Re: Shrinked axes

[mmuetzel]

Pantxo wrote
> mmuetzel wrote
>> On 11 Oct 2019 12:00:09 +0300, Valdas Jankūnas wrote:
>> 
>> I see the same in Octave 5.1.0 and Octave 4.4.1. So it doesn't look like
>> a
>> recent regression.
>> I also think this is a bug. Can you please open a report on savannah?
>> https://savannah.gnu.org/bugs/?group=octave
>> 
>> Also setting the outerposition in pixel units seems to be off if the
>> figure is larger.
>> It looks like the area that is accounted for the labels and title is
>> proportional to the figure extent instead of that of the outerpostion.
>> 
>> Fwiw, the axes have the expected dimensions in Matlab with your original
>> example.
>> 
>> Markus
> 
> I don't think this is a bug per se, but rather a consequence of the way we
> impose a minimal margin around axes boxes. Since the "looseinset" property
> also exists as a hidden property in Matlab (or it used to), I expect the
> very same result there.
> 
> Pantxo  

I tried the example in Matlab R2019b and the results are like the OP
expected:
The axes in figure 2 have the same size as in figure 1.
The same result with Matlab R2016a and R2012a.
The results in Octave are different.

Markus




--
Sent from: https://octave.1599824.n4.nabble.com/Octave-General-f1599825.html