i am reviewing older emails and found this one.

It raise the question should the code be changed to agree with the manual 
or should the manual be changed.  I am sure convenience is not the 
correct decider so I thought I would ask the developer community what they 
think.

---------- Forwarded message ----------
Date: Wed, 20 Mar 2019 16:33:01 -0700
From: burton 
Reply-To: A discussion list for users of Csound 
To: CSOUND@LISTSERV.HEANET.IE
Subject: [Csnd] OSCsend: docs discrepancy

https://csound.com/docs/manual/OSCsend, section Performance:

 	kwhen — a message is sent whenever this value changes. A message will
always be sent on the first call.

but a message is not sent if kwhen = 0 (sound 6.13 from git 40a21168c)

so either (a) the manual should be change and “always” rephrased towards
“unless 0” or (b) the opcode should send on the first call even if kwhen=0.

i guess it’s simpler to fix the docs… it will not invalidate current usage
and it preserves the perhaps useful option of not sending on the first call.

alex.

csound_test_OSCsend_kwhen.csd

i'd vote to add the note in the manual.
	joachim



On 28/03/19 18:51, john wrote:
> i am reviewing older emails and found this one.
>
> It raise the question should the code be changed to agree with the
> manual or should the manual be changed.  I am sure convenience is not
> the correct decider so I thought I would ask the developer community
> what they think.
>
> ---------- Forwarded message ----------
> Date: Wed, 20 Mar 2019 16:33:01 -0700
> From: burton 
> Reply-To: A discussion list for users of Csound 
> To: CSOUND@LISTSERV.HEANET.IE
> Subject: [Csnd] OSCsend: docs discrepancy
>
> https://csound.com/docs/manual/OSCsend, section Performance:
>
>     kwhen — a message is sent whenever this value changes. A message will
> always be sent on the first call.
>
> but a message is not sent if kwhen = 0 (sound 6.13 from git 40a21168c)
>
> so either (a) the manual should be change and “always” rephrased towards
> “unless 0” or (b) the opcode should send on the first call even if kwhen=0.
>
> i guess it’s simpler to fix the docs… it will not invalidate current usage
> and it preserves the perhaps useful option of not sending on the first
> call.
>
> alex.
>
> csound_test_OSCsend_kwhen.csd
>