printarray treats %d correctly.
> 
> ### Utilities
> 
> - hetro had a number of fixes and improvements.
> 
> 
> ### Frontends
> 
> - Belacsound:
> 
> - CsoundQt:
> 
> ### General Usage
> 
> - There have been a number of improvements in the semantics for multicore;
>   most of these are corrections with a few efficiency gains.
> 
> - There is a maximum number of arguments for an opcode which was
>   neither explicit nor policed.  Attempts to use too many arguments
>   now gives a syntax error.
> 
> 
> ## Bugs Fixed
> 
> - FLgetsnap fixed.
> 
> - directory fixed regarding file extensions.
> 
> - FLsetText reused a string incorrectly which led to incorrect values.
> 
> - fmb3 failed to initialise the lfo rate in some cases.
> 
> - ftaudio at i-rate was totally broken.
> 
> - following a reinit printks could be skipped; fixed
> 
> - printks fixed sot prints at correct times.
> 
> # SYSTEM LEVEL CHANGES
> 
> - Hash Table implementation modified to expand on load for better performance
>   when map contains large number of entries
> 
> ### System Changes
> 
> - plugin GEN functions can have a zero length, but the code must check
>   for this and act accordingly.  This allows for deferred allocations.
> 
> - schedule reports undefined instr numbers/names and continues, rater than 
>   causing an error.
> 
> - allow multiple calls to midi out controls.
> 
> ### Translations
> 
> ### API
> 
> - find_opcode_new and find_opcode_exact now exposed in API.
> 
> - After a reset a default message string callback handle is configured.
> 
> - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> 
> ### Platform Specific
> 
> - WebAudio: libsndfile now compiled with FLAC and OGG support.
> 
> - iOS
> 
> - Android
> 
> - Windows
>  - in both orchestra and score the path tracking of #include expects a \ separator.
> 
> - MacOS
> 
> - GNU/Linux
> 
> - Haiku port now available.
> 
> - Bela
>  - allow analog in and out with different channel numbers.
> 
> ==END==
> 
> -----------------------------------------------------------------------
> The following may need an entry above
> ------------------------------------------------------------------------
> commit a0d0caf2f642a788a47b5dc138380d086fac058a
> Merge: 35873dd434 175a790712
> Author: John ffitch 
> Date:   Wed Jun 26 15:05:03 2019 +0100
> 
> commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> Author: Felipe Sateler 
> Date:   Sun Feb 3 18:54:08 2019 -0300
> 
>     perfThread: wait the recording thread before waiting for the performance thread
> 
>     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
>     thread but it has not been stopped yet.
> 
>     Fixes #

Hi,

I uploaded the CsoundQt 0.9.6 release to:

https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6

For Mac please use this one (without pythonqt):

https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg

Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.

Best greetings,

tarmo

Kontakt Tarmo Johannes (<trmjhnns@gmail.com>) kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:

Hi, Thanks!

I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.

I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.

 In CsoundQt

 there will be some more small fixes, like how templates are opened and some more.

Greetings,

tarmo

Kontakt Steven Yi (<stevenyi@gmail.com>) kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:

That build didn't work, but after a correction, we have:

https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe

It'd be nice if someone with a clean Windows system could test this.

On Sat, Jul 6, 2019 at 6:36 PM Steven Yi <stevenyi@gmail.com> wrote:

I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:

https://ci.appveyor.com/project/csound/csound/builds/25794751

for testing.  (Currently queued up)

One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name.

On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes <trmjhnns@gmail.com> wrote:

Hi,

I uploaded the binaries to:

https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1

Have not done release notes yet.

tarmo

Kontakt Victor Lazzarini (<Victor.Lazzarini@mu.ie>) kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:

Happy to test it.
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952

> On 4 Jul 2019, at 15:27, Tarmo Johannes <trmjhnns@GMAIL.COM> wrote:
>
> Hi,
>
> Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> I will make a release candidate soon.
>
> tarmo
>
> Kontakt John (<jpff@codemist.co.uk>) kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> There have been many changed since 6.12 and we are getting close to a
> new release.  If anyone has an issue that must be resolved before 6.13
> or other reasons now is the time to speak.  Draft release notes
> below.  Again any corrections or improvements welcome.
>
> ==John ffitch
> ========================================================================
> <!---
>
> To maintain this document use the following markdown:
>
> # First level heading
> ## Second level heading
> ### Third level heading
>
> - First level bullet point
>  - Second level bullet point
>   - Third level bullet point
>
> `inline code`
>
> ``` pre-formatted text etc.  ```
>
> [hyperlink](url for the hyperlink)
>
> Any valid HTML can also be used.
>
>  --->
>
> ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> # CSOUND VERSION 6.13 RELEASE NOTES
>
> Not many new opcodes but there are a significant number of opcodes being
> extended to use arrays in a variety of ways, widening the options for
> users.  There have been many fixes to the core code as well as opcodes.
> -- The Developers
>
> ## USER-LEVEL CHANGES
>
> ### New opcodes
>
> - string2array is a variant of fillarray with he data coming from a
>   string of space separated values.
>
> - nstrstr returns the name string of an instrument number or an empty
>   string if the number does not refer to a named instrument.
>
> - ntof converts notename to frequency at i- and k-time.
>
> ### New Gen and Macros
>
> - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
>   output gain factor with a maximum value of 1, modifying the output gain by
>   a dynamic range and a shaping exponent.
>
> ### Orchestra
>
> - The consistency of kr, sr and ksmps reworked especially when there
>   is overriding.
>
> - corrected default 0bdfs usage.
>
> - Resolving the path for #include files reworked to be more liberal.
>
> - reading and writing to multidimensional arrays was very wrong.  This
>   is now correct.
>
> - Better checking for unknown array types (issue #1124)
>
> - In all array operations the size of an array is determined at init time and no
>   allocation happens at perf time.
>
> - array arithmetic now respects --sample-accurate.
>
> ### Score
>
> - The characters n and m could erroneously get ignored in scores.
>
> - Resolving the path for #include files reworked to be more liberal.
>
> - After an error the backtrace of files and macros incorrectly read
>   the information for orchestra rather than scores; fixed.
>
> - The end of an r (repeated) section was not always correct.
>
> - Nested {} parts of a score could lead to errors.
>
> - After an s statement a newline was required; no longer needed.
>
> - The {} score loops have been reworked to allow macros and expressions
>   in the loop count.
>
> ### Options
>
> - The new option --use-system-sr st the sample rate to the hardware/system value.
>
> ### Modified Opcodes and Gens
>
> - Sending failure in OSC is now a warning rather than an error.
>
> - passign can now have an array as the target.
>
> - version of bpf/bpfcos added to allow points defined via arrays.
>
> - grain can now use tables of any size, which was only a power of 2.
>
> - Changing colours in FL widgets now works (it previously did not redraw
>   the colour).
>
> - fillarray can read from a file of values overcoming the argument
>   limit.
>
> - sumarray now works for audio arrays as well as for scalar values.
>
> - assignment of an audio value to an audio array now works.
>
> - monitor was broken in the array form.
>
> - gendyc now respects sample-accurate mode.
>
> - mtof and ftom now have array versions.
>
> - sc_lag and sc_lagud now use the first k- or a-rate input when no
>   initial value is given.
>
> - printarray now works for string arrays.
>
> - changed2 now works for strings.
>
> - diskgrain, syncgrain and syncloop now can do sample rate scaling.
>
> - GEN01 correctly reads raw audio files when requested.
>
> - ftaudio can now take two additional optional argument`s for the start
>   and end of the table data being written to file.
>
> - sensekey recoded in the 'key down' mode.
>
> - loscilx can return an audio array.
>
> - schedule opcode reports undefined instruments in all cases.
>
> - event_i now accepts tagged instrument numbers.
>
> - printarray treats %d correctly.
>
> ### Utilities
>
> - hetro had a number of fixes and improvements.
>
>
> ### Frontends
>
> - Belacsound:
>
> - CsoundQt:
>
> ### General Usage
>
> - There have been a number of improvements in the semantics for multicore;
>   most of these are corrections with a few efficiency gains.
>
> - There is a maximum number of arguments for an opcode which was
>   neither explicit nor policed.  Attempts to use too many arguments
>   now gives a syntax error.
>
>
> ## Bugs Fixed
>
> - FLgetsnap fixed.
>
> - directory fixed regarding file extensions.
>
> - FLsetText reused a string incorrectly which led to incorrect values.
>
> - fmb3 failed to initialise the lfo rate in some cases.
>
> - ftaudio at i-rate was totally broken.
>
> - following a reinit printks could be skipped; fixed
>
> - printks fixed sot prints at correct times.
>
> # SYSTEM LEVEL CHANGES
>
> - Hash Table implementation modified to expand on load for better performance
>   when map contains large number of entries
>
> ### System Changes
>
> - plugin GEN functions can have a zero length, but the code must check
>   for this and act accordingly.  This allows for deferred allocations.
>
> - schedule reports undefined instr numbers/names and continues, rater than
>   causing an error.
>
> - allow multiple calls to midi out controls.
>
> ### Translations
>
> ### API
>
> - find_opcode_new and find_opcode_exact now exposed in API.
>
> - After a reset a default message string callback handle is configured.
>
> - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
>
> ### Platform Specific
>
> - WebAudio: libsndfile now compiled with FLAC and OGG support.
>
> - iOS
>
> - Android
>
> - Windows
>  - in both orchestra and score the path tracking of #include expects a \ separator.
>
> - MacOS
>
> - GNU/Linux
>
> - Haiku port now available.
>
> - Bela
>  - allow analog in and out with different channel numbers.
>
> ==END==
>
> -----------------------------------------------------------------------
> The following may need an entry above
> ------------------------------------------------------------------------
> commit a0d0caf2f642a788a47b5dc138380d086fac058a
> Merge: 35873dd434 175a790712
> Author: John ffitch <jpff@codemist.co.uk>
> Date:   Wed Jun 26 15:05:03 2019 +0100
>
> commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> Author: Felipe Sateler <fsateler@gmail.com>
> Date:   Sun Feb 3 18:54:08 2019 -0300
>
>     perfThread: wait the recording thread before waiting for the performance thread
>
>     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
>     thread but it has not been stopped yet.
>
>     Fixes #1103

