Content-Type: multipart/alternative; boundary="973390219-2082574744-1394191713=:56905" --973390219-2082574744-1394191713=:56905 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable =A0=0AHi Alan=0AThe (very)simple example is attached - it can be run simply= with cmake "dir/of/example" -G "Visual Studio 11 Win64"=0AYou might well b= e able to run this on Wine or something as I'm told=A0visual studio generat= ors do not actually need access to the compiler like the nmake generator do= es.=0A=0A2) I wasn't aware of the svn line ending issues. I didn't notice a= ny options within TortoiseSVN to set them so assumed it was dealt with auto= magically. I'll try to hunt down the options next time.=0A=0A3) That was me= just playing around to see if I could affect the CMAKE_CL_64 or CMAKE_C_CO= MPILER or CMAKE_CXX_COMPILER variables at all. When it comes to this stuff = I have no real idea how it all works so was just trying things almost at ra= ndom. To be honest when it got to that point I came to the conclusion that = to fix things=A0I needed to properly learn CMake, which I probably don't ha= ve time for.=0A=0A4) I had to make this change to get the module to run cor= rectly. I think otherwise it was looking in the Plplot build tree for this = file rather than the CMake directory. The change was based solely on a comm= ent on a forum somewhere where someone reported a similar problem to that w= hich I was experiencing.=0A=0A5) Yes that is correct. According to the test= and Brad's expectation the CMAKE_CL_64 variable should be=A0defined for 64= bit Visual Studio and NMake projects and=A0not for 32 bit. For some reason= when this variable and the others you see listed in the commented out mess= ages they indicate 32=A0bit projects. This is despite=A0output from other C= Make files during the generation process that appear to indicate otherwise.= =0A=0AMy main issue with CMake is that I don't really know how the executio= n flow works, or how scope works, or how it interacts with existing or pred= efined variables. So modifying existing code is not to bad, but when it com= es to starting things from scratch or working through multiple files I'm ra= ther lost. As much as I would love to learn, I feel like I have other prior= ities (both Plplot related and not). However, if there are some tests you c= an do to guide things then I am happy to keep working on this.=0A=0APhil --973390219-2082574744-1394191713=:56905 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
 
Hi A= lan
The (very)simple example is attache= d - it can be run simply with cmake "dir/of/example" -G "Visual Studio 11 W= in64"
You might well be able to run thi= s on Wine or something as I'm told visual studio generators do not act= ually need access to the compiler like the nmake generator does.

2) I was= n't aware of the svn line ending issues. I didn't notice any options within TortoiseSVN= to set them so assumed it was dealt with automagically. I'll try to hunt d= own the options next time.

3) That was me just playing around to see if I= could affect the CMAKE_CL_64 or CMAKE_C_COMPILER or CMAKE_CXX_COMPILER var= iables at all. When it comes to this stuff I have no real idea how it all w= orks so was just trying things almost at random. To be honest when it got t= o that point I came to the conclusion that to fix things I needed to p= roperly learn CMake, which I probably don't have time for.

4) I had to m= ake this change to get the module to run correctly. I think otherwise it wa= s looking in the Plplot build tree for this file rather than the CMake dire= ctory. The change was based solely on a comment on a forum somewhere where someone reported a similar problem to that which I was experiencing.
=

5) = Yes that is correct. According to the test and Brad's expectation the CMAKE= _CL_64 variable should be defined for 64 bit Visual Studio and NMake p= rojects and not for 32 bit. For some reason when this variable and the= others you see listed in the commented out messages they indicate 32 = bit projects. This is despite output from other CMake files during the= generation process that appear to indicate otherwise.

My main issue with= CMake is that I don't really know how the execution flow works, or how sco= pe works, or how it interacts with existing or predefined variables. So mod= ifying existing code is not to bad, but when it comes to starting things fr= om scratch or working through multiple files I'm rather lost. As much as I would love to learn, I feel like I have other priorities (both Plplot rela= ted and not). However, if there are some tests you can do to guide things t= hen I am happy to keep working on this.

Phil
--973390219-2082574744-1394191713=:56905--