strange behavior

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

strange behavior

Nitecki, Zbigniew H.
I’m mystified by a phenomenon that has been an irritant while creating a particular document, illustrated by the attached files.
The only difference between the two source files is on line 80: in version (a), the page break is commented out, and in version (b) it is present.
In the pdf for version (a), the text coming after the figure is run together (as if I had failed to exit math mode, except that the font is not the math mode font) but when I introduce a page break, the problem disappears.
In the context in which I am running into this, the page break is not a problem, but I am curious what it is about the figure that creates the problem.
I have left the full figure, which invokes the pst-ode macro/program for plotting solutions of a system of two linear (homogeneous) o.d.e.and in fact when I comment out lines 35-52 and 56-67 (which leaves only the commands for drawing the axes) the problem also goes away (even without a page break).

Any idea what is going on, and if possible, how to fix it without invoking a page break?

Zbigniew Nitecki
Department of Mathematics
Tufts University
Medford, MA 02155

telephones:
Office    (617)627-3843
Dept.    (617)627-3234
Dept. fax    (617)627-3966
http://www.tufts.edu/~znitecki/






----------- Please Consult the Following Before Posting -----------
TeX FAQ: http://www.tex.ac.uk/faq
List Reminders and Etiquette: https://sites.esm.psu.edu/~gray/tex/
List Archives: http://dir.gmane.org/gmane.comp.tex.macosx
                https://email.esm.psu.edu/pipermail/macosx-tex/
TeX on Mac OS X Website: http://mactex-wiki.tug.org/
List Info: https://email.esm.psu.edu/mailman/listinfo/macosx-tex

Test_of_figure(a).tex (4K) Download Attachment
Test_of_figure(a).pdf (500K) Download Attachment
Test_of_figure(b).tex (4K) Download Attachment
Test_of_figure(b).pdf (501K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: strange behavior

Axel E. Retif-4
On 5/21/20 12:31 PM, Nitecki, Zbigniew H. wrote:

> Any idea what is going on, and if possible, how to fix it without
> invoking a page break?

The problem seems to be with Ghostscript.

If I typeset your Test_of_figure(a) in a terminal and then convert the
DVI to PDF with dvipdf I get something similar to your output (see
Test_figure->dvipdf); but if I open the DVI with Okular the DVI is OK
(see Test_figure_DVI-on-Okular). Also note the right spacing in the formula.

If I convert the DVI to PS with dvips and open the PS in a text editor I
see (line 6773)

      12 192 a 112 275 a Fe(Some)30 b(text)p 0 TeXcolorgray

and (lines 13945 and 13946)

      Fe(=)f(2)p 0 TeXcolorgray 0 TeXcolorgray 112 2245 a(And)k(no)m(w)i
      (some)f(new)g(text.)p 0 TeXcolorgray 1779 4912 a Fa(1)p



You can view the DVI with xdvi and text is OK, but you can't see the figure.

I'm not using a Mac, but Ubuntu Mate 19.10 with TeXLive 2019 (I haven't
installed TL 2020 yet) and Ghostscript 9.27.


Best

Axel

----------- Please Consult the Following Before Posting -----------
TeX FAQ: http://www.tex.ac.uk/faq
List Reminders and Etiquette: https://sites.esm.psu.edu/~gray/tex/
List Archives: http://dir.gmane.org/gmane.comp.tex.macosx
                https://email.esm.psu.edu/pipermail/macosx-tex/
TeX on Mac OS X Website: http://mactex-wiki.tug.org/
List Info: https://email.esm.psu.edu/mailman/listinfo/macosx-tex

Test_figure->dvipdf.png (29K) Download Attachment
Test_figure_DVI-on-Okular.png (11K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: strange behavior

Axel E. Retif-4
On 5/22/20 2:41 AM, Axel E. Retif wrote:

> The problem seems to be with Ghostscript.

[...]

> but if I open the DVI with Okular the DVI is OK

I read here

     https://tug.org/PSTricks/main.cgi?file=pdf/pdfoutput

that «dvipdf is no alternative, because it cannot handle the PostScript
prologue files. Use always dvips and then ps2pdf».

So I did that, first commenting out

     % \usepackage{epstopdf}
     % \DeclareGraphicsRule{.tif}{png}{.png}{`convert #1 `dirname
#1`/`basename #1 .tif`.png}

because we're not using pdfLaTex, but with the same bad result, as it is
using also GhostScript (see Test_of_figure_latex->dvips->ps2pdf).

But I also read on the Web that XeLaTeX can do the job, and indeed it
does! (See Test_of_figure_xelatex.)

It takes several seconds with lots of

     pstODEsolve (RKF45),
     -/+ failed/successful step, "o" output step, "!" step size
underflow (stop)

but it seems to be OK.


Best

Axel

----------- Please Consult the Following Before Posting -----------
TeX FAQ: http://www.tex.ac.uk/faq
List Reminders and Etiquette: https://sites.esm.psu.edu/~gray/tex/
List Archives: http://dir.gmane.org/gmane.comp.tex.macosx
                https://email.esm.psu.edu/pipermail/macosx-tex/
TeX on Mac OS X Website: http://mactex-wiki.tug.org/
List Info: https://email.esm.psu.edu/mailman/listinfo/macosx-tex

Test_of_figure_latex->dvips->ps2pdf.pdf (503K) Download Attachment
Test_of_figure_xelatex.pdf (502K) Download Attachment