I switch the installer build to use 0.9.6. Building now on Appveyor:

https://ci.appveyor.com/project/csound/csound/history

On Wed, Jul 10, 2019 at 6:03 PM Victor Lazzarini  wrote:
>
> Will do, thx
>
> Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy
> Maynooth University
> Ireland
>
> On 10 Jul 2019, at 22:37, Tarmo Johannes  wrote:
>
> Hi,
>
> I uploaded the CsoundQt 0.9.6 release to:
>
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> For Mac please use this one (without pythonqt):
> https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
>
> Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
>
> Best greetings,
> tarmo
>
> Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
>>
>> Hi, Thanks!
>>
>> I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
>>
>> I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
>>  In CsoundQt
>>  there will be some more small fixes, like how templates are opened and some more.
>>
>> Greetings,
>> tarmo
>>
>> Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
>>>
>>> That build didn't work, but after a correction, we have:
>>>
>>> https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
>>>
>>> It'd be nice if someone with a clean Windows system could test this.
>>>
>>> On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
>>>>
>>>> I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
>>>>
>>>> https://ci.appveyor.com/project/csound/csound/builds/25794751
>>>>
>>>> for testing.  (Currently queued up)
>>>>
>>>> One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name.
>>>>
>>>>
>>>>
>>>>
>>>> On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> I uploaded the binaries to:
>>>>> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
>>>>> Have not done release notes yet.
>>>>>
>>>>> tarmo
>>>>>
>>>>>
>>>>> Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
>>>>>>
>>>>>> Happy to test it.
>>>>>> ========================
>>>>>> Prof. Victor Lazzarini
>>>>>> Dean of Arts, Celtic Studies, and Philosophy,
>>>>>> Maynooth University,
>>>>>> Maynooth, Co Kildare, Ireland
>>>>>> Tel: 00 353 7086936
>>>>>> Fax: 00 353 1 7086952
>>>>>>
>>>>>> > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
>>>>>> >
>>>>>> > Hi,
>>>>>> >
>>>>>> > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
>>>>>> > I will make a release candidate soon.
>>>>>> >
>>>>>> > tarmo
>>>>>> >
>>>>>> > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
>>>>>> > There have been many changed since 6.12 and we are getting close to a
>>>>>> > new release.  If anyone has an issue that must be resolved before 6.13
>>>>>> > or other reasons now is the time to speak.  Draft release notes
>>>>>> > below.  Again any corrections or improvements welcome.
>>>>>> >
>>>>>> > ==John ffitch
>>>>>> > ========================================================================
>>>>>> > 
>>>>>> >
>>>>>> > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
>>>>>> > # CSOUND VERSION 6.13 RELEASE NOTES
>>>>>> >
>>>>>> > Not many new opcodes but there are a significant number of opcodes being
>>>>>> > extended to use arrays in a variety of ways, widening the options for
>>>>>> > users.  There have been many fixes to the core code as well as opcodes.
>>>>>> > -- The Developers
>>>>>> >
>>>>>> > ## USER-LEVEL CHANGES
>>>>>> >
>>>>>> > ### New opcodes
>>>>>> >
>>>>>> > - string2array is a variant of fillarray with he data coming from a
>>>>>> >   string of space separated values.
>>>>>> >
>>>>>> > - nstrstr returns the name string of an instrument number or an empty
>>>>>> >   string if the number does not refer to a named instrument.
>>>>>> >
>>>>>> > - ntof converts notename to frequency at i- and k-time.
>>>>>> >
>>>>>> > ### New Gen and Macros
>>>>>> >
>>>>>> > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
>>>>>> >   output gain factor with a maximum value of 1, modifying the output gain by
>>>>>> >   a dynamic range and a shaping exponent.
>>>>>> >
>>>>>> > ### Orchestra
>>>>>> >
>>>>>> > - The consistency of kr, sr and ksmps reworked especially when there
>>>>>> >   is overriding.
>>>>>> >
>>>>>> > - corrected default 0bdfs usage.
>>>>>> >
>>>>>> > - Resolving the path for #include files reworked to be more liberal.
>>>>>> >
>>>>>> > - reading and writing to multidimensional arrays was very wrong.  This
>>>>>> >   is now correct.
>>>>>> >
>>>>>> > - Better checking for unknown array types (issue #1124)
>>>>>> >
>>>>>> > - In all array operations the size of an array is determined at init time and no
>>>>>> >   allocation happens at perf time.
>>>>>> >
>>>>>> > - array arithmetic now respects --sample-accurate.
>>>>>> >
>>>>>> > ### Score
>>>>>> >
>>>>>> > - The characters n and m could erroneously get ignored in scores.
>>>>>> >
>>>>>> > - Resolving the path for #include files reworked to be more liberal.
>>>>>> >
>>>>>> > - After an error the backtrace of files and macros incorrectly read
>>>>>> >   the information for orchestra rather than scores; fixed.
>>>>>> >
>>>>>> > - The end of an r (repeated) section was not always correct.
>>>>>> >
>>>>>> > - Nested {} parts of a score could lead to errors.
>>>>>> >
>>>>>> > - After an s statement a newline was required; no longer needed.
>>>>>> >
>>>>>> > - The {} score loops have been reworked to allow macros and expressions
>>>>>> >   in the loop count.
>>>>>> >
>>>>>> > ### Options
>>>>>> >
>>>>>> > - The new option --use-system-sr st the sample rate to the hardware/system value.
>>>>>> >
>>>>>> > ### Modified Opcodes and Gens
>>>>>> >
>>>>>> > - Sending failure in OSC is now a warning rather than an error.
>>>>>> >
>>>>>> > - passign can now have an array as the target.
>>>>>> >
>>>>>> > - version of bpf/bpfcos added to allow points defined via arrays.
>>>>>> >
>>>>>> > - grain can now use tables of any size, which was only a power of 2.
>>>>>> >
>>>>>> > - Changing colours in FL widgets now works (it previously did not redraw
>>>>>> >   the colour).
>>>>>> >
>>>>>> > - fillarray can read from a file of values overcoming the argument
>>>>>> >   limit.
>>>>>> >
>>>>>> > - sumarray now works for audio arrays as well as for scalar values.
>>>>>> >
>>>>>> > - assignment of an audio value to an audio array now works.
>>>>>> >
>>>>>> > - monitor was broken in the array form.
>>>>>> >
>>>>>> > - gendyc now respects sample-accurate mode.
>>>>>> >
>>>>>> > - mtof and ftom now have array versions.
>>>>>> >
>>>>>> > - sc_lag and sc_lagud now use the first k- or a-rate input when no
>>>>>> >   initial value is given.
>>>>>> >
>>>>>> > - printarray now works for string arrays.
>>>>>> >
>>>>>> > - changed2 now works for strings.
>>>>>> >
>>>>>> > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
>>>>>> >
>>>>>> > - GEN01 correctly reads raw audio files when requested.
>>>>>> >
>>>>>> > - ftaudio can now take two additional optional argument`s for the start
>>>>>> >   and end of the table data being written to file.
>>>>>> >
>>>>>> > - sensekey recoded in the 'key down' mode.
>>>>>> >
>>>>>> > - loscilx can return an audio array.
>>>>>> >
>>>>>> > - schedule opcode reports undefined instruments in all cases.
>>>>>> >
>>>>>> > - event_i now accepts tagged instrument numbers.
>>>>>> >
>>>>>> > - printarray treats %d correctly.
>>>>>> >
>>>>>> > ### Utilities
>>>>>> >
>>>>>> > - hetro had a number of fixes and improvements.
>>>>>> >
>>>>>> >
>>>>>> > ### Frontends
>>>>>> >
>>>>>> > - Belacsound:
>>>>>> >
>>>>>> > - CsoundQt:
>>>>>> >
>>>>>> > ### General Usage
>>>>>> >
>>>>>> > - There have been a number of improvements in the semantics for multicore;
>>>>>> >   most of these are corrections with a few efficiency gains.
>>>>>> >
>>>>>> > - There is a maximum number of arguments for an opcode which was
>>>>>> >   neither explicit nor policed.  Attempts to use too many arguments
>>>>>> >   now gives a syntax error.
>>>>>> >
>>>>>> >
>>>>>> > ## Bugs Fixed
>>>>>> >
>>>>>> > - FLgetsnap fixed.
>>>>>> >
>>>>>> > - directory fixed regarding file extensions.
>>>>>> >
>>>>>> > - FLsetText reused a string incorrectly which led to incorrect values.
>>>>>> >
>>>>>> > - fmb3 failed to initialise the lfo rate in some cases.
>>>>>> >
>>>>>> > - ftaudio at i-rate was totally broken.
>>>>>> >
>>>>>> > - following a reinit printks could be skipped; fixed
>>>>>> >
>>>>>> > - printks fixed sot prints at correct times.
>>>>>> >
>>>>>> > # SYSTEM LEVEL CHANGES
>>>>>> >
>>>>>> > - Hash Table implementation modified to expand on load for better performance
>>>>>> >   when map contains large number of entries
>>>>>> >
>>>>>> > ### System Changes
>>>>>> >
>>>>>> > - plugin GEN functions can have a zero length, but the code must check
>>>>>> >   for this and act accordingly.  This allows for deferred allocations.
>>>>>> >
>>>>>> > - schedule reports undefined instr numbers/names and continues, rater than
>>>>>> >   causing an error.
>>>>>> >
>>>>>> > - allow multiple calls to midi out controls.
>>>>>> >
>>>>>> > ### Translations
>>>>>> >
>>>>>> > ### API
>>>>>> >
>>>>>> > - find_opcode_new and find_opcode_exact now exposed in API.
>>>>>> >
>>>>>> > - After a reset a default message string callback handle is configured.
>>>>>> >
>>>>>> > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
>>>>>> >
>>>>>> > ### Platform Specific
>>>>>> >
>>>>>> > - WebAudio: libsndfile now compiled with FLAC and OGG support.
>>>>>> >
>>>>>> > - iOS
>>>>>> >
>>>>>> > - Android
>>>>>> >
>>>>>> > - Windows
>>>>>> >  - in both orchestra and score the path tracking of #include expects a \ separator.
>>>>>> >
>>>>>> > - MacOS
>>>>>> >
>>>>>> > - GNU/Linux
>>>>>> >
>>>>>> > - Haiku port now available.
>>>>>> >
>>>>>> > - Bela
>>>>>> >  - allow analog in and out with different channel numbers.
>>>>>> >
>>>>>> > ==END==
>>>>>> >
>>>>>> > -----------------------------------------------------------------------
>>>>>> > The following may need an entry above
>>>>>> > ------------------------------------------------------------------------
>>>>>> > commit a0d0caf2f642a788a47b5dc138380d086fac058a
>>>>>> > Merge: 35873dd434 175a790712
>>>>>> > Author: John ffitch 
>>>>>> > Date:   Wed Jun 26 15:05:03 2019 +0100
>>>>>> >
>>>>>> > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
>>>>>> > Author: Felipe Sateler 
>>>>>> > Date:   Sun Feb 3 18:54:08 2019 -0300
>>>>>> >
>>>>>> >     perfThread: wait the recording thread before waiting for the performance thread
>>>>>> >
>>>>>> >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
>>>>>> >     thread but it has not been stopped yet.
>>>>>> >
>>>>>> >     Fixes #1103

