path: root/package/ola/ola-0002-move-python-sub-check-to-configure.ac.patch
Commit message (Collapse)AuthorAgeFilesLines
* ola: bump to 0.9.3Gravatar Simon Marchi2014-12-081-48/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This update requires a few changes. First, good news, the patches that you guys have submitted to OLA have been merged, so we can drop those: ola-0001-fix-build-warning.patch -> ea375582b0bfee93d66608ffc807078ffc48e961 ola-0002-move-python-sub-check-to-configure.ac.patch -> 673a7602a6bf7a6aa8a8461ebd9362d59f6e21df ola-0003-fix-check-for-python-module-for-cross-compilation.patch -> b51b48be81ec38bc7d1229be0c7d3189c5ddbafa Less good news: OLA now builds a small protoc wrapper (ola_protoc) that is built and executed at compile-time on the host. If we don't change anything, ola_protoc is built with the target toolchain and therefore can't run on the host. Explanation for ola_protoc is here To solve this, I created a package host-ola, which builds and provides an ola_protoc for the host. It tries to disable as much as possible of things that we won't need at configure time. Only ola_protoc is built and installed so it's not that long. The change has been built-tested only. [1] https://github.com/OpenLightingProject/ola/blob/master/protoc/ola-protoc.cpp#L20 [Thomas: - add HOST_OLA_DEPENDENCIES variable, so that host-ola doesn't get all the dependencies of the target ola package: certain target ola dependencies do not have corresponding host packages (and it doesn't make sense to have them as dependencies just to build ola_protoc) - improve the commit log to mention which upstream commits correspond to our patches.] Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
* package/ola: fix check for google.protobufGravatar Yann E. MORIN2014-01-141-0/+48
Fix the google.protobuf check in an upstreamable way. As suggested by Arnout, use an autoconf ac_cv_have_* variable so we can pass it on the command line. This is generic behaviour for a lot of autoconf internals, so it makes sense to do it this way. Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Arnout Vandecappelle <arnout@mind.be> Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be> Signed-off-by: Peter Korsgaard <peter@korsgaard.com>