Re: Re: org-babel-gnuplot broken today?

[Nick Dokos <nicholas.dokos@hp.com>]

Eric Schulte <schulte.eric@gmail.com> wrote:

> Hi Nigel,
> 
> Org-mode gnuplot code blocks do make use of user variables [1] for
> passing arguments into gnuplot.  I believe that this feature may only be
> available in later versions of gnuplot.  I'm using gnuplot 4.4 locally
> and I've had no problems using gnuplot from Org-mode code blocks which I
> do most every day.
> 

Did you try the example that Nigel posted? I think babel *is* broken.
I tried a bisect and can verify that

commit f16c46c856afbd41b115a6a50306c0c002cdb333 is good (release_7.01h-750-gf16c46c)
commit b664510ec4811491e4611791c24591153abdf5f3 is bad  (release_7.01h-800-gb664510)

but the sequence is not bisectable: when it tries the middle commit
2699f4e9bc7695432cf6aaf1dd48e182b10c2847 (release_7.01h-775-g2699f4e)
it gets "Symbol's value as variable is void: result-type" and it's
not clear to me how to proceed with the bisection at that point. I tried
skipping a bit, but the results were mixed at best.

So it looks to me as if something in those 50 commits broke it.

HTH,
Nick


> the syntax for defining variables is fairly simple, try the following
> directly in the gnuplot terminal, if your gnuplot does have variable
> support it should graph a straight line of y=3D10
> #+begin_src gnuplot
>   a=3D10
>   plot a
> #+end_src
> 
> Best -- Eric
> 
> Nigel Beck <me@nigelbeck.com> writes:
> 
> > John - thanks for the quick reply (and, btw, the very useful tutorial).
> >
> > I ought to have noted that the example works fine when invoked directly
> > from the command line, with the only change being that I move the data
> > into a file (mainly because I don't know the correct gnuplot syntax for
> > putting table data into a variable!).
> >
> > I haven't updated gnuplot since installing it, so I believe the org-mode
> > <-> gnuplot unfriendliness lies in some effect that a recent pull of
> > 7.01trans has had. (I did updated emacs-w3m from cvs the other day, so I
> > suppose thats another possible source of error, but it seems unlikely).
> >
> > I agree that it appears that the table data being handed to gnuplot
> > from org-mode is somehow being formatted in a way that gnuplot does not u=
> nderstand.
> >
> > John Hendy <jw.hendy@gmail.com> writes:
> >
> >> On Tue, Oct 26, 2010 at 3:10 PM, Nigel Beck <me@nigelbeck.com> wrote:
> >>
> >>     I refreshed today 7.01trans and noticed I couldn't get my gnuplots to
> >>     plot anymore: data from a table within the org file is no longer
> >>     digested nicely by gnuplot.
> >>
> >> That's odd. I wrote the tutorial and thus my heart skipped a beat for a =
> moment!
> >> =C2=A0
> >>
> >>     To test, I used the snippet below from the worg
> >>     http://orgmode.org/worg/org-contrib/babel/languages/ob-doc-gnuplot.p=
> hp:
> >>
> >>     ** Data Table Plot
> >>=20=20=20=20
> >>     Plotting data points from a table could look like this:
> >>     #+tblname: basic-plot
> >>
> >> ---- 8< ----
> >> =C2=A0
> >>
> >>     Exporting or executing the gnuplot piece, Gnuplot (Gnuplot 4.4 patch
> >>     =C2=A0level 1) complains:
> >>=20=20=20=20
> >>     gnuplot> plot data u 1:2 w p lw 2 title 'x vs. y1', \
> >>
> >> ---- 8< ----
> >> =C2=A0
> >>
> >>     =C2=A0 =C2=A0 =C2=A0 =C2=A0 No data in plot
> >>
> >>     So of course no basic-plot.png is created....
> >>=20=20=20=20
> >>     As far as I know, my gnuplots were happy sometime within the past we=
> ek
> >>     refreshing org-mode from git basically daily...
> >>
> >> That's odd indeed. I copied and pasted your exact table and code from th=
> e email and was able to generate the plot. The error is
> >> quite odd. It almost makes me wonder if something is different about how=
>  gnuplot is receiving the data. Like a syntax error or
> >> something. I would say that at this point it's not the table or code, pe=
> r se, since I can generate the plot. Hopefully someone will
> >> chime in with more experience who may have seen this error?
> >>
> >> One thing to try in the meantime... Gather up some gnuplot test data fil=
> e and try to run it from the command line perhaps? It would
> >> at least be one way to see if it's working outside of org-mode?
> >>
> >> Best regards,
> >> John
> >> =C2=A0
> >>
> >>     _______________________________________________
> >>     Emacs-orgmode mailing list
> >>     Please use `Reply All' to send replies to the list.
> >>     Emacs-orgmode@gnu.org
> >>     http://lists.gnu.org/mailman/listinfo/emacs-orgmode
> >>
> >> _______________________________________________
> >> Emacs-orgmode mailing list
> >> Please use `Reply All' to send replies to the list.
> >> Emacs-orgmode@gnu.org
> >> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
> 
> Footnotes:=20
> [1]  http://www.gnuplot.info/docs/gnuplot.html#x1-3100013.4
> 
> 
> _______________________________________________
> Emacs-orgmode mailing list
> Please use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
>