Re: Bug plot after calling plotyy

[Ben Abbott]

On Jan 27, 2013, at 4:57 PM, Philip Nienhuis wrote:

> Jordi Gutiérrez Hermoso-2 wrote
>> On 26 January 2013 20:05, Lourenço <
> 
>> lourencojunior@
> 
>> > wrote:
>>> On 26 January 2013 21:58, Jordi Gutiérrez Hermoso <
> 
>> jordigh@
> 
>> > wrote:
>>>> On 26 January 2013 18:01, Lourenco <
> 
>> lourenco.usp@
> 
>> > wrote:
>>>>> I reported this issue at stackoverflow:
>>>>> 
>>>>> http://stackoverflow.com/questions/14533543/error-plotting-data-read-from-csv-file
>>>> 
>>>> We're doing something wrong if your first impression is to report bugs
>>>> to Stack Overflow instead of to us:
>>>> 
>>>>    http://www.gnu.org/software/octave/bugs.html
>>>> 
>>>> Do you have any suggestions on how to make it clearer how to report
>>>> Octave bugs? When you start Octave, there is a message that tells you
>>>> how to report bugs. How can we make it more likely for you to read
>>>> this message?
>>> 
>>> Thank you so much for you explanation, Jordi. It's quite likely to
>>> read that message, no doubt. However, I think I did not write
>>> perfectly what I mean. When I wrote ``reported this issue at'', I
>>> would like to mean that the details of the problem I face are
>>> described there. I must apologize for my bad words.
>> 
>> It appears I am still not being clear.
>> 
>> You think it's a bug. I agree. The page linked above suggests that
>> 
>>    If Octave produces an error message for valid input, that is a bug.
>> 
>> which seems to be precisely the situation here.
>> 
>> If you think it is a bug, please report it. You may choose not to, but
>> bug reporting is a helpful activity for the Octave project. The point
>> of reporting bugs isn't for them to be fixed for you. It's a courtesy
>> to other users and to developers so that we may have a public record
>> of Octave's shortcomings. Someone may choose to fix that bug, as a
>> further courtesy.
>> 
>> So, let me reiterate: bug reports are very welcome. Is there a way we
>> can make this clearer? We want bug reports, and we want them here:
>> 
>>    http://savannah.gnu.org/bugs/?func=additem&group=octave
>> 
>> So please, please, please, report bugs!
> 
> Please please please first check if this is a duplicate of bug #37290  :-)
> 
> (which has already been solved in the dev sources)
> 
> Philip

The following triggers the "bug" in stable

figure (1)
clf ()
x = 0:10;
plotyy (x, x, x, -x)
plot (x)

But does not in default.

Looking at the changeset ...

     2.1 --- a/scripts/plot/private/__go_draw_axes__.m
     2.2 +++ b/scripts/plot/private/__go_draw_axes__.m
     2.3 @@ -47,7 +47,7 @@
     2.4        else
     2.5          h = axis_obj.__plotyy_axes__;
     2.6          h = h(ishandle (h));
     2.7 -        h = h(isprop (h, "__ploty_axes__"));
     2.8 +        h = h(isprop (h, "__plotyy_axes__"));
     2.9          rmappdata (h, "__plotyy_axes__");
    2.10        endif
    2.11      endif

I don't think the "rmappdata(...)" command is correct.  Is there an appdata 
property named "__plotyy_axes__"?  Is this a second bug?

Ben