Hi Tarmo,

I got this error

Dyld Error Message:
  Library not loaded: @rpath/QtWebEngineWidgets.framework/Versions/5/QtWebEngineWidgets
  Referenced from: /private/var/folders/*/CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6
  Reason: image not found

best
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 10 Jul 2019, at 22:36, Tarmo Johannes  wrote:
> 
> Hi,
> 
> I uploaded the CsoundQt 0.9.6 release to:
> 
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> For Mac please use this one (without pythonqt):
> https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> 
> Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
> 
> Best greetings,
> tarmo
> 
> Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
> Hi, Thanks!
> 
> I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
> 
> I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
>  In CsoundQt
>  there will be some more small fixes, like how templates are opened and some more.
> 
> Greetings,
> tarmo
> 
> Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
> That build didn't work, but after a correction, we have:
> 
> https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
> 
> It'd be nice if someone with a clean Windows system could test this. 
> 
> On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
> I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
> 
> https://ci.appveyor.com/project/csound/csound/builds/25794751
> 
> for testing.  (Currently queued up)
> 
> One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name. 
> 
> 
> 
> 
> On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
> Hi,
> 
> I uploaded the binaries to:
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
> Have not done release notes yet.
> 
> tarmo
> 
> 
> Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
> Happy to test it.
> ========================
> Prof. Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy,
> Maynooth University,
> Maynooth, Co Kildare, Ireland
> Tel: 00 353 7086936
> Fax: 00 353 1 7086952 
> 
> > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> > 
> > Hi,
> > 
> > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> > I will make a release candidate soon.
> > 
> > tarmo
> > 
> > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> > There have been many changed since 6.12 and we are getting close to a
> > new release.  If anyone has an issue that must be resolved before 6.13
> > or other reasons now is the time to speak.  Draft release notes
> > below.  Again any corrections or improvements welcome.
> > 
> > ==John ffitch
> > ========================================================================
> > 
> > 
> > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> > # CSOUND VERSION 6.13 RELEASE NOTES
> > 
> > Not many new opcodes but there are a significant number of opcodes being
> > extended to use arrays in a variety of ways, widening the options for
> > users.  There have been many fixes to the core code as well as opcodes.
> > -- The Developers
> > 
> > ## USER-LEVEL CHANGES
> > 
> > ### New opcodes
> > 
> > - string2array is a variant of fillarray with he data coming from a
> >   string of space separated values.
> > 
> > - nstrstr returns the name string of an instrument number or an empty
> >   string if the number does not refer to a named instrument.
> > 
> > - ntof converts notename to frequency at i- and k-time.
> > 
> > ### New Gen and Macros
> > 
> > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
> >   output gain factor with a maximum value of 1, modifying the output gain by
> >   a dynamic range and a shaping exponent.
> > 
> > ### Orchestra
> > 
> > - The consistency of kr, sr and ksmps reworked especially when there
> >   is overriding.
> > 
> > - corrected default 0bdfs usage.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - reading and writing to multidimensional arrays was very wrong.  This
> >   is now correct.
> > 
> > - Better checking for unknown array types (issue #1124)
> > 
> > - In all array operations the size of an array is determined at init time and no 
> >   allocation happens at perf time.
> > 
> > - array arithmetic now respects --sample-accurate.
> > 
> > ### Score
> > 
> > - The characters n and m could erroneously get ignored in scores.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - After an error the backtrace of files and macros incorrectly read
> >   the information for orchestra rather than scores; fixed.
> > 
> > - The end of an r (repeated) section was not always correct.
> > 
> > - Nested {} parts of a score could lead to errors.
> > 
> > - After an s statement a newline was required; no longer needed.
> > 
> > - The {} score loops have been reworked to allow macros and expressions
> >   in the loop count.
> > 
> > ### Options
> > 
> > - The new option --use-system-sr st the sample rate to the hardware/system value.
> > 
> > ### Modified Opcodes and Gens
> > 
> > - Sending failure in OSC is now a warning rather than an error.
> > 
> > - passign can now have an array as the target.
> > 
> > - version of bpf/bpfcos added to allow points defined via arrays.
> > 
> > - grain can now use tables of any size, which was only a power of 2.
> > 
> > - Changing colours in FL widgets now works (it previously did not redraw
> >   the colour).
> > 
> > - fillarray can read from a file of values overcoming the argument
> >   limit.
> > 
> > - sumarray now works for audio arrays as well as for scalar values.
> > 
> > - assignment of an audio value to an audio array now works.
> > 
> > - monitor was broken in the array form.
> > 
> > - gendyc now respects sample-accurate mode.
> > 
> > - mtof and ftom now have array versions.
> > 
> > - sc_lag and sc_lagud now use the first k- or a-rate input when no
> >   initial value is given.
> > 
> > - printarray now works for string arrays.
> > 
> > - changed2 now works for strings.
> > 
> > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> > 
> > - GEN01 correctly reads raw audio files when requested.
> > 
> > - ftaudio can now take two additional optional argument`s for the start
> >   and end of the table data being written to file.
> > 
> > - sensekey recoded in the 'key down' mode.
> > 
> > - loscilx can return an audio array.
> > 
> > - schedule opcode reports undefined instruments in all cases.
> > 
> > - event_i now accepts tagged instrument numbers.
> > 
> > - printarray treats %d correctly.
> > 
> > ### Utilities
> > 
> > - hetro had a number of fixes and improvements.
> > 
> > 
> > ### Frontends
> > 
> > - Belacsound:
> > 
> > - CsoundQt:
> > 
> > ### General Usage
> > 
> > - There have been a number of improvements in the semantics for multicore;
> >   most of these are corrections with a few efficiency gains.
> > 
> > - There is a maximum number of arguments for an opcode which was
> >   neither explicit nor policed.  Attempts to use too many arguments
> >   now gives a syntax error.
> > 
> > 
> > ## Bugs Fixed
> > 
> > - FLgetsnap fixed.
> > 
> > - directory fixed regarding file extensions.
> > 
> > - FLsetText reused a string incorrectly which led to incorrect values.
> > 
> > - fmb3 failed to initialise the lfo rate in some cases.
> > 
> > - ftaudio at i-rate was totally broken.
> > 
> > - following a reinit printks could be skipped; fixed
> > 
> > - printks fixed sot prints at correct times.
> > 
> > # SYSTEM LEVEL CHANGES
> > 
> > - Hash Table implementation modified to expand on load for better performance
> >   when map contains large number of entries
> > 
> > ### System Changes
> > 
> > - plugin GEN functions can have a zero length, but the code must check
> >   for this and act accordingly.  This allows for deferred allocations.
> > 
> > - schedule reports undefined instr numbers/names and continues, rater than 
> >   causing an error.
> > 
> > - allow multiple calls to midi out controls.
> > 
> > ### Translations
> > 
> > ### API
> > 
> > - find_opcode_new and find_opcode_exact now exposed in API.
> > 
> > - After a reset a default message string callback handle is configured.
> > 
> > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> > 
> > ### Platform Specific
> > 
> > - WebAudio: libsndfile now compiled with FLAC and OGG support.
> > 
> > - iOS
> > 
> > - Android
> > 
> > - Windows
> >  - in both orchestra and score the path tracking of #include expects a \ separator.
> > 
> > - MacOS
> > 
> > - GNU/Linux
> > 
> > - Haiku port now available.
> > 
> > - Bela
> >  - allow analog in and out with different channel numbers.
> > 
> > ==END==
> > 
> > -----------------------------------------------------------------------
> > The following may need an entry above
> > ------------------------------------------------------------------------
> > commit a0d0caf2f642a788a47b5dc138380d086fac058a
> > Merge: 35873dd434 175a790712
> > Author: John ffitch 
> > Date:   Wed Jun 26 15:05:03 2019 +0100
> > 
> > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> > Author: Felipe Sateler 
> > Date:   Sun Feb 3 18:54:08 2019 -0300
> > 
> >     perfThread: wait the recording thread before waiting for the performance thread
> > 
> >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
> >     thread but it has not been stopped yet.
> > 
> >     Fixes #1103
>

