Hi Alan
I'm making slow but steady progress between tidying up after the kids as the
evening wears on. I am doing a full install at the moment just so I can follow
things through. I have located the DLLs on the lib directory and added their
directory to my path, but still they don't seem to be loading. Not sure why.
Running through the debugger so I can see the loadlibrary call, but it just
returns null.
By the way while searching for solutions I found a number of security warnings
against loading a dll just by its , with no path. It is easy for someone to add
a spoof dll to the system that will be loaded and executed instead. Not sure
how real any threat really would be to a plplot user though.
Phil
-----Original Message-----
From: "Alan W. Irwin" <ir...@beluga.phys.uvic.ca>
Sent: 22/09/2014 19:49
To: "Phil Rosenberg" <philip_rosenb...@yahoo.com>
Cc: "PLplot development list" <Plplot-devel@lists.sourceforge.net>
Subject: RE: [Plplot-devel] tcl build problem
On 2014-09-22 19:10+0100 Phil Rosenberg wrote:
> Just realized I only replied to Alan, not the list - see below if you are
> interested.
>
> Alan I added my bin directory to the PATH.
Hi Phil:
See my recent off-list comment about that. For build-tree tests
(which I think is what you are doing), the dll subdirectory within the
build tree must be on your PATH, not some bin directory.
> The examples now run, giving me a list of drivers, however, when I
select a driver, plplot exits saying it cannot load the driver.
Suggestions from any of our Windows users welcome as I'm not used to
using dlls. I wonder if Plplot is trying to load dlls that don't
exist. E.g. I probably don't have wxWidgets dlls built, just static
Libs. Although even if I select svg, which has no external
dependencies I get the same error.
Have you run the "all" target in the build tree? One of the broad
results of running that target is all your built dll's (including the
plplot dll and the svn dll) should be located in the dll subdirectory
in that build tree, and the svg dll should be dynamically loaded without
issues by the examples if that dll subdirectory is on your PATH.
If for some reason, the all target does not automatically collect all
dll's in the dll subdirectory for your visual studio generator (like
it does for our better-tested make-based and nmake based generators),
then for now until we can address that issue, you should copy all
dll's created in the build tree into that dll subdirectory by hand,
put that subdirectory on your PATH, and then see whether the dynamic
load of the svg device works properly.
Alan
>
> Phil
>
> -----Original Message-----
> From: "Phil Rosenberg" <philip_rosenb...@yahoo.com>
> Sent: 22/09/2014 17:44
> To: "Alan W. Irwin" <ir...@beluga.phys.uvic.ca>
> Subject: RE: [Plplot-devel] tcl build problem
>
> Hi Alan
> I have started this testing. It all works fine with my standard static lib,
> static runtime visual studio, cmake 3 setup.
> However if I try dynamic runtime with dynamic Libs then I get errors for all
> the test_xxx_dyndriver projects. The errors look like:
> Generating test_dyndrivers_dir/xxx.driver_info
> Could not open driver module xxx
> CUSTOMBUILD : libltdl error : No error information
>
> The test_dyndrovers_dir is full of 0 byte files and if I try to run an
> example I get a popup saying plplot.dll is missing from your system. Although
> I do have the dlls in my bin directory. Maybe it is just a case of adding the
> bin directory to my path.
>
> I will keep you updated
>
> Phil
>
>
> From: Alan W. Irwin
> Sent: 05/09/2014 12:40
> To: phil rosenberg
> Cc: PLplot development list
> Subject: Re: [Plplot-devel] tcl build problem
>
>
> On 2014-09-04 13:51-0700 Alan W. Irwin wrote:
>
>> On 2014-09-01 11:08-0700 Alan W. Irwin wrote:
>>
>>> I now plan to do a comprehensive run-time test using
>>> scripts/comprehensive_test.sh for the same Tcl/Tk/Itcl/Itk install
>>> locations, and such a comprehensive test (involving run-time tests of
>>> both the build tree and install tree for our 3 major build
>>> configurations) will almost guarantee finding more changes that have
>>> to be done for our three different build systems (core build +
>>> CMake-based and traditional build systems for the installed examples).
>>
>> That prediction was spot on, and over the last few days I have been
>> plugging away at all the issues that were found by that comprehensive
>> testing. Today I finally got a breakthrough on what had been an
>> intractable problem (dealing with blanks in pathnames for the Makefile
>> + pkg-config traditional build system for the installed examples).
>>
>> Propagating that solution from one of our libraries to the rest
>> requires extensive and non-trivial build-system changes that all need
>> to be comprehensively tested, but I think the end is finally in sight,
>> so I might even get this pushed by late Friday.
>
> Hi Phil:
>
> Actually, I just made that push (commit 6f7f855) a little ahead of my
> ETA. This should complete my project to get comprehensive run-time
> testing (done with the scripts/comprehensive_testing.sh script) to
> work completely on Linux for a test case where
> Tcl/Tk/itcl/itk/iwidgets were installed with a pathname prefix that
> had a blank. This should clear out many of the blank-in-path issues
> that would have hindered you for your own comprehensive run-time
> testing.
>
> @everyone: As part of these series of commits I also took the
> opportunity to remove all mention of the unused LIB_TAG variable
> from our build-system logic. This change makes that logic
> much easier to read.
>
> Alan
>
> __________________________
> Alan W. Irwin
>
> Astronomical research affiliation with Department of Physics and Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
>
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
>
> Linux-powered Science
> __________________________
__________________________
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________
Linux-powered Science
__________________________
------------------------------------------------------------------------------
Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
_______________________________________________
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
