Re: '__gnuplot_has_feature__' undefined

[Rik]

On 09/02/2016 10:32 AM, Daniel J Sebald wrote:
> On 09/02/2016 12:20 PM, Mike Miller wrote:
>> On Fri, Sep 02, 2016 at 11:57:08 -0500, Daniel J Sebald wrote:
>>> My thinking is "yes" and "no".  Yes on the bigger items, such as things
>>> that
>>> might cause a v4.0 series gnuplot to fail to interpret properly and
>>> therefore not create a plot.  No on fairly special cases that might be
>>> outside the ordinary type of plot.
>>>
>>> By disallowing older versions of gnuplot, that's ruling out users who are
>>> looking to put in some simple data and create a simple plot (say a line
>>> plot), e.g., the non-power-user class.  Say, for example students
>>> creating a
>>> lab report.
>>
>> That is a reasonable position to take, but the logic must be put into
>> the scripts to support that. It is broken as it is right now and not
>> ready for release IMHO.
>>
>> There is already a set of checks and functions in place for whether
>> gnuplot supports certain features. That needs to be updated to work with
>> the current baseline. There are many places now where Octave sends a
>> command to gnuplot, and gnuplot responds with an arcane (to the Octave
>> user) error message about unsupported syntax or options.
>
> The main one has revolved around gnuplot's "line style" vs. "line type",
> the similarity of syntax leading to confusion.  I've tried fixing a few
> of those.  (I know the waterfall demo still has an issue with some dotted
> lines that should be solid.)

Even a simple plot fails to execute correctly with 4.6.6

plot (1:10, '--')

It doesn't throw an error, but the line is always solid regardless of the
linestyle set in Octave.

--Rik