Re: [Plplot-devel] Comprehensive testing on Cygwin with Tcl/Tk/Itcl/Itk

[Arjen M. <Arj...@de...>]

Hi Alan,



> -----Original Message-----
> From: Alan W. Irwin [mailto:ir...@be...]
> Sent: Sunday, July 05, 2015 10:48 PM
> To: Arjen Markus
> Cc: plp...@li...
> Subject: RE: [Plplot-devel] Comprehensive testing on Cygwin with Tcl/Tk/Itcl/Itk
>
> Hi Arjen:
>
> Thanks for that report.
>
> You said:
>
> > I have attached the result of the comprehensive_test.sh script on
> Cygwin. This time I had an X server running, so that Tk is properly recognised as a
> valid option.
>
> Good.  I hope this part of the X setup is put into your script that sets up and runs
> scripts/comprehensive_test.sh so this issue will not come back to haunt us again.
>
> > The "shared" build succeeds, but the
> "nondynamic" build fails:
>
> There are some lingering but important Cygwin setup issues that you should address
> just as soon as that is convenient for you.
>
> I. As I mentioned before you are running scripts/comprehensive_test.sh with
>
> --cmake_added_options "-DENABLE_octave=OFF -DENABLE_ada=OFF -
> DENABLE_itk=OFF"
>
> Please drop that last option, i.e., use
>
> --cmake_added_options "-DENABLE_octave=OFF -DENABLE_ada=OFF"
>
> instead to test my fix for the (default) -DENABLE_itk=ON case that I implemented
> some time ago.
>
> II. As I mentioned before these tests should be based strictly on Cygwin packages
> rather than something you built and installed in /usr/local at some unknown time in
> the past.  And given a choice, CMake will always prefer to use the /usr/local version.
> So I strongly suggest you completely move /usr/local to somewhere else for the time
> being to make sure you have a pure Cygwin result.
>
Oh dear, I had hidden the Tcl libraries, but not the header files. Will redo that test now that these are also hidden.

> Here are environment variables that are set that involve usr/local.
>
> irwin@raven> grep usr/local comprehensive_test_env.out PATH=/lib/python2.7/site-
> packages:/usr/local/bin:/usr/bin:/cygdrive/c/ProgramData/Oracle/Java/javapath:/cygd
> rive/d/CAF/bin:/cygdrive/c/GTK/bin:/cygdrive/c/tcl/bin:/cygdrive/c/windows/system32:
> /usr/lib/lapack
> INFOPATH=/usr/local/info:/usr/share/info:/usr/info
>
> These environmental variable results (which I assume are setup by default on
> Cygwin) show that /usr/local/bin is on the PATH and /usr/local/info is on the
> INFOPATH. That will be fine once you do the above suggested move of /usr/local to
> somewhere else, but with your present setup with /usr/local containing old results
> you have compiled yourself, these settings for PATH and INFOPATH do
> contaminate your results as seen below.
>
> Here are the CMake results concerning usr/local.
>
> irwin@raven> grep usr/local shared/output_tree/cmake.out
> -- TCL_INCLUDE_PATH = /usr/local/include
> -- FindShapelib: Found shapelib header directory, /usr/local/include, and library,
> /usr/local/lib/libshp.a.
> -- tk_COMPILE_FLAGS = -I"/usr/local/include" -I"/usr/include" -I"/usr/include" -
> I"/cygdrive/d/plplot-svn/plplot-git"/bindings/tcl -I"/cygdrive/d/plplot-
> svn/comprehensive_test_disposeable/shared/build_tree"/bindings/tcl -
> I"/cygdrive/d/plplot-svn/plplot-git"/bindings/tk
> -- ntk_COMPILE_FLAGS = -I"/usr/local/include" -I"/usr/include" -I"/usr/include"
> -- tkwin_COMPILE_FLAGS = -I"/usr/local/include" -I"/usr/include" -I"/usr/include" -
> I"/cygdrive/d/plplot-svn/plplot-git"/bindings/tcl -I"/cygdrive/d/plplot-
> svn/comprehensive_test_disposeable/shared/build_tree"/bindings/tcl -
> I"/cygdrive/d/plplot-svn/plplot-git"/bindings/tk-x-plat -I"/cygdrive/d/plplot-svn/plplot-
> git"/bindings/tk
> libplplot_LINK_FLAGS =
> /usr/lib/libltdl.dll.a;/usr/lib/libdl.a;/usr/local/lib/libshp.a;/usr/lib/libfreetype.dll.a;-
> lcsirocsa;-lcsironn;-lqhull;-lqsastime
>
> I assume all those /usr/local results being found by CMake will be replaced by the
> correct pure Cygwin results once you do the suggested renaming of /usr/local.  But,
> of course, you should check shared/output_tree/cmake.out (generated very early in
> the test) to make sure there are no warnings about missing components of PLplot
> that correspond to a Cygwin package (e.g., the development package for
> shapelib) that is not installed yet.
>
> Please run another noninteractive test with the above issues completely addressed
> as soon as is convenient.  You may find the undefined-symbol issue you discovered
> for this test completely disappears for that further test because there is no longer any
> chance for inconsistency between pure Cygwin packages and /usr/local results you
> have built yourself.
>
Hm, it seems that the shapelib library was installed under /usr/local, as well as the LibLASi static library. I did not compile them myself.

I have renamed the original /usr/local and created an empty one now, let us see what this brings.

Regards,

Arjen

DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.