Re: [Plplot-devel] Mac OS/X PLPLOT installation

[Koen van der Drift <kvddrift@ea...>]

On Aug 19, 2004, at 1:15 AM, Alan W. Irwin wrote:
> Could you give us the result of running the following command after you
> have built a copy of the installed examples:
>
> /tmp/examples/c/x10c -debug -dev xwin
>
>

D'oh! I should have started the command from an XWindow, not the 
terminal.

I will commit plplot 5.3.1 to fink later today or tomorrow.



- Koen.

Re: [Plplot-devel] PDL interface to PLplot

[Rafael Laboissiere <rlaboiss@us...>]

* Alan W. Irwin <irwin@...> [2004-08-20 17:42]:

> This took a half hour rather than a "few minutes" because I encountered a
> stubborn syntax problem with cf/pdl.ac.  I want nothing to happen if
> enable_pdl is disabled, but configure wouldn't work until I put a useless
> executable statement in what previously was an empty if block.  Does bash
> not allow empty if blocks?

No, Bourne shells (including bash) do not accept empty if blocks.

> Rafael, if there is a more elegant way to have an empty if block, please
> change cf/pdl.ac appropriately.

I fixed it in CVS by inverting the test from "= no" to "= yes".  Very
simple.

-- 
Rafael

Re: [Plplot-devel] make check python woes for separated build tree

[Alan W. Irwin <irwin@be...>]

On 2004-07-15 19:05-0700 Alan W. Irwin wrote:

> I have been spending a lot of time trying to get make check to
> work for python in the separated build tree case, and I cannot get it to work
> in general.

> [...] The only other alternative I can see is to copy all the scripts from the
> source tree to the build tree, but that solution seems pretty ugly to me.

Actually, I have now rethought this, and I think this is the correct choice
because the alternative I was looking at for executing the scripts in the
source tree contradicts all autotool assumptions and is bound to be difficult
to maintain even if I could ever get it to work.

Therefore, I need the x?? and xw??.py scripts to always be available in the
build tree so I can execute make check in the build tree for the python
examples.

So what is needed is a rule to test whether the x?? and xw??.py scripts
exist in the build tree examples/python directory and if not (this only
occurs for a clean build tree that is separated from the source tree)
establish symlinks for the source tree x?? and xw??.py scripts within the
build tree.

Rafael, I am thinking of using similar logic to what is done for
examples/tk/Makefile.am for the x01.tcl: target.  But instead of removing
the symbolic link each time before reestablishing it inside the loop over
file (not a good thing to do if the build tree is the source tree), I plan
to use if test ! -e $$file to test for the non-existence of the symbolic
link (or actual file if source tree = build tree).  Is this the way to go,
or would you recommend some other approach?

Come to think of it, if you like the if test ! -e $$file approach, we could
also use it for examples/tk/Makefile.am.

Alan
__________________________
Alan W. Irwin
email: irwin@...
phone: 250-727-2902

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the PLplot scientific plotting software
package (plplot.org), the Yorick front-end to PLplot (yplot.sf.net), the
Loads of Linux Links project (loll.sf.net), and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-devel] PDL interface to PLplot

[Alan W. Irwin <irwin@be...>]

On 2004-08-21 00:26+0200 Rafael Laboissiere wrote:

> * Alan W. Irwin <irwin@...> [2004-08-20 14:28]:
>
>> This is so short let's do it now rather than wasting time putting it on
>> some ToDo list.  All that is really needed from you is a decision about
>> whether to use the "enable" or "with" form of parameter.  Once that decision
>> is made, I can easily do the rest in a few minutes following what is done
>> elsewhere in the cf/* files for either PL_ARG_ENABLE or PL_ARG_WITH.
>
> According to the Autoconf documentation, "enable"will be more appropriate in
> this case.

OK.  Thanks for that decision.  I have implemented the enable-pdl version,
minimally tested it (works with build-tree=source tree and installed
examples), and committed it.

This took a half hour rather than a "few minutes" because I encountered a
stubborn syntax problem with cf/pdl.ac.  I want nothing to happen if
enable_pdl is disabled, but configure wouldn't work until I put a useless
executable statement in what previously was an empty if block.  Does bash
not allow empty if blocks?

Rafael, if there is a more elegant way to have an empty if block, please
change cf/pdl.ac appropriately.

Alan
__________________________
Alan W. Irwin
email: irwin@...
phone: 250-727-2902

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the PLplot scientific plotting software
package (plplot.org), the Yorick front-end to PLplot (yplot.sf.net), the
Loads of Linux Links project (loll.sf.net), and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________