Hi all -

I discovered another serious bug in Csound 5.12.1. If any \ is used,  
Csound crahes. For instance:

instr 1
a1 oscils \
10000, 500, 0
out a1
endin

This breaks my Stockhausen example in the QuteCsound Examples menu,  
and I guess many other CSDs.
Can this be fixed?

Thanks -

	joachim


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Is this a parser issue? Have tried running the examples with both
parsers, old and new?

On 17 February 2010 10:43, joachim heintz  wrote:
> Hi all -
>
> I discovered another serious bug in Csound 5.12.1. If any \ is used, Csound
> crahes. For instance:
>
> instr 1
> a1 oscils \
> 10000, 500, 0
> out a1
> endin
>
> This breaks my Stockhausen example in the QuteCsound Examples menu, and I
> guess many other CSDs.
> Can this be fixed?
>
> Thanks -
>
>        joachim
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

--new-parser is ok, but who knows it outside the list? the usual user  
will just try and see it crashes and his result is that Csound is  
unstable ...


Am 17.02.2010 um 11:46 schrieb Rory Walsh:

> Is this a parser issue? Have tried running the examples with both
> parsers, old and new?
>
> On 17 February 2010 10:43, joachim heintz  wrote:
>> Hi all -
>>
>> I discovered another serious bug in Csound 5.12.1. If any \ is  
>> used, Csound
>> crahes. For instance:
>>
>> instr 1
>> a1 oscils \
>> 10000, 500, 0
>> out a1
>> endin
>>
>> This breaks my Stockhausen example in the QuteCsound Examples menu,  
>> and I
>> guess many other CSDs.
>> Can this be fixed?
>>
>> Thanks -
>>
>>        joachim
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

It is a known bug, but thanks for reporting. Andres had mentioned it  
before.
On 17 Feb 2010, at 10:43, joachim heintz wrote:

> Hi all -
>
> I discovered another serious bug in Csound 5.12.1. If any \ is used,  
> Csound crahes. For instance:
>
> instr 1
> a1 oscils \
> 10000, 500, 0
> out a1
> endin
>
> This breaks my Stockhausen example in the QuteCsound Examples menu,  
> and I guess many other CSDs.
> Can this be fixed?
>
> Thanks -
>
> 	joachim
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

ah, maybe that's what it is. I had some students crashing 5.12 in a
lab session yesterday, but did not have the time to get to the bottom
of it so we reverted to 5.11 for that session.
Does it report as "Bus error" or just "segfault" ?

Oeyvind


2010/2/17 joachim heintz :
> Hi all -
>
> I discovered another serious bug in Csound 5.12.1. If any \ is used, Csound
> crahes. For instance:
>
> instr 1
> a1 oscils \
> 10000, 500, 0
> out a1
> endin
>
> This breaks my Stockhausen example in the QuteCsound Examples menu, and I
> guess many other CSDs.
> Can this be fixed?
>
> Thanks -
>
>        joachim
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

In the Terminal as Bus error. QuteCsound quits immediately without any  
message.
This bug breaks (among others) the partikkel manual example, by the  
way ...
(please test for having fun with the output - here it produces tons of  
printouts)
:((((((

	j


Am 17.02.2010 um 20:51 schrieb Oeyvind Brandtsegg:

> ah, maybe that's what it is. I had some students crashing 5.12 in a
> lab session yesterday, but did not have the time to get to the bottom
> of it so we reverted to 5.11 for that session.
> Does it report as "Bus error" or just "segfault" ?
>
> Oeyvind
>
>
> 2010/2/17 joachim heintz :
>> Hi all -
>>
>> I discovered another serious bug in Csound 5.12.1. If any \ is  
>> used, Csound
>> crahes. For instance:
>>
>> instr 1
>> a1 oscils \
>> 10000, 500, 0
>> out a1
>> endin
>>
>> This breaks my Stockhausen example in the QuteCsound Examples menu,  
>> and I
>> guess many other CSDs.
>> Can this be fixed?
>>
>> Thanks -
>>
>>        joachim
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Thanks for taking the time to find the offender,
I felt I should try to find the error, but did not have any specific
ideas on where to start looking.
Oeyvind

