octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: terminals_with_size


From: Rik
Subject: Re: terminals_with_size
Date: Wed, 14 Sep 2016 18:37:34 -0700

On 09/14/2016 06:00 PM, Rik wrote:
> On 09/14/2016 05:44 PM, Tatsuro MATSUOKA wrote:
>> ----- Original Message -----
>>
>>> From: Rik 
>>> To: John W. Eaton ; Mike Miller 
>>> Cc: Octave-Maintainers 
>>> Date: 2016/9/15, Thu 05:20
>>> Subject: Re: Minimum gnuplot version for Octave 4.2?
>>>
>>> On 09/14/2016 11:16 AM, John W. Eaton wrote:
>>>>  On 09/14/2016 01:26 PM, Mike Miller wrote:
>>>>>  On Wed, Sep 14, 2016 at 09:52:14 -0700, Rik wrote:
>>>>>>  Is there any objection to requiring a minimum gnuplot version of 
>>> 4.4 for
>>>>>>  the next release?  The release date was March 2010, or 6.5 years 
>>> ago.  The
>>>>>>  4.2 series had a release date of April 2007, but I don't think 
>>> anyone has
>>>>>>  tried to use Octave with gnuplot 4.2.X and I'm not really sure 
>>> it would
>>>>>>  work anymore.  In our own README.gnuplot we say, "Octave works 
>>> best with
>>>>>>  gnuplot 4.4 or higher, which is available from 
>>> http://www.gnuplot.info.";
>>>>>>  I happened to notice in the documentation for image() that we still 
>>> have
>>>>>>  "*Warning:* X and Y are ignored when using gnuplot 4.0 or 
>>> earlier." which
>>>>>>  is horrendously out of date.
>>>>>  None here.
>>>>>
>>>>>  RHEL/CentOS 5 has gnuplot 4.0 and RHEL/CentOS 6 has gnuplot 4.2. All
>>>>>  other current supported versions of distros that I know of have 4.4 or
>>>>>  later. I think it's safe to drop checks and warnings messages for 
>>> any
>>>>>  version lower than 4.4. Gnuplot should be very easy to build on older
>>>>>  GNU or UNIX distros that may need a newer version.
>>>>  It's OK with me as well.
>>>>
>>>>  jwe
>>>>
>>> Okay, done in this cset
>>> (http://hg.savannah.gnu.org/hgweb/octave/rev/a666e3ee6af8).
>>>
>>> --Rik
>> In the above changeset.
>>
>> --- a/scripts/plot/util/__gnuplot_drawnow__.m
>> +++ b/scripts/plot/util/__gnuplot_drawnow__.m
>> @@ -193,16 +193,8 @@
>>                                   "epscairo", "epslatex", "fig", "gif", ...
>>                                   "jpeg", "latex", "pbm", "pdf", "pdfcairo", 
>> ...
>>                                   "postscript", "png", "pngcairo", "pstex", 
>> ...
>> -                                 "pslatex", "svg", "tikz"};
>> -          if (__gnuplot_has_feature__ ("windows_figure_position"))
>> -            terminals_with_size{end+1} = "windows";
>> -          endif
>> -          if (__gnuplot_has_feature__ ("x11_figure_position"))
>> -            terminals_with_size{end+1} = "x11";
>> -          endif
>> -          if (__gnuplot_has_feature__ ("wxt_figure_size"))
>> -            terminals_with_size{end+1} = "wxt";
>> -          endif
>> +                                 "pslatex", "svg", "tikz", "windows", ...
>> +                                 "wxt", "x11"};
>>  
>>
>> Attached code
>>         if (all (gnuplot_size > 0))
>>           terminals_with_size = {"canvas", "cairolatex", "eepic", "emf", ...
>>                                  "epscairo", "epslatex", "fig", "gif", ...
>>                                  "jpeg", "latex", "pbm", "pdf", "pdfcairo", 
>> ...
>>                                  "postscript", "png", "pngcairo", "pstex", 
>> ...
>>                                  "pslatex", "svg", "tikz", "windows", ...
>>                                  "wxt", "x11"};
>>
>>
>>
>>
>> I do not see "qt". 
>>
>> The qt terminal is implemented from gnuplot 4.6 and default terminal from 
>> gnuplot 5.0.
> Is qt a terminal_with_size?  I don't know.  Someone who understands gnuplot
> and Octave can chime in on whether it should be added to the list.
>
> --Rik

I read through the gnuplot documentation and checked all the terminals and
certainly qt has the size attribute.  I added a new feature "qt_terminal"
in __gnuplot_has_feature__ and if "qt" is available it is added to the list
of terminals_with_size.  I can't push right now because the repository has
gone down again (second time today).

--Rik



reply via email to

[Prev in Thread] Current Thread [Next in Thread]