The library is in the App bundle, so I think it’s the rpath that is not set correctly in the build.
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 10 Jul 2019, at 22:36, Tarmo Johannes  wrote:
> 
> Hi,
> 
> I uploaded the CsoundQt 0.9.6 release to:
> 
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> For Mac please use this one (without pythonqt):
> https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> 
> Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
> 
> Best greetings,
> tarmo
> 
> Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
> Hi, Thanks!
> 
> I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
> 
> I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
>  In CsoundQt
>  there will be some more small fixes, like how templates are opened and some more.
> 
> Greetings,
> tarmo
> 
> Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
> That build didn't work, but after a correction, we have:
> 
> https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
> 
> It'd be nice if someone with a clean Windows system could test this. 
> 
> On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
> I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
> 
> https://ci.appveyor.com/project/csound/csound/builds/25794751
> 
> for testing.  (Currently queued up)
> 
> One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name. 
> 
> 
> 
> 
> On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
> Hi,
> 
> I uploaded the binaries to:
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
> Have not done release notes yet.
> 
> tarmo
> 
> 
> Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
> Happy to test it.
> ========================
> Prof. Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy,
> Maynooth University,
> Maynooth, Co Kildare, Ireland
> Tel: 00 353 7086936
> Fax: 00 353 1 7086952 
> 
> > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> > 
> > Hi,
> > 
> > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> > I will make a release candidate soon.
> > 
> > tarmo
> > 
> > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> > There have been many changed since 6.12 and we are getting close to a
> > new release.  If anyone has an issue that must be resolved before 6.13
> > or other reasons now is the time to speak.  Draft release notes
> > below.  Again any corrections or improvements welcome.
> > 
> > ==John ffitch
> > ========================================================================
> > 
> > 
> > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> > # CSOUND VERSION 6.13 RELEASE NOTES
> > 
> > Not many new opcodes but there are a significant number of opcodes being
> > extended to use arrays in a variety of ways, widening the options for
> > users.  There have been many fixes to the core code as well as opcodes.
> > -- The Developers
> > 
> > ## USER-LEVEL CHANGES
> > 
> > ### New opcodes
> > 
> > - string2array is a variant of fillarray with he data coming from a
> >   string of space separated values.
> > 
> > - nstrstr returns the name string of an instrument number or an empty
> >   string if the number does not refer to a named instrument.
> > 
> > - ntof converts notename to frequency at i- and k-time.
> > 
> > ### New Gen and Macros
> > 
> > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
> >   output gain factor with a maximum value of 1, modifying the output gain by
> >   a dynamic range and a shaping exponent.
> > 
> > ### Orchestra
> > 
> > - The consistency of kr, sr and ksmps reworked especially when there
> >   is overriding.
> > 
> > - corrected default 0bdfs usage.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - reading and writing to multidimensional arrays was very wrong.  This
> >   is now correct.
> > 
> > - Better checking for unknown array types (issue #1124)
> > 
> > - In all array operations the size of an array is determined at init time and no 
> >   allocation happens at perf time.
> > 
> > - array arithmetic now respects --sample-accurate.
> > 
> > ### Score
> > 
> > - The characters n and m could erroneously get ignored in scores.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - After an error the backtrace of files and macros incorrectly read
> >   the information for orchestra rather than scores; fixed.
> > 
> > - The end of an r (repeated) section was not always correct.
> > 
> > - Nested {} parts of a score could lead to errors.
> > 
> > - After an s statement a newline was required; no longer needed.
> > 
> > - The {} score loops have been reworked to allow macros and expressions
> >   in the loop count.
> > 
> > ### Options
> > 
> > - The new option --use-system-sr st the sample rate to the hardware/system value.
> > 
> > ### Modified Opcodes and Gens
> > 
> > - Sending failure in OSC is now a warning rather than an error.
> > 
> > - passign can now have an array as the target.
> > 
> > - version of bpf/bpfcos added to allow points defined via arrays.
> > 
> > - grain can now use tables of any size, which was only a power of 2.
> > 
> > - Changing colours in FL widgets now works (it previously did not redraw
> >   the colour).
> > 
> > - fillarray can read from a file of values overcoming the argument
> >   limit.
> > 
> > - sumarray now works for audio arrays as well as for scalar values.
> > 
> > - assignment of an audio value to an audio array now works.
> > 
> > - monitor was broken in the array form.
> > 
> > - gendyc now respects sample-accurate mode.
> > 
> > - mtof and ftom now have array versions.
> > 
> > - sc_lag and sc_lagud now use the first k- or a-rate input when no
> >   initial value is given.
> > 
> > - printarray now works for string arrays.
> > 
> > - changed2 now works for strings.
> > 
> > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> > 
> > - GEN01 correctly reads raw audio files when requested.
> > 
> > - ftaudio can now take two additional optional argument`s for the start
> >   and end of the table data being written to file.
> > 
> > - sensekey recoded in the 'key down' mode.
> > 
> > - loscilx can return an audio array.
> > 
> > - schedule opcode reports undefined instruments in all cases.
> > 
> > - event_i now accepts tagged instrument numbers.
> > 
> > - printarray treats %d correctly.
> > 
> > ### Utilities
> > 
> > - hetro had a number of fixes and improvements.
> > 
> > 
> > ### Frontends
> > 
> > - Belacsound:
> > 
> > - CsoundQt:
> > 
> > ### General Usage
> > 
> > - There have been a number of improvements in the semantics for multicore;
> >   most of these are corrections with a few efficiency gains.
> > 
> > - There is a maximum number of arguments for an opcode which was
> >   neither explicit nor policed.  Attempts to use too many arguments
> >   now gives a syntax error.
> > 
> > 
> > ## Bugs Fixed
> > 
> > - FLgetsnap fixed.
> > 
> > - directory fixed regarding file extensions.
> > 
> > - FLsetText reused a string incorrectly which led to incorrect values.
> > 
> > - fmb3 failed to initialise the lfo rate in some cases.
> > 
> > - ftaudio at i-rate was totally broken.
> > 
> > - following a reinit printks could be skipped; fixed
> > 
> > - printks fixed sot prints at correct times.
> > 
> > # SYSTEM LEVEL CHANGES
> > 
> > - Hash Table implementation modified to expand on load for better performance
> >   when map contains large number of entries
> > 
> > ### System Changes
> > 
> > - plugin GEN functions can have a zero length, but the code must check
> >   for this and act accordingly.  This allows for deferred allocations.
> > 
> > - schedule reports undefined instr numbers/names and continues, rater than 
> >   causing an error.
> > 
> > - allow multiple calls to midi out controls.
> > 
> > ### Translations
> > 
> > ### API
> > 
> > - find_opcode_new and find_opcode_exact now exposed in API.
> > 
> > - After a reset a default message string callback handle is configured.
> > 
> > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> > 
> > ### Platform Specific
> > 
> > - WebAudio: libsndfile now compiled with FLAC and OGG support.
> > 
> > - iOS
> > 
> > - Android
> > 
> > - Windows
> >  - in both orchestra and score the path tracking of #include expects a \ separator.
> > 
> > - MacOS
> > 
> > - GNU/Linux
> > 
> > - Haiku port now available.
> > 
> > - Bela
> >  - allow analog in and out with different channel numbers.
> > 
> > ==END==
> > 
> > -----------------------------------------------------------------------
> > The following may need an entry above
> > ------------------------------------------------------------------------
> > commit a0d0caf2f642a788a47b5dc138380d086fac058a
> > Merge: 35873dd434 175a790712
> > Author: John ffitch 
> > Date:   Wed Jun 26 15:05:03 2019 +0100
> > 
> > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> > Author: Felipe Sateler 
> > Date:   Sun Feb 3 18:54:08 2019 -0300
> > 
> >     perfThread: wait the recording thread before waiting for the performance thread
> > 
> >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
> >     thread but it has 

