Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.graphics.apps.gnuplot > #2757

Re: plimsoll / standard enthalpy of formation

Show key headers only | View raw

Jörg Buchholz wrote:

> On 18.02.2015 05:16, Ethan A Merritt wrote:
>> Jörg Buchholz wrote:
>>
>>> On 17.02.2015 06:10, Ethan A Merritt wrote:
>>>> Jörg Buchholz wrote:
>>>>
>>>>> On 14.02.2015 20:47, Ethan A Merritt wrote:
>>>>>> Jörg Buchholz wrote:
>>>>>>
>>>>>>> On 14.02.2015 04:38, Ethan A Merritt wrote:
>>>>>>>> Jörg Buchholz wrote:
>>>>>>>>
>>>>>>>>> Hello,
>>>>>>>>>
>>>>>>>>> I will set a label and/or a xlabel with a "plimsoll" at
>>>>>>>>> superscript character.
>>>>>>>>> In LaTeX, with "\usepackage{chemmacros}", you can do it with
>>>>>>>>> something like this:
>>>>>>>>>
>>>>>>>>> $\Delta H_{\mathrm{f}}^{\standardstate}$
>>>>>>>>>
>>>>>>>>> For the term postscript enhanced I found two workarounds.
>>>>>>>>>
>>>>>>>>> set xlabel "{/Symbol D}H@^{~\261__o}_f"
>>>>>>>>>
>>>>>>>>> lx=0
>>>>>>>>> ly=-0.0
>>>>>>>>> set label 1 '{/Symbol D}H_f' at lx,ly font ",14" right
>>>>>>>>> set label 2 '{/Symbol=6 \306}' at lx,ly rotate by -45 offset
>>>>>>>>> char 0,0.6 left
>>>>>>>>>
>>>>>>>>> The first one (xlabel) did not work with term pdfcairo enhanced
>>>>>>>>> and the second one (label) needs some other offsets.
>>>>>>>>
>>>>>>>> All of the cairo terminals, and pretty much all terminals newer
>>>>>>>> than
>>>>>>>> PostScript, work best with UTF8 encoding.  No need to change the
>>>>>>>> font to /Symbol, no need to construct special positioning.
>>>>>>>> Latex is another option of course.
>>>>>>>>
>>>>>>>>       set encoding utf8
>>>>>>>>       set label 1  "ΔH_{f^⦵}"
>>>>>>>>
>>>>>>>
>>>>>>> Thanks a lot. Is it the U+29B5 sign?
>>>>>>
>>>>>> Yes.
>>>>>
>>>>> Is there a way to type it in directly into the gnuplot-console? I
>>>>> can load it from a file but can not type it in directly.
>>>>
>>>> I think this is entirely up the the local configuration of your
>>>> desktop environment.
>>>>
>>>>> If I load it from a file than "show xlabel" shows me the following:
>>>>> xlabel is "ΔH_{f}^{⦵}"
>>>>>
>>>>> If I copy only the U+29B5 sign and paste it into the console I get a
>>>>> "\U+FFE2\U+FFA6\U+FFB5" and "show xlabel" gives me xlabel is "".
>>>>> "show encoding nominal character encoding is utf8
>>>>> however LC_CTYPE in current locale is de_DE.UTF-8"
>>>>
>>>> The "set/show encoding" command in gnuplot controls how gnuplot
>>>> handles the output.  It does not affect input.
>>>>
>>>>> What goes wrong? It is a Debian testing with KDE. I can copy the ⦵
>>>>> into the Terminal (bash) but not into the gnuplot-console.
>>>>
>>>> That is very strange.  I would have thought that any input method
>>>> that worked for your terminal shell prompt would also work for an
>>>> instance of gnuplot running in that same terminal.
>>>
>>> I tested a little bit. The gnuplot-console gives to me the
>>> hexadecimal-code of every non standard character. Some examples:
>>>
>>> ä gives me \U+FFC3\U+FFA4 c3 a4 is the hex-code for U+00E4 (I have the
>>> ä on my Keybord)
>>> ⦵ gives me \U+FFE2\U+FFA6\U+FFB5 e2 a6 b5 is the hex-code for U+29B5
>>>
>>> It is an on all Terminals (LXterminal, gnome-terminal,
>>> standard-KDE-terminal) the same, on Debian stable and testing with
>>> KDE. And all Terminals can display every unicode sign.
>>>
>>> On Fedora 21 it works "normal", there I can use non standard
>>> characters inside the gnuplot-console.
>>>
>>> Is there a config-file?
>>
>> A config file for what, exactly?
>  >
>> The only thing I can think of is ~/.inputrc and/or /etc/inputrc
>>
>> I've never touched these files, but they are mentioned in "man readline".
>> If you have these files at all, try renaming them temporarily to see
>> if that improves the behavior.  You might have to start a fresh session.
> 
> I compiled gnuplot 5.0 from source. After this I have the same
> functionality in the terminal and in the gnuplot-console.
> 
> So it looks like the other effects comes from the "Debian gnuplot package"

Your finding is interesting, but I am not sure that readline is the 
operative difference here.

To test, I built gnuplot 5.0 from source after configuring using 
   ./configure --with-readline-builtin

For me UTF8 input works the same with the builtin version of readline
as it does with gnu readline.  So I am not 100% convinced that this is
the true source of the strange behavior you saw, at least not by
itself.  I suspect an interaction with some additional layer of the
input stream processing. uim? ncurses? anthy? 
But in any case it's good that you have it working now.

> 
> The /usr/share/doc/gnuplot/README.Debian says:
> 
> "libreadline
> -----------
> 
> Yes, the built in readline of gnuplot is bad. However, libreadline
> cannot be used instead because it is licensed under the GPL, whereas
> gnuplot has special licenses (patches only). Distribution of those
> programs linked together is legally impossible but you may rebuild
> your own custom package with readline. Please don't file bugs telling
> me to use libreadline in gnuplot..."

For what it's worth,  I disagree with the quoted statement above that
there would be a legal restriction on distributing a gnuplot binary
that dynamically linked to use gnu libreadline.  Not that my opinion is
likely to sway Debian packaging policy.  Fortunately the builtin
readline is not bad at all (modulo this current strangeness).
Its primary deficiency relative to gnu libreadline is that it does
a poor job of editing input lines that are longer than the terminal
display width.

	Ethan

Back to comp.graphics.apps.gnuplot | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-12 08:41 +0100
  Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-12 20:37 +0100
  Re: plimsoll / standard enthalpy of formation Ethan A Merritt <EAMerritt@gmail.com> - 2015-02-13 19:38 -0800
    Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-14 19:57 +0100
      Re: plimsoll / standard enthalpy of formation Ethan A Merritt <EAMerritt@gmail.com> - 2015-02-14 11:47 -0800
        Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-16 20:25 +0100
          Re: plimsoll / standard enthalpy of formation Ethan A Merritt <EAMerritt@gmail.com> - 2015-02-16 21:10 -0800
            Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-17 21:13 +0100
              Re: plimsoll / standard enthalpy of formation Ethan A Merritt <EAMerritt@gmail.com> - 2015-02-17 20:16 -0800
                Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-18 20:33 +0100
                Re: plimsoll / standard enthalpy of formation Ethan A Merritt <sfeam@users.sourceforge.net> - 2015-02-18 16:01 -0800
                Re: plimsoll / standard enthalpy of formation Jörg Buchholz <bookwood4news@freenet.de> - 2015-02-19 19:47 +0100

csiph-web