This opcode gives an init error when passed an empty string. Can its
behavior be changed so that it returns 0 in the case of empty strings?

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Not sure what others think, but that seems reasonable to me. Just for
reference, I checked strtod and found:

http://www.cplusplus.com/reference/cstdlib/strtod/

and it mentions:

"If the first sequence of non-whitespace characters in str does not
form a valid floating-point number as just described, or if no such
sequence exists because either str is empty or contains only
whitespace characters, no conversion is performed and the function
returns a zero value."

I guess the flipside is that the current behavior might be useful to
help detect coding errors.

On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
> This opcode gives an init error when passed an empty string. Can its
> behavior be changed so that it returns 0 in the case of empty strings?
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I see your point. My work around was to add a preceding 0 to the
string I'm testing. I guess if this is how the underlying functions
work, it might be best to have them work in the same way in Csound.
Case closed.

On 31 July 2015 at 12:32, Steven Yi  wrote:
> Not sure what others think, but that seems reasonable to me. Just for
> reference, I checked strtod and found:
>
> http://www.cplusplus.com/reference/cstdlib/strtod/
>
> and it mentions:
>
> "If the first sequence of non-whitespace characters in str does not
> form a valid floating-point number as just described, or if no such
> sequence exists because either str is empty or contains only
> whitespace characters, no conversion is performed and the function
> returns a zero value."
>
> I guess the flipside is that the current behavior might be useful to
> help detect coding errors.
>
> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>> This opcode gives an init error when passed an empty string. Can its
>> behavior be changed so that it returns 0 in the case of empty strings?
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well the C strtod *does* allow for empty strings and returning 0. Then
again, it also appears to return 0 even with invalid data. I was just
putting out there what benefit the current behavior provides.  (In
other words, just wanted to add more information, but not really
voting on one or the other.)

On Fri, Jul 31, 2015 at 6:41 AM, Rory Walsh  wrote:
> I see your point. My work around was to add a preceding 0 to the
> string I'm testing. I guess if this is how the underlying functions
> work, it might be best to have them work in the same way in Csound.
> Case closed.
>
> On 31 July 2015 at 12:32, Steven Yi  wrote:
>> Not sure what others think, but that seems reasonable to me. Just for
>> reference, I checked strtod and found:
>>
>> http://www.cplusplus.com/reference/cstdlib/strtod/
>>
>> and it mentions:
>>
>> "If the first sequence of non-whitespace characters in str does not
>> form a valid floating-point number as just described, or if no such
>> sequence exists because either str is empty or contains only
>> whitespace characters, no conversion is performed and the function
>> returns a zero value."
>>
>> I guess the flipside is that the current behavior might be useful to
>> help detect coding errors.
>>
>> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>>> This opcode gives an init error when passed an empty string. Can its
>>> behavior be changed so that it returns 0 in the case of empty strings?
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

It's not the first time I've been caught out with this when working
with the API. I'm not sure what's best. Considering Csoud is written
in C, perhaps it should go the way of C rather than C++?

On 31 July 2015 at 13:47, Steven Yi  wrote:
> Well the C strtod *does* allow for empty strings and returning 0. Then
> again, it also appears to return 0 even with invalid data. I was just
> putting out there what benefit the current behavior provides.  (In
> other words, just wanted to add more information, but not really
> voting on one or the other.)
>
> On Fri, Jul 31, 2015 at 6:41 AM, Rory Walsh  wrote:
>> I see your point. My work around was to add a preceding 0 to the
>> string I'm testing. I guess if this is how the underlying functions
>> work, it might be best to have them work in the same way in Csound.
>> Case closed.
>>
>> On 31 July 2015 at 12:32, Steven Yi  wrote:
>>> Not sure what others think, but that seems reasonable to me. Just for
>>> reference, I checked strtod and found:
>>>
>>> http://www.cplusplus.com/reference/cstdlib/strtod/
>>>
>>> and it mentions:
>>>
>>> "If the first sequence of non-whitespace characters in str does not
>>> form a valid floating-point number as just described, or if no such
>>> sequence exists because either str is empty or contains only
>>> whitespace characters, no conversion is performed and the function
>>> returns a zero value."
>>>
>>> I guess the flipside is that the current behavior might be useful to
>>> help detect coding errors.
>>>
>>> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>>>> This opcode gives an init error when passed an empty string. Can its
>>>> behavior be changed so that it returns 0 in the case of empty strings?
>>>>
>>>> ------------------------------------------------------------------------------
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I just read through the strtod opcode's source:

https://github.com/csound/csound/blob/develop/OOps/str_ops.c#L536-L575

and it seems to be doing the right thing to me.  I'm going to vote to
keep the current behavior.

On Fri, Jul 31, 2015 at 7:54 AM, Rory Walsh  wrote:
> It's not the first time I've been caught out with this when working
> with the API. I'm not sure what's best. Considering Csoud is written
> in C, perhaps it should go the way of C rather than C++?
>
> On 31 July 2015 at 13:47, Steven Yi  wrote:
>> Well the C strtod *does* allow for empty strings and returning 0. Then
>> again, it also appears to return 0 even with invalid data. I was just
>> putting out there what benefit the current behavior provides.  (In
>> other words, just wanted to add more information, but not really
>> voting on one or the other.)
>>
>> On Fri, Jul 31, 2015 at 6:41 AM, Rory Walsh  wrote:
>>> I see your point. My work around was to add a preceding 0 to the
>>> string I'm testing. I guess if this is how the underlying functions
>>> work, it might be best to have them work in the same way in Csound.
>>> Case closed.
>>>
>>> On 31 July 2015 at 12:32, Steven Yi  wrote:
>>>> Not sure what others think, but that seems reasonable to me. Just for
>>>> reference, I checked strtod and found:
>>>>
>>>> http://www.cplusplus.com/reference/cstdlib/strtod/
>>>>
>>>> and it mentions:
>>>>
>>>> "If the first sequence of non-whitespace characters in str does not
>>>> form a valid floating-point number as just described, or if no such
>>>> sequence exists because either str is empty or contains only
>>>> whitespace characters, no conversion is performed and the function
>>>> returns a zero value."
>>>>
>>>> I guess the flipside is that the current behavior might be useful to
>>>> help detect coding errors.
>>>>
>>>> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>>>>> This opcode gives an init error when passed an empty string. Can its
>>>>> behavior be changed so that it returns 0 in the case of empty strings?
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>> ------------------------------------------------------------------------------
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks Steven, that's fine. If I clash with it again, and come undone,
I'll get back to you.

On 31 July 2015 at 14:45, Steven Yi  wrote:
> I just read through the strtod opcode's source:
>
> https://github.com/csound/csound/blob/develop/OOps/str_ops.c#L536-L575
>
> and it seems to be doing the right thing to me.  I'm going to vote to
> keep the current behavior.
>
> On Fri, Jul 31, 2015 at 7:54 AM, Rory Walsh  wrote:
>> It's not the first time I've been caught out with this when working
>> with the API. I'm not sure what's best. Considering Csoud is written
>> in C, perhaps it should go the way of C rather than C++?
>>
>> On 31 July 2015 at 13:47, Steven Yi  wrote:
>>> Well the C strtod *does* allow for empty strings and returning 0. Then
>>> again, it also appears to return 0 even with invalid data. I was just
>>> putting out there what benefit the current behavior provides.  (In
>>> other words, just wanted to add more information, but not really
>>> voting on one or the other.)
>>>
>>> On Fri, Jul 31, 2015 at 6:41 AM, Rory Walsh  wrote:
>>>> I see your point. My work around was to add a preceding 0 to the
>>>> string I'm testing. I guess if this is how the underlying functions
>>>> work, it might be best to have them work in the same way in Csound.
>>>> Case closed.
>>>>
>>>> On 31 July 2015 at 12:32, Steven Yi  wrote:
>>>>> Not sure what others think, but that seems reasonable to me. Just for
>>>>> reference, I checked strtod and found:
>>>>>
>>>>> http://www.cplusplus.com/reference/cstdlib/strtod/
>>>>>
>>>>> and it mentions:
>>>>>
>>>>> "If the first sequence of non-whitespace characters in str does not
>>>>> form a valid floating-point number as just described, or if no such
>>>>> sequence exists because either str is empty or contains only
>>>>> whitespace characters, no conversion is performed and the function
>>>>> returns a zero value."
>>>>>
>>>>> I guess the flipside is that the current behavior might be useful to
>>>>> help detect coding errors.
>>>>>
>>>>> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>>>>>> This opcode gives an init error when passed an empty string. Can its
>>>>>> behavior be changed so that it returns 0 in the case of empty strings?
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> _______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>> ------------------------------------------------------------------------------
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

