Re: wait_for_file ?? [print via pipes - help testing on Windows?]

[bpabbott]

On Sep 17, 2010, at 05:17 AM, Marco Atzeri <address@hidden> wrote:

--- Ven 17/9/10, Michael D Godfrey <address@hidden> ha scritto:

Da: Michael D Godfrey <address@hidden>
Oggetto: Re: wait_for_file ?? [print via pipes - help testing on Windows?]
A: "Ben Abbott" <address@hidden>
Cc: "octave maintainers mailing list" <address@hidden>
Data: Venerdì 17 settembre 2010, 08:16

On 09/16/2010 06:58 PM, Ben Abbott wrote:

I no longer have any errors with testing the printing on MacOS 10.6.4.

For complete functionality it is necessary to have pstoedit, fig2dev (from TransFig), and epstool installed.

I'm unable to do any testing for Windows. So it would be appreciated if a Windows user can test this changeset.

There's been some recent changesets that have effected my results, so I suggest pulling the current sources, and a build fresh.

My tip is ...

	http://hg.savannah.gnu.org/hgweb/octave/rev/88687577519f

Ben

After installs of epstool and pstoedit I get:
octave:1> test_printing_20100916
warning: function name `test_printing' does not agree with function file name `/data/d/src/octave/hg/test_printing_20100916.m'
Xlib:  extension "NV-GLX" missing on display ":1175.0".
backend is fltk
print fltk-aifm  -daifm
error: fltk_backend: filename should be fid
warning: print.m: No such file or directory, '/tmp/oct-f9T0RH.eps'.
error: called from:
error:   /data/d/src/octave/hg/__fltk_print__.m at line 193, column 7
error:   /data/d/src/octave/hg/__fltk_print__.m at line 121, column 14
error:   /d/src/octave/hg/octave/scripts/plot/printm at line 344, column 12
error:   /data/d/src/octave/hg/test_printing_20100916.m at line 65, column 9
========================================
This is current build and your changeset.patch of today.

Michael

same on cygwin

octave:3> source ./test_printing.m
octave:4> test_printing
backend is fltk
print fltk-aifm  -daifm
error: fltk_backend: filename should be fid
error: called from:
error:   /pub/hg/octave_build/../octave/scripts/plot/__fltk_print__.m at line 98, column 5
warning: print.m: No such file or directory, '/tmp/oct-22eccc.0.eps'.
error:   /pub/hg/octave_build/../octave/scripts/plot/print.m at line 338, column 6
error:   test_printing at line 65, column 9

 hg tip
changeset:   10992:88687577519f
tag:         tip
user:        Ben Abbott <address@hidden>
date:        Thu Sep 16 20:30:01 2010 -0400
summary:     __go_draw_axes__.m: Ensure text objects have units of "data".

Marco

Thanks for the quick response.

Unfortunately, I should have been more clear. The changeset attached to the prior email has not been pushed. You'll need to patch your sources with that changeset.

After applying the changeset __fltk_print__ will include the lines below. Since line 98 is part of the fig2dev_devices and not "aifm", I assume that the sources have not been patched.

 81   case fig2dev_devices

 82     cmd_pstoedit = opts.pstoedit_cmd (opts, "fig");

 83     cmd_fig2dev = opts.fig2dev_cmd (opts, opts.devopt);

 84     if (strcmp (opts.devopt, "pstex"))

 85       [~, ~, ext] = fileparts (opts.name);

 86       if (any (strcmpi (ext, {".ps", ".tex", "."})))

 87         opts.name = opts.name(1:end-numel(ext));

 88       endif

 89       opts.name = strcat (opts.name, ".ps");

 90       cmd = sprintf ("%s | %s > %s", cmd_pstoedit, cmd_fig2dev, opts.name);

 91       gl2ps_device = {"eps"};

 92       pipeline = {cmd};

 93       cmd_fig2dev = opts.fig2dev_cmd (opts, "pstex_t");

 94       gl2ps_device{2} = "eps";

 95       pipeline{2} = sprintf ("%s | %s > %s", cmd_pstoedit,

 96                              cmd_fig2dev, strrep(opts.name, ".ps", ".tex"));

 97     

 98       cmd = sprintf ("%s | %s > %s", cmd_pstoedit, cmd_fig2dev, opts.name);

 99       gl2ps_device = {"eps"};

100       pipeline = {cmd};

101     endif

102   case {"aifm"}

103     cmd = opts.pstoedit_cmd (opts, "ps2ai");

104     gl2ps_device = {"eps"};

105     pipeline = {sprintf("%s > %s", cmd, opts.name)};

106   case {"dxf", "emf", "fig", "hpgl"}

Ben