Hi Tarmo,

here’s the dependency list you are using. You see it fails with the first lib that depends on the rpath

otool -L CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6 
CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6:
	/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 1560.12.0)
	/System/Library/Frameworks/CoreMIDI.framework/Versions/A/CoreMIDI (compatibility version 1.0.0, current version 69.0.0)
	/System/Library/Frameworks/CoreAudio.framework/Versions/A/CoreAudio (compatibility version 1.0.0, current version 1.0.0)
	CsoundLib64.framework/Versions/6.0/CsoundLib64 (compatibility version 6.0.0, current version 0.0.0)
	libcsnd6.6.0.dylib (compatibility version 6.0.0, current version 0.0.0)
	@rpath/QtWebEngineWidgets.framework/Versions/5/QtWebEngineWidgets (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtWebEngineCore.framework/Versions/5/QtWebEngineCore (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtQuick.framework/Versions/5/QtQuick (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtQml.framework/Versions/5/QtQml (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtNetwork.framework/Versions/5/QtNetwork (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtCore.framework/Versions/5/QtCore (compatibility version 5.12.0, current version 5.12.0)
	/System/Library/Frameworks/DiskArbitration.framework/Versions/A/DiskArbitration (compatibility version 1.0.0, current version 1.0.0)
	/System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility version 1.0.0, current version 275.0.0)
	@rpath/QtGui.framework/Versions/5/QtGui (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtWebChannel.framework/Versions/5/QtWebChannel (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtPositioning.framework/Versions/5/QtPositioning (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtPrintSupport.framework/Versions/5/QtPrintSupport (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtWidgets.framework/Versions/5/QtWidgets (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtQuickWidgets.framework/Versions/5/QtQuickWidgets (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtConcurrent.framework/Versions/5/QtConcurrent (compatibility version 5.12.0, current version 5.12.0)
	@rpath/QtXml.framework/Versions/5/QtXml (compatibility version 5.12.0, current version 5.12.0)
	/System/Library/Frameworks/OpenGL.framework/Versions/A/OpenGL (compatibility version 1.0.0, current version 1.0.0)
	/System/Library/Frameworks/AGL.framework/Versions/A/AGL (compatibility version 1.0.0, current version 1.0.0)
	/usr/lib/libc++.1.dylib (compatibility version 1.0.0, current version 400.9.4)
	/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1252.200.5)

You could try setting this as a relative path

../frameworks/

instead of @rpath

best
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 10 Jul 2019, at 22:36, Tarmo Johannes  wrote:
> 
> Hi,
> 
> I uploaded the CsoundQt 0.9.6 release to:
> 
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> For Mac please use this one (without pythonqt):
> https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> 
> Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
> 
> Best greetings,
> tarmo
> 
> Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
> Hi, Thanks!
> 
> I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
> 
> I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
>  In CsoundQt
>  there will be some more small fixes, like how templates are opened and some more.
> 
> Greetings,
> tarmo
> 
> Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
> That build didn't work, but after a correction, we have:
> 
> https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
> 
> It'd be nice if someone with a clean Windows system could test this. 
> 
> On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
> I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
> 
> https://ci.appveyor.com/project/csound/csound/builds/25794751
> 
> for testing.  (Currently queued up)
> 
> One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name. 
> 
> 
> 
> 
> On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
> Hi,
> 
> I uploaded the binaries to:
> https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
> Have not done release notes yet.
> 
> tarmo
> 
> 
> Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
> Happy to test it.
> ========================
> Prof. Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy,
> Maynooth University,
> Maynooth, Co Kildare, Ireland
> Tel: 00 353 7086936
> Fax: 00 353 1 7086952 
> 
> > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> > 
> > Hi,
> > 
> > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> > I will make a release candidate soon.
> > 
> > tarmo
> > 
> > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> > There have been many changed since 6.12 and we are getting close to a
> > new release.  If anyone has an issue that must be resolved before 6.13
> > or other reasons now is the time to speak.  Draft release notes
> > below.  Again any corrections or improvements welcome.
> > 
> > ==John ffitch
> > ========================================================================
> > 
> > 
> > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> > # CSOUND VERSION 6.13 RELEASE NOTES
> > 
> > Not many new opcodes but there are a significant number of opcodes being
> > extended to use arrays in a variety of ways, widening the options for
> > users.  There have been many fixes to the core code as well as opcodes.
> > -- The Developers
> > 
> > ## USER-LEVEL CHANGES
> > 
> > ### New opcodes
> > 
> > - string2array is a variant of fillarray with he data coming from a
> >   string of space separated values.
> > 
> > - nstrstr returns the name string of an instrument number or an empty
> >   string if the number does not refer to a named instrument.
> > 
> > - ntof converts notename to frequency at i- and k-time.
> > 
> > ### New Gen and Macros
> > 
> > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
> >   output gain factor with a maximum value of 1, modifying the output gain by
> >   a dynamic range and a shaping exponent.
> > 
> > ### Orchestra
> > 
> > - The consistency of kr, sr and ksmps reworked especially when there
> >   is overriding.
> > 
> > - corrected default 0bdfs usage.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - reading and writing to multidimensional arrays was very wrong.  This
> >   is now correct.
> > 
> > - Better checking for unknown array types (issue #1124)
> > 
> > - In all array operations the size of an array is determined at init time and no 
> >   allocation happens at perf time.
> > 
> > - array arithmetic now respects --sample-accurate.
> > 
> > ### Score
> > 
> > - The characters n and m could erroneously get ignored in scores.
> > 
> > - Resolving the path for #include files reworked to be more liberal.
> > 
> > - After an error the backtrace of files and macros incorrectly read
> >   the information for orchestra rather than scores; fixed.
> > 
> > - The end of an r (repeated) section was not always correct.
> > 
> > - Nested {} parts of a score could lead to errors.
> > 
> > - After an s statement a newline was required; no longer needed.
> > 
> > - The {} score loops have been reworked to allow macros and expressions
> >   in the loop count.
> > 
> > ### Options
> > 
> > - The new option --use-system-sr st the sample rate to the hardware/system value.
> > 
> > ### Modified Opcodes and Gens
> > 
> > - Sending failure in OSC is now a warning rather than an error.
> > 
> > - passign can now have an array as the target.
> > 
> > - version of bpf/bpfcos added to allow points defined via arrays.
> > 
> > - grain can now use tables of any size, which was only a power of 2.
> > 
> > - Changing colours in FL widgets now works (it previously did not redraw
> >   the colour).
> > 
> > - fillarray can read from a file of values overcoming the argument
> >   limit.
> > 
> > - sumarray now works for audio arrays as well as for scalar values.
> > 
> > - assignment of an audio value to an audio array now works.
> > 
> > - monitor was broken in the array form.
> > 
> > - gendyc now respects sample-accurate mode.
> > 
> > - mtof and ftom now have array versions.
> > 
> > - sc_lag and sc_lagud now use the first k- or a-rate input when no
> >   initial value is given.
> > 
> > - printarray now works for string arrays.
> > 
> > - changed2 now works for strings.
> > 
> > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> > 
> > - GEN01 correctly reads raw audio files when requested.
> > 
> > - ftaudio can now take two additional optional argument`s for the start
> >   and end of the table data being written to file.
> > 
> > - sensekey recoded in the 'key down' mode.
> > 
> > - loscilx can return an audio array.
> > 
> > - schedule opcode reports undefined instruments in all cases.
> > 
> > - event_i now accepts tagged instrument numbers.
> > 
> > - printarray treats %d correctly.
> > 
> > ### Utilities
> > 
> > - hetro had a number of fixes and improvements.
> > 
> > 
> > ### Frontends
> > 
> > - Belacsound:
> > 
> > - CsoundQt:
> > 
> > ### General Usage
> > 
> > - There have been a number of improvements in the semantics for multicore;
> >   most of these are corrections with a few efficiency gains.
> > 
> > - There is a maximum number of arguments for an opcode which was
> >   neither explicit nor policed.  Attempts to use too many arguments
> >   now gives a syntax error.
> > 
> > 
> > ## Bugs Fixed
> > 
> > - FLgetsnap fixed.
> > 
> > - directory fixed regarding file extensions.
> > 
> > - FLsetText reused a string incorrectly which led to incorrect values.
> > 
> > - fmb3 failed to initialise the lfo rate in some cases.
> > 
> > - ftaudio at i-rate was totally broken.
> > 
> > - following a reinit printks could be skipped; fixed
> > 
> > - printks fixed sot prints at correct times.
> > 
> > # SYSTEM LEVEL CHANGES
> > 
> > - Hash Table implementation modified to expand on load for better performance
> >   when map contains large number of entries
> > 
> > ### System Changes
> > 
> > - plugin GEN functions can have a zero length, but the code must check
> >   for this and act accordingly.  This allows for deferred allocations.
> > 
> > - schedule reports undefined instr numbers/names and continues, rater than 
> >   causing an error.
> > 
> > - allow multiple calls to midi out controls.
> > 
> > ### Translations
> > 
> > ### API
> > 
> > - find_opcode_new and find_opcode_exact now exposed in API.
> > 
> > - After a reset a default message string callback handle is configured.
> > 
> > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> > 
> > ### Platform Specific
> > 
> > - WebAudio: libsndfile now compiled with FLAC and OGG support.
> > 
> > - iOS
> > 
> > - Android
> > 
> > - Windows
> >  - in both orchestra and score the path tracking of #include expects a \ separator.
> > 
> > - MacOS
> > 
> > - GNU/Linux
> > 
> > - Haiku port now available.
> > 
> > - Bela
> >  - allow analog in and out with different channel numbers.
> > 
> > ==END==
> > 
> > -----------------------------------------------------------------------
> > The following may need an entry above
> > ------------------------------------------------------------------------
> > commit a0d0caf2f642a788a47b5dc138380d086fac058a
> > Merge: 35873dd434 175a790712
> > Author: John ffitch 
> > Date:   Wed Jun 26 15:05:03 2019 +0100
> > 
> > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> > Author: Felipe Sateler 
> > Date:   Sun Feb 3 18:54:08 2019 -0300
> > 
> >     perfThread: wait the recording thread before waiting for the performance thread
> > 
> >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
> >     thread but it has not 

