On Dec 24 2013 11:01 AM, EBo wrote: > On Dec 24 2013 6:15 AM, Michael Haberler wrote: >> Am 24.12.2013 um 13:26 schrieb EBo <[email protected]>: >> >>> On Nov 21 2013 2:37 AM, andy pugh wrote: >>>> On 21 November 2013 01:10, Michael Haberler <[email protected]> >>>> wrote: >>>> >>>>> looks like the libraries for boost and python-dev are missing in >>>>> the >>>>> Makefile >>>> >>>> I guess it is possible that the "syntactical error" was actually a >>>> case of getting the code good enough to get far enough to notice >>>> missing BP. >>>> >>>> I am using it without problems elsewhere in the code (and have >>>> been >>>> for months) but, now you mention it, not in that code sub-folder. >>> >>> not sure that this is completely related, but I wanted to try >>> building >>> the unified-build branch and found I have a: >>> >>> "configure: error: boost::python is required to build LinuxCNC" >>> >>> error. I am currently using Boost-1.52.0 with boost-numpy and >>> boost-m4 >>> built for good measure. >>> >>> BTW, I used the following configure command: >>> >>> "./configure --without-libmodbus --enable-simulator --with-posix" >>> >>> So, has this been fixed elsewhere, or possibly I need to upgrade my >>> Boost version to 1.55? >> >> >> not sure what to fix since it's hard to tell where the error is - >> for >> instance we dont know where your boost library is installed, which >> it >> seems you built from source >> >> boost::python does not have a package config to derive flags, so the >> configure code is a tad specific, see lines 43-95 here: >> >> >> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=blob;f=src/configure.in;h=71d44790041256645fa02ac6f90cdfeeff32d775;hb=refs/heads/master >> >> if your installation is in a standard place, a patch would be great! > > I'll take a look at this after the impending holiday feast (YUM ;-) > > Anyway, yes I am running on Gentoo -- which by default is build from > source. The includes are in /usr/include/boost/... and the libraries > are in /usr/lib/libboost* (including python, and many others). I > will > have to dig through the logs and maybe hack a little to sort this > out, > but I thought I would ask before spending a couple of hours crawling > in > the code... > > Patches are likely impending as well ;-)
OK... I tracked it down (just before passing out from all the turkey... The problem on my distro is that the boost_python libs are slotted (meaning that they are all appended with specific version. So the -lboost_python should be something like -lboost_python-2.7, etc. So, the problem with checking the specific version of boost python that appears to have been sorted out around line 65 of configure.in does not appear to have been propagated lower down in the tests (such as around line 90). I will try to look at this some more, but it might not be the next couple of days... EBo -- ------------------------------------------------------------------------------ Rapidly troubleshoot problems before they affect your business. Most IT organizations don't have a clear picture of how application performance affects their revenue. With AppDynamics, you get 100% visibility into your Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro! http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
