Hi All,

I worked with Stephen today to get csoundErrCnt changes fixed to build
on Windows. However, Travis is continuing to fail from the initial
commit, which the build errors and logs are here:

https://travis-ci.org/csound/csound/builds/617779505

The change to add csoundErrCnt has caused 3 of the CSD tests to fail.

I'm unable to do much investigating over the next few days due to
travel and conference presentation prep.  Could someone take a look at
the errors?

Note: csoundErrCnt had to work on Windows due to PUBLIC causing
issues. One thing I noticed was that I wasn't sure if this was meant
to be a part of the Csound API or if it was meant to be internal. If
it's meant to be a part of the API, I think the forward declaration in
Top/csound.c should be removed and the function should be declared in
include/csound.h.  (Also the other file that has the extern could be
removed then too.)

Following up, Travis is still failing due to the change in
csoundErrCnt and return code. The python tests show error codes of 256
rather than 1 which I'm not quite sure how that happens, considering
if I run the test individually then use "echo $?" I get the expected
error number.  I think the unit test suite also needs to be updated to
check for the number of errors reported.

On Sun, Dec 1, 2019 at 11:32 PM Steven Yi  wrote:
>
> Hi All,
>
> I worked with Stephen today to get csoundErrCnt changes fixed to build
> on Windows. However, Travis is continuing to fail from the initial
> commit, which the build errors and logs are here:
>
> https://travis-ci.org/csound/csound/builds/617779505
>
> The change to add csoundErrCnt has caused 3 of the CSD tests to fail.
>
> I'm unable to do much investigating over the next few days due to
> travel and conference presentation prep.  Could someone take a look at
> the errors?
>
> Note: csoundErrCnt had to work on Windows due to PUBLIC causing
> issues. One thing I noticed was that I wasn't sure if this was meant
> to be a part of the Csound API or if it was meant to be internal. If
> it's meant to be a part of the API, I think the forward declaration in
> Top/csound.c should be removed and the function should be declared in
> include/csound.h.  (Also the other file that has the extern could be
> removed then too.)
>

On Tue, 3 Dec 2019, Steven Yi wrote:

> Following up, Travis is still failing due to the change in
> csoundErrCnt and return code. The python tests show error codes of 256
> rather than 1 which I'm not quite sure how that happens, considering
> if I run the test individually then use "echo $?" I get the expected
> error number.  I think the unit test suite also needs to be updated to
> check for the number of errors reported.
>
> On Sun, Dec 1, 2019 at 11:32 PM Steven Yi  wrote:
>>
>> Hi All,
>>
>> I worked with Stephen today to get csoundErrCnt changes fixed to build
>> on Windows. However, Travis is continuing to fail from the initial
>> commit, which the build errors and logs are here:
>>
>> https://travis-ci.org/csound/csound/builds/617779505
>>
>> The change to add csoundErrCnt has caused 3 of the CSD tests to fail.
>>
>> I'm unable to do much investigating over the next few days due to
>> travel and conference presentation prep.  Could someone take a look at
>> the errors?
>>
>> Note: csoundErrCnt had to work on Windows due to PUBLIC causing
>> issues. One thing I noticed was that I wasn't sure if this was meant
>> to be a part of the Csound API or if it was meant to be internal. If
>> it's meant to be a part of the API, I think the forward declaration in
>> Top/csound.c should be removed and the function should be declared in
>> include/csound.h.  (Also the other file that has the extern could be
>> removed then too.)

it was supposed to be part of theAPI but I could not get csound->ErrorCnt 
to work.

I looked at te egression test failures (bugs,csd and bugi.csd) but 
couldnot seeanything wrong.  Perhas the ython harness needs chaging but I 
a not te right person for that.
==John ff

>>
>> steven

One other error seems to be  a missing beats.wav which cannot be 

Great :)

I'm glad the return code for errors in performance isn't 0 anymore.

I didn't test this yet, I find it strange that this should be a part of the Csound API but not internal.

On Tue, 3 Dec 2019 at 20:51, Steven Yi <stevenyi@gmail.com> wrote:

Following up, Travis is still failing due to the change in
csoundErrCnt and return code. The python tests show error codes of 256
rather than 1 which I'm not quite sure how that happens, considering
if I run the test individually then use "echo $?" I get the expected
error number.  I think the unit test suite also needs to be updated to
check for the number of errors reported.

On Sun, Dec 1, 2019 at 11:32 PM Steven Yi <stevenyi@gmail.com> wrote:
>
> Hi All,
>
> I worked with Stephen today to get csoundErrCnt changes fixed to build
> on Windows. However, Travis is continuing to fail from the initial
> commit, which the build errors and logs are here:
>
> https://travis-ci.org/csound/csound/builds/617779505
>
> The change to add csoundErrCnt has caused 3 of the CSD tests to fail.
>
> I'm unable to do much investigating over the next few days due to
> travel and conference presentation prep.  Could someone take a look at
> the errors?
>
> Note: csoundErrCnt had to work on Windows due to PUBLIC causing
> issues. One thing I noticed was that I wasn't sure if this was meant
> to be a part of the Csound API or if it was meant to be internal. If
> it's meant to be a part of the API, I think the forward declaration in
> Top/csound.c should be removed and the function should be declared in
> include/csound.h.  (Also the other file that has the extern could be
> removed then too.)
>
> steven

I’ve removed the Python tests from the build for the moment until the script is fixed.
========================
Prof. Victor Lazzarini
Maynooth University
Ireland

> On 3 Dec 2019, at 20:29, john  wrote:
> 
> One other error seems to be  a missing beats.wav which cannot be comitted to github because of .gitign