My suggestion was wrong, actually what works is to use @executable_path/../frameworks instead of @rpath.

I changed it using install_name_tool and it worked for that library.
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 11 Jul 2019, at 09:02, Tarmo Johannes  wrote:
> 
> Thanks,
> 
> @rpath itsself should work but there seems something to be wrong in the version in dmg. The app itsself here and the one with pythonqt (also when opened from dmg) work. I will rebuild.
> 
> tarmo
> 
> Kontakt Victor Lazzarini () kirjutas kuupäeval N, 11. juuli 2019 kell 10:56:
> Hi Tarmo,
> 
> here’s the dependency list you are using. You see it fails with the first lib that depends on the rpath
> 
> otool -L CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6 
> CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6:
>         /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 1560.12.0)
>         /System/Library/Frameworks/CoreMIDI.framework/Versions/A/CoreMIDI (compatibility version 1.0.0, current version 69.0.0)
>         /System/Library/Frameworks/CoreAudio.framework/Versions/A/CoreAudio (compatibility version 1.0.0, current version 1.0.0)
>         CsoundLib64.framework/Versions/6.0/CsoundLib64 (compatibility version 6.0.0, current version 0.0.0)
>         libcsnd6.6.0.dylib (compatibility version 6.0.0, current version 0.0.0)
>         @rpath/QtWebEngineWidgets.framework/Versions/5/QtWebEngineWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWebEngineCore.framework/Versions/5/QtWebEngineCore (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQuick.framework/Versions/5/QtQuick (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQml.framework/Versions/5/QtQml (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtNetwork.framework/Versions/5/QtNetwork (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtCore.framework/Versions/5/QtCore (compatibility version 5.12.0, current version 5.12.0)
>         /System/Library/Frameworks/DiskArbitration.framework/Versions/A/DiskArbitration (compatibility version 1.0.0, current version 1.0.0)
>         /System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility version 1.0.0, current version 275.0.0)
>         @rpath/QtGui.framework/Versions/5/QtGui (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWebChannel.framework/Versions/5/QtWebChannel (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtPositioning.framework/Versions/5/QtPositioning (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtPrintSupport.framework/Versions/5/QtPrintSupport (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWidgets.framework/Versions/5/QtWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQuickWidgets.framework/Versions/5/QtQuickWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtConcurrent.framework/Versions/5/QtConcurrent (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtXml.framework/Versions/5/QtXml (compatibility version 5.12.0, current version 5.12.0)
>         /System/Library/Frameworks/OpenGL.framework/Versions/A/OpenGL (compatibility version 1.0.0, current version 1.0.0)
>         /System/Library/Frameworks/AGL.framework/Versions/A/AGL (compatibility version 1.0.0, current version 1.0.0)
>         /usr/lib/libc++.1.dylib (compatibility version 1.0.0, current version 400.9.4)
>         /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1252.200.5)
> 
> You could try setting this as a relative path
> 
> ../frameworks/
> 
> instead of @rpath
> 
> best
> ========================
> Prof. Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy,
> Maynooth University,
> Maynooth, Co Kildare, Ireland
> Tel: 00 353 7086936
> Fax: 00 353 1 7086952 
> 
> > On 10 Jul 2019, at 22:36, Tarmo Johannes  wrote:
> > 
> > Hi,
> > 
> > I uploaded the CsoundQt 0.9.6 release to:
> > 
> > https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> > For Mac please use this one (without pythonqt):
> > https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> > 
> > Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
> > 
> > Best greetings,
> > tarmo
> > 
> > Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
> > Hi, Thanks!
> > 
> > I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
> > 
> > I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
> >  In CsoundQt
> >  there will be some more small fixes, like how templates are opened and some more.
> > 
> > Greetings,
> > tarmo
> > 
> > Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
> > That build didn't work, but after a correction, we have:
> > 
> > https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
> > 
> > It'd be nice if someone with a clean Windows system could test this. 
> > 
> > On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
> > I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
> > 
> > https://ci.appveyor.com/project/csound/csound/builds/25794751
> > 
> > for testing.  (Currently queued up)
> > 
> > One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name. 
> > 
> > 
> > 
> > 
> > On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
> > Hi,
> > 
> > I uploaded the binaries to:
> > https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
> > Have not done release notes yet.
> > 
> > tarmo
> > 
> > 
> > Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
> > Happy to test it.
> > ========================
> > Prof. Victor Lazzarini
> > Dean of Arts, Celtic Studies, and Philosophy,
> > Maynooth University,
> > Maynooth, Co Kildare, Ireland
> > Tel: 00 353 7086936
> > Fax: 00 353 1 7086952 
> > 
> > > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> > > 
> > > Hi,
> > > 
> > > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> > > I will make a release candidate soon.
> > > 
> > > tarmo
> > > 
> > > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> > > There have been many changed since 6.12 and we are getting close to a
> > > new release.  If anyone has an issue that must be resolved before 6.13
> > > or other reasons now is the time to speak.  Draft release notes
> > > below.  Again any corrections or improvements welcome.
> > > 
> > > ==John ffitch
> > > ========================================================================
> > > 
> > > 
> > > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> > > # CSOUND VERSION 6.13 RELEASE NOTES
> > > 
> > > Not many new opcodes but there are a significant number of opcodes being
> > > extended to use arrays in a variety of ways, widening the options for
> > > users.  There have been many fixes to the core code as well as opcodes.
> > > -- The Developers
> > > 
> > > ## USER-LEVEL CHANGES
> > > 
> > > ### New opcodes
> > > 
> > > - string2array is a variant of fillarray with he data coming from a
> > >   string of space separated values.
> > > 
> > > - nstrstr returns the name string of an instrument number or an empty
> > >   string if the number does not refer to a named instrument.
> > > 
> > > - ntof converts notename to frequency at i- and k-time.
> > > 
> > > ### New Gen and Macros
> > > 
> > > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
> > >   output gain factor with a maximum value of 1, modifying the output gain by
> > >   a dynamic range and a shaping exponent.
> > > 
> > > ### Orchestra
> > > 
> > > - The consistency of kr, sr and ksmps reworked especially when there
> > >   is overriding.
> > > 
> > > - corrected default 0bdfs usage.
> > > 
> > > - Resolving the path for #include files reworked to be more liberal.
> > > 
> > > - reading and writing to multidimensional arrays was very wrong.  This
> > >   is now correct.
> > > 
> > > - Better checking for unknown array types (issue #1124)
> > > 
> > > - In all array operations the size of an array is determined at init time and no 
> > >   allocation happens at perf time.
> > > 
> > > - array arithmetic now respects --sample-accurate.
> > > 
> > > ### Score
> > > 
> > > - The characters n and m could erroneously get ignored in scores.
> > > 
> > > - Resolving the path for #include files reworked to be more liberal.
> > > 
> > > - After an error the backtrace of files and macros incorrectly read
> > >   the information for orchestra rather than scores; fixed.
> > > 
> > > - The end of an r (repeated) section was not always correct.
> > > 
> > > - Nested {} parts of a score could lead to errors.
> > > 
> > > - After an s statement a newline was required; no longer needed.
> > > 
> > > - The {} score loops have been reworked to allow macros and expressions
> > >   in the loop count.
> > > 
> > > ### Options
> > > 
> > > - The new option --use-system-sr st the sample rate to the hardware/system value.
> > > 
> > > ### Modified Opcodes and Gens
> > > 
> > > - Sending failure in OSC is now a warning rather than an error.
> > > 
> > > - passign can now have an array as the target.
> > > 
> > > - version of bpf/bpfcos added to allow points defined via arrays.
> > > 
> > > - grain can now use tables of any size, which was only a power of 2.
> > > 
> > > - Changing colours in FL widgets now works (it previously did not redraw
> > >   the colour).
> > > 
> > > - fillarray can read from a file of values overcoming the argument
> > >   limit.
> > > 
> > > - sumarray now works for audio arrays as well as for scalar values.
> > > 
> > > - assignment of an audio value to an audio array now works.
> > > 
> > > - monitor was broken in the array form.
> > > 
> > > - gendyc now respects sample-accurate mode.
> > > 
> > > - mtof and ftom now have array versions.
> > > 
> > > - sc_lag and sc_lagud now use the first k- or a-rate input when no
> > >   initial value is given.
> > > 
> > > - printarray now works for string arrays.
> > > 
> > > - changed2 now works for strings.
> > > 
> > > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> > > 
> > > - GEN01 correctly reads raw audio files when requested.
> > > 
> > > - ftaudio can now take two additional optional argument`s for the start
> > >   and end of the table data being written to file.
> > > 
> > > - sensekey recoded in the 'key down' mode.
> > > 
> > > - loscilx can return an audio array.
> > > 
> > > - schedule opcode reports undefined instruments in all cases.
> > > 
> > > - event_i now accepts tagged instrument numbers.
> > > 
> > > - printarray treats %d correctly.
> > > 
> > > ### Utilities
> > > 
> > > - hetro had a number of fixes and improvements.
> > > 
> > > 
> > > ### Frontends
> > > 
> > > - Belacsound:
> > > 
> > > - CsoundQt:
> > > 
> > > ### General Usage
> > > 
> > > - There have been a number of improvements in the semantics for multicore;
> > >   most of these are corrections with a few efficiency gains.
> > > 
> > > - There is a maximum number of arguments for an opcode which was
> > >   neither explicit nor policed.  Attempts to use too many arguments
> > >   now gives a syntax error.
> > > 
> > > 
> > > ## Bugs Fixed
> > > 
> > > - FLgetsnap fixed.
> > > 
> > > - directory fixed regarding file extensions.
> > > 
> > > - FLsetText reused a string incorrectly which led to incorrect values.
> > > 
> > > - fmb3 failed to initialise the lfo rate in some cases.
> > > 
> > > - ftaudio at i-rate was totally broken.
> > > 
> > > - following a reinit printks could be skipped; fixed
> > > 
> > > - printks fixed sot prints at correct times.
> > > 
> > > # SYSTEM LEVEL CHANGES
> > > 
> > > - Hash Table implementation modified to expand on load for better performance
> > >   when map contains large number of entries
> > > 
> > > ### System Changes
> > > 
> > > - plugin GEN functions can have a zero length, but the code must check
> > >   for this and act accordingly.  This allows for deferred allocations.
> > > 
> > > - schedule reports undefined instr numbers/names and continues, rater than 
> > >   causing an error.
> > > 
> > > - allow multiple calls to midi out controls.
> > > 
> > > ### Translations
> > > 
> > > ### API
> > > 
> > > - find_opcode_new and find_opcode_exact now exposed in API.
> > > 
> > > - After a reset a default message string callback handle is configured.
> > > 
> > > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> > > 
> > > ### Platform Specific
> > > 
> > > - WebAudio: libsndfile now compiled with FLAC and OGG support.
> > > 
> > > - iOS
> > > 
> > > - Android
> > > 
> > > - Windows
> > >  - in both orchestra and score the path tracking of #include expects a \ separator.
> > > 
> > > - MacOS
> > > 
> > > - GNU/Linux
> > > 
> > > - Haiku port now available.
> > > 
> > > - Bela
> > >  - allow analog in and out with different channel numbers.
> > > 
> > > ==END==
> > > 
> > > -----------------------------------------------------------------------
> > > The following may need an entry above
> > > ------------------------------------------------------------------------
> > > commit a0d0caf2f642a788a47b5dc138380d086fac058a
> > > Merge: 35873dd434 175a790712
> > > Author: John ffitch 
> > > Date:   Wed Jun 26 15:05:03 2019 +0100
> > > 
> > > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> > > Author: Felipe Sateler 
> > > Date:   Sun Feb 3 18:54:08 2019 -0300
> > > 
> > >     perfThread: wait the recording thread before waiting for the performance thread
> > > 
> > >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
> > >     thread but it has not been stopped yet.
> 