another reason for keeping the current behaviour may be that otherwise 
their would be no distiction between an empty string input and "0" as 
input?!

	joachim


Am 31.07.2015 um 14:54 schrieb Rory Walsh:
> Thanks Steven, that's fine. If I clash with it again, and come undone,
> I'll get back to you.
>
> On 31 July 2015 at 14:45, Steven Yi  wrote:
>> I just read through the strtod opcode's source:
>>
>> https://github.com/csound/csound/blob/develop/OOps/str_ops.c#L536-L575
>>
>> and it seems to be doing the right thing to me.  I'm going to vote to
>> keep the current behavior.
>>
>> On Fri, Jul 31, 2015 at 7:54 AM, Rory Walsh  wrote:
>>> It's not the first time I've been caught out with this when working
>>> with the API. I'm not sure what's best. Considering Csoud is written
>>> in C, perhaps it should go the way of C rather than C++?
>>>
>>> On 31 July 2015 at 13:47, Steven Yi  wrote:
>>>> Well the C strtod *does* allow for empty strings and returning 0. Then
>>>> again, it also appears to return 0 even with invalid data. I was just
>>>> putting out there what benefit the current behavior provides.  (In
>>>> other words, just wanted to add more information, but not really
>>>> voting on one or the other.)
>>>>
>>>> On Fri, Jul 31, 2015 at 6:41 AM, Rory Walsh  wrote:
>>>>> I see your point. My work around was to add a preceding 0 to the
>>>>> string I'm testing. I guess if this is how the underlying functions
>>>>> work, it might be best to have them work in the same way in Csound.
>>>>> Case closed.
>>>>>
>>>>> On 31 July 2015 at 12:32, Steven Yi  wrote:
>>>>>> Not sure what others think, but that seems reasonable to me. Just for
>>>>>> reference, I checked strtod and found:
>>>>>>
>>>>>> http://www.cplusplus.com/reference/cstdlib/strtod/
>>>>>>
>>>>>> and it mentions:
>>>>>>
>>>>>> "If the first sequence of non-whitespace characters in str does not
>>>>>> form a valid floating-point number as just described, or if no such
>>>>>> sequence exists because either str is empty or contains only
>>>>>> whitespace characters, no conversion is performed and the function
>>>>>> returns a zero value."
>>>>>>
>>>>>> I guess the flipside is that the current behavior might be useful to
>>>>>> help detect coding errors.
>>>>>>
>>>>>> On Fri, Jul 31, 2015 at 6:18 AM, Rory Walsh  wrote:
>>>>>>> This opcode gives an init error when passed an empty string. Can its
>>>>>>> behavior be changed so that it returns 0 in the case of empty strings?
>>>>>>>
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> _______________________________________________
>>>>>>> Csound-devel mailing list
>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> _______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> _______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>
>>>> ------------------------------------------------------------------------------
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>

------------------------------------------------------------------------------
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Touche :)