2010/2/17 joachim heintz :
> In the Terminal as Bus error. QuteCsound quits immediately without any
> message.
> This bug breaks (among others) the partikkel manual example, by the way ...
> (please test for having fun with the output - here it produces tons of
> printouts)
> :((((((
>
>        j
>
>
> Am 17.02.2010 um 20:51 schrieb Oeyvind Brandtsegg:
>
>> ah, maybe that's what it is. I had some students crashing 5.12 in a
>> lab session yesterday, but did not have the time to get to the bottom
>> of it so we reverted to 5.11 for that session.
>> Does it report as "Bus error" or just "segfault" ?
>>
>> Oeyvind
>>
>>
>> 2010/2/17 joachim heintz :
>>>
>>> Hi all -
>>>
>>> I discovered another serious bug in Csound 5.12.1. If any \ is used,
>>> Csound
>>> crahes. For instance:
>>>
>>> instr 1
>>> a1 oscils \
>>> 10000, 500, 0
>>> out a1
>>> endin
>>>
>>> This breaks my Stockhausen example in the QuteCsound Examples menu, and I
>>> guess many other CSDs.
>>> Can this be fixed?
>>>
>>> Thanks -
>>>
>>>       joachim
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

joachim heintz wrote:
> 
> This breaks my Stockhausen example in the QuteCsound Examples menu,  
> and I guess many other CSDs.
> 
In another freeware area that I'm involved in (Inform 7 for writing
interactive fiction) the developers have a set of test files that they run
through before issuing a new release, in order to make sure that nothing has
broken. If this isn't being done with Csound, perhaps it would be useful to
develop such a suite.

Of course, there are a lot of features to test, but it should be possible to
write a set of 10 of 15 .csd files that use most of the standard features of
the system. Before a new release, these files could be executed, and then
(assuming there are no crashes), the output files could be compared using a
diff program of some sort, to make sure they're byte-identical.

There hasn't been a new release of Inform in almost a year; that's the
downside. (Well, there are probably other contributing factors at work
there, such as the fact that the whole development process is rather
shrouded in secrecy.) Csound development seems to proceed more quickly. So
there might be a tradeoff in terms of how often new versions would be
released. Or maybe not. I'm just saying ... hmm, that's a pretty basic bug,
and I'm surprised there's no procedure in place that would have caught it.

--JA

Another thing that breaks your example is the line (in instr 1):

p3	 =	 (iend / 76.2) + 5

I don't think this assignment should be allowed anyway. It should  
report an error, though,
instead of crashing (I guess the crash is the line continuation issue.).

Victor
On 18 Feb 2010, at 01:18, Jim Aikin wrote:

> joachim heintz wrote:
>>
>> This breaks my Stockhausen example in the QuteCsound Examples menu,
>> and I guess many other CSDs.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Sorry, I don't understand. Do you mean this statement is incorrect, or  
it is not executed correctly by the current version?


Am 18.02.2010 um 09:21 schrieb Victor Lazzarini:

> Another thing that breaks your example is the line (in instr 1):
>
> p3	 =	 (iend / 76.2) + 5
>
> I don't think this assignment should be allowed anyway. It should  
> report an error, though,
> instead of crashing (I guess the crash is the line continuation  
> issue.).
>
> Victor
> On 18 Feb 2010, at 01:18, Jim Aikin wrote:
>
>> joachim heintz wrote:
>>>
>>> This breaks my Stockhausen example in the QuteCsound Examples menu,
>>> and I guess many other CSDs.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I mean I think it should raise an error; instead it will crash Csound  
(with the new
parser; the old parser gets knocked off by the line continuation  
problem).
John is aware of the problem, because I had seen it on Monday and  
reported it to him.

Victor


On 18 Feb 2010, at 10:05, joachim heintz wrote:

> Sorry, I don't understand. Do you mean this statement is incorrect,  
> or it is not executed correctly by the current version?
>
>
> Am 18.02.2010 um 09:21 schrieb Victor Lazzarini:
>
>> Another thing that breaks your example is the line (in instr 1):
>>
>> p3	 =	 (iend / 76.2) + 5
>>
>> I don't think this assignment should be allowed anyway. It should  
>> report an error, though,
>> instead of crashing (I guess the crash is the line continuation  
>> issue.).
>>
>> Victor
>> On 18 Feb 2010, at 01:18, Jim Aikin wrote:
>>
>>> joachim heintz wrote:
>>>>
>>>> This breaks my Stockhausen example in the QuteCsound Examples menu,
>>>> and I guess many other CSDs.
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"