That’s working fine and it will go on the package.
========================
Prof. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 11 Jul 2019, at 09:20, Tarmo Johannes  wrote:
> 
> I rebuilt and re-uploaded, dowloaded for testing and now it seems to be OK:
> https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> Sorry for the mess and thanks for noticing!
> 
> tarmo
> 
> 
> Kontakt Tarmo Johannes () kirjutas kuupäeval N, 11. juuli 2019 kell 11:02:
> Thanks,
> 
> @rpath itsself should work but there seems something to be wrong in the version in dmg. The app itsself here and the one with pythonqt (also when opened from dmg) work. I will rebuild.
> 
> tarmo
> 
> Kontakt Victor Lazzarini () kirjutas kuupäeval N, 11. juuli 2019 kell 10:56:
> Hi Tarmo,
> 
> here’s the dependency list you are using. You see it fails with the first lib that depends on the rpath
> 
> otool -L CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6 
> CsoundQt-d-html-cs6.app/Contents/MacOS/CsoundQt-d-html-cs6:
>         /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation (compatibility version 150.0.0, current version 1560.12.0)
>         /System/Library/Frameworks/CoreMIDI.framework/Versions/A/CoreMIDI (compatibility version 1.0.0, current version 69.0.0)
>         /System/Library/Frameworks/CoreAudio.framework/Versions/A/CoreAudio (compatibility version 1.0.0, current version 1.0.0)
>         CsoundLib64.framework/Versions/6.0/CsoundLib64 (compatibility version 6.0.0, current version 0.0.0)
>         libcsnd6.6.0.dylib (compatibility version 6.0.0, current version 0.0.0)
>         @rpath/QtWebEngineWidgets.framework/Versions/5/QtWebEngineWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWebEngineCore.framework/Versions/5/QtWebEngineCore (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQuick.framework/Versions/5/QtQuick (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQml.framework/Versions/5/QtQml (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtNetwork.framework/Versions/5/QtNetwork (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtCore.framework/Versions/5/QtCore (compatibility version 5.12.0, current version 5.12.0)
>         /System/Library/Frameworks/DiskArbitration.framework/Versions/A/DiskArbitration (compatibility version 1.0.0, current version 1.0.0)
>         /System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility version 1.0.0, current version 275.0.0)
>         @rpath/QtGui.framework/Versions/5/QtGui (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWebChannel.framework/Versions/5/QtWebChannel (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtPositioning.framework/Versions/5/QtPositioning (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtPrintSupport.framework/Versions/5/QtPrintSupport (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtWidgets.framework/Versions/5/QtWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtQuickWidgets.framework/Versions/5/QtQuickWidgets (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtConcurrent.framework/Versions/5/QtConcurrent (compatibility version 5.12.0, current version 5.12.0)
>         @rpath/QtXml.framework/Versions/5/QtXml (compatibility version 5.12.0, current version 5.12.0)
>         /System/Library/Frameworks/OpenGL.framework/Versions/A/OpenGL (compatibility version 1.0.0, current version 1.0.0)
>         /System/Library/Frameworks/AGL.framework/Versions/A/AGL (compatibility version 1.0.0, current version 1.0.0)
>         /usr/lib/libc++.1.dylib (compatibility version 1.0.0, current version 400.9.4)
>         /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1252.200.5)
> 
> You could try setting this as a relative path
> 
> ../frameworks/
> 
> instead of @rpath
> 
> best
> ========================
> Prof. Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy,
> Maynooth University,
> Maynooth, Co Kildare, Ireland
> Tel: 00 353 7086936
> Fax: 00 353 1 7086952 
> 
> > On 10 Jul 2019, at 22:36, Tarmo Johannes  wrote:
> > 
> > Hi,
> > 
> > I uploaded the CsoundQt 0.9.6 release to:
> > 
> > https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6
> > For Mac please use this one (without pythonqt):
> > https://github.com/CsoundQt/CsoundQt/releases/download/0.9.6/CsoundQt-0.9.6-MacOS.dmg
> > 
> > Victor, Steven, please  test shortly if it works for you, I think those are fine and can be used in installer.
> > 
> > Best greetings,
> > tarmo
> > 
> > Kontakt Tarmo Johannes () kirjutas kuupäeval P, 7. juuli 2019 kell 13:19:
> > Hi, Thanks!
> > 
> > I downloaded in on a Windows 7 computer and doing some short tests, everything seems to work.
> > 
> > I forgot to add extended examples (Ian McCurdy collection etc) to the CsoundQt zip, I will do that with next release candidate soon.
> >  In CsoundQt
> >  there will be some more small fixes, like how templates are opened and some more.
> > 
> > Greetings,
> > tarmo
> > 
> > Kontakt Steven Yi () kirjutas kuupäeval P, 7. juuli 2019 kell 05:40:
> > That build didn't work, but after a correction, we have:
> > 
> > https://ci.appveyor.com/api/buildjobs/u9jrejggs5rra6qf/artifacts/msvc%2FCsound6_x64-6.13.0-beta-317.exe
> > 
> > It'd be nice if someone with a clean Windows system could test this. 
> > 
> > On Sat, Jul 6, 2019 at 6:36 PM Steven Yi  wrote:
> > I updated the windows installer to use this build of CsoundQt.  Appveyor will build a release here:
> > 
> > https://ci.appveyor.com/project/csound/csound/builds/25794751
> > 
> > for testing.  (Currently queued up)
> > 
> > One thing to note is that the installer and directory within the installer has a typo of "CsdounQt" but the executable has the correct name. 
> > 
> > 
> > 
> > 
> > On Fri, Jul 5, 2019 at 10:41 AM Tarmo Johannes  wrote:
> > Hi,
> > 
> > I uploaded the binaries to:
> > https://github.com/CsoundQt/CsoundQt/releases/tag/0.9.6-rc1
> > Have not done release notes yet.
> > 
> > tarmo
> > 
> > 
> > Kontakt Victor Lazzarini () kirjutas kuupäeval R, 5. juuli 2019 kell 12:14:
> > Happy to test it.
> > ========================
> > Prof. Victor Lazzarini
> > Dean of Arts, Celtic Studies, and Philosophy,
> > Maynooth University,
> > Maynooth, Co Kildare, Ireland
> > Tel: 00 353 7086936
> > Fax: 00 353 1 7086952 
> > 
> > > On 4 Jul 2019, at 15:27, Tarmo Johannes  wrote:
> > > 
> > > Hi,
> > > 
> > > Just to let know that I would like to release CsoundQt 0.9.6 this time and suggest it to be bundled with the Csound installer.
> > > I will make a release candidate soon.
> > > 
> > > tarmo
> > > 
> > > Kontakt John () kirjutas kuupäeval T, 2. juuli 2019 kell 19:41:
> > > There have been many changed since 6.12 and we are getting close to a
> > > new release.  If anyone has an issue that must be resolved before 6.13
> > > or other reasons now is the time to speak.  Draft release notes
> > > below.  Again any corrections or improvements welcome.
> > > 
> > > ==John ffitch
> > > ========================================================================
> > > 
> > > 
> > > ========== DRAFT ========== DRAFT ========== DRAFT ========== DRAFT ========
> > > # CSOUND VERSION 6.13 RELEASE NOTES
> > > 
> > > Not many new opcodes but there are a significant number of opcodes being
> > > extended to use arrays in a variety of ways, widening the options for
> > > users.  There have been many fixes to the core code as well as opcodes.
> > > -- The Developers
> > > 
> > > ## USER-LEVEL CHANGES
> > > 
> > > ### New opcodes
> > > 
> > > - string2array is a variant of fillarray with he data coming from a
> > >   string of space separated values.
> > > 
> > > - nstrstr returns the name string of an instrument number or an empty
> > >   string if the number does not refer to a named instrument.
> > > 
> > > - ntof converts notename to frequency at i- and k-time.
> > > 
> > > ### New Gen and Macros
> > > 
> > > - ampmidicurve is a new opcode that maps an input MIDI velocity number to an
> > >   output gain factor with a maximum value of 1, modifying the output gain by
> > >   a dynamic range and a shaping exponent.
> > > 
> > > ### Orchestra
> > > 
> > > - The consistency of kr, sr and ksmps reworked especially when there
> > >   is overriding.
> > > 
> > > - corrected default 0bdfs usage.
> > > 
> > > - Resolving the path for #include files reworked to be more liberal.
> > > 
> > > - reading and writing to multidimensional arrays was very wrong.  This
> > >   is now correct.
> > > 
> > > - Better checking for unknown array types (issue #1124)
> > > 
> > > - In all array operations the size of an array is determined at init time and no 
> > >   allocation happens at perf time.
> > > 
> > > - array arithmetic now respects --sample-accurate.
> > > 
> > > ### Score
> > > 
> > > - The characters n and m could erroneously get ignored in scores.
> > > 
> > > - Resolving the path for #include files reworked to be more liberal.
> > > 
> > > - After an error the backtrace of files and macros incorrectly read
> > >   the information for orchestra rather than scores; fixed.
> > > 
> > > - The end of an r (repeated) section was not always correct.
> > > 
> > > - Nested {} parts of a score could lead to errors.
> > > 
> > > - After an s statement a newline was required; no longer needed.
> > > 
> > > - The {} score loops have been reworked to allow macros and expressions
> > >   in the loop count.
> > > 
> > > ### Options
> > > 
> > > - The new option --use-system-sr st the sample rate to the hardware/system value.
> > > 
> > > ### Modified Opcodes and Gens
> > > 
> > > - Sending failure in OSC is now a warning rather than an error.
> > > 
> > > - passign can now have an array as the target.
> > > 
> > > - version of bpf/bpfcos added to allow points defined via arrays.
> > > 
> > > - grain can now use tables of any size, which was only a power of 2.
> > > 
> > > - Changing colours in FL widgets now works (it previously did not redraw
> > >   the colour).
> > > 
> > > - fillarray can read from a file of values overcoming the argument
> > >   limit.
> > > 
> > > - sumarray now works for audio arrays as well as for scalar values.
> > > 
> > > - assignment of an audio value to an audio array now works.
> > > 
> > > - monitor was broken in the array form.
> > > 
> > > - gendyc now respects sample-accurate mode.
> > > 
> > > - mtof and ftom now have array versions.
> > > 
> > > - sc_lag and sc_lagud now use the first k- or a-rate input when no
> > >   initial value is given.
> > > 
> > > - printarray now works for string arrays.
> > > 
> > > - changed2 now works for strings.
> > > 
> > > - diskgrain, syncgrain and syncloop now can do sample rate scaling.
> > > 
> > > - GEN01 correctly reads raw audio files when requested.
> > > 
> > > - ftaudio can now take two additional optional argument`s for the start
> > >   and end of the table data being written to file.
> > > 
> > > - sensekey recoded in the 'key down' mode.
> > > 
> > > - loscilx can return an audio array.
> > > 
> > > - schedule opcode reports undefined instruments in all cases.
> > > 
> > > - event_i now accepts tagged instrument numbers.
> > > 
> > > - printarray treats %d correctly.
> > > 
> > > ### Utilities
> > > 
> > > - hetro had a number of fixes and improvements.
> > > 
> > > 
> > > ### Frontends
> > > 
> > > - Belacsound:
> > > 
> > > - CsoundQt:
> > > 
> > > ### General Usage
> > > 
> > > - There have been a number of improvements in the semantics for multicore;
> > >   most of these are corrections with a few efficiency gains.
> > > 
> > > - There is a maximum number of arguments for an opcode which was
> > >   neither explicit nor policed.  Attempts to use too many arguments
> > >   now gives a syntax error.
> > > 
> > > 
> > > ## Bugs Fixed
> > > 
> > > - FLgetsnap fixed.
> > > 
> > > - directory fixed regarding file extensions.
> > > 
> > > - FLsetText reused a string incorrectly which led to incorrect values.
> > > 
> > > - fmb3 failed to initialise the lfo rate in some cases.
> > > 
> > > - ftaudio at i-rate was totally broken.
> > > 
> > > - following a reinit printks could be skipped; fixed
> > > 
> > > - printks fixed sot prints at correct times.
> > > 
> > > # SYSTEM LEVEL CHANGES
> > > 
> > > - Hash Table implementation modified to expand on load for better performance
> > >   when map contains large number of entries
> > > 
> > > ### System Changes
> > > 
> > > - plugin GEN functions can have a zero length, but the code must check
> > >   for this and act accordingly.  This allows for deferred allocations.
> > > 
> > > - schedule reports undefined instr numbers/names and continues, rater than 
> > >   causing an error.
> > > 
> > > - allow multiple calls to midi out controls.
> > > 
> > > ### Translations
> > > 
> > > ### API
> > > 
> > > - find_opcode_new and find_opcode_exact now exposed in API.
> > > 
> > > - After a reset a default message string callback handle is configured.
> > > 
> > > - New function csoundSystemSr added to the API to read hardware-imposed sample rate.
> > > 
> > > ### Platform Specific
> > > 
> > > - WebAudio: libsndfile now compiled with FLAC and OGG support.
> > > 
> > > - iOS
> > > 
> > > - Android
> > > 
> > > - Windows
> > >  - in both orchestra and score the path tracking of #include expects a \ separator.
> > > 
> > > - MacOS
> > > 
> > > - GNU/Linux
> > > 
> > > - Haiku port now available.
> > > 
> > > - Bela
> > >  - allow analog in and out with different channel numbers.
> > > 
> > > ==END==
> > > 
> > > -----------------------------------------------------------------------
> > > The following may need an entry above
> > > ------------------------------------------------------------------------
> > > commit a0d0caf2f642a788a47b5dc138380d086fac058a
> > > Merge: 35873dd434 175a790712
> > > Author: John ffitch 
> > > Date:   Wed Jun 26 15:05:03 2019 +0100
> > > 
> > > commit 0b4c6edb9f8631b2a1d207b25432326802518a04
> > > Author: Felipe Sateler 
> > > Date:   Sun Feb 3 18:54:08 2019 -0300
> > > 
> > >     perfThread: wait the recording thread before waiting for the performance thread
> > > 
> > >     If we don't tell the recording thread to stop, we might enter a deadlock as the perf thread waits for the record
> > >     thread but it has not been stopped yet.
> > > 
> > >     F