This commit also drops the patch provided upstream, and since
version 0.5.1, the ar_track_alvar package now provides an archive
with more default directory structure.
This commit also corrects the license information in the recipe to
LGPL-2.1.
I noticed a build failure for yaml-cpp (#328), that the boost
headers could not be found. Probably since the update to 0.5.2 [1],
yaml-cpp depends on boost. This commit resolves the issue by adding
this dependency.
[1] da10180fc5
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
Due to the recent commit 6eddc9e2@openembedded-core [1], which
updated python-nose from 1.2.1 to 1.3.6, the bbappend for
python-nose 1.2.1 leads to an error during parsing. Instead of
making the bbappend version-independent, this commit simply drops
the bbappend file, as python-nose is provided as native package
already since 78711c88@openembedded-core [2] (since January 2014).
[1] 6eddc9e222
[2] 78711c8875
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
As per Section 24.5.15 of the Yocto Manual [1], use of 'virtclass'
overrides has been deprecated since Yocto version 1.6. Update to
the new syntax.
[1] http://www.yoctoproject.org/docs/1.8/mega-manual/mega-manual.html
Signed-off-by: Ash Charles <ashcharles@gmail.com>
Bump from Velodyne version 1.1.2 (Nov 2013) to the latest release,
1.2.0 (Aug 2014). This version includes a fix to support yaml-cpp
1.5+ so resolves#325 [1].
[1] https://github.com/bmwcarit/meta-ros/issues/325
Signed-off-by: Ash Charles <ashcharles@gmail.com>
The 'remove_packaging_data_files' post-processing command generates a
warning for Poky version 1.5+ and is no longer needed. Section 24.4.9
of the Yocto manual (v1.8) explains that these files are automatically
removed for images not using the package-management feature.
This has been maintained for backwards-compatibility but as of the 0.2
release of this layer, Poky version 1.4 is several years old so we can
reasonably remove this post-processing command (discussion [1]).
[1] https://www.mail-archive.com/meta-ros%40googlegroups.com/msg00086.html
Signed-off-by: Ash Charles <ashcharles@gmail.com>
During some code inspections, I discovered that two consecutive
empty lines slipped in with the commit 185882428c.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
The frontier-exploration recipe stated that the software is under
BSD license, probably due to a copy-and-paste inattention. The
commit now points to the license line for version 0.2.2 and changes
the license to GPLv3, which is stated in the package.xml.
Strangely, the package.xml states that the frontier_exploration
ROS package is licensed under GPLv3, but the license text in the
LICENSE file is the GPLv2.1 terms and conditions; so the actual
intended license by the copyright holders remain unclear.
However, assuming the conditions for GPLv3 must be fulfilled
for usage and distribution is a 'safe' approximation, even if
the conditions for GPLv2.1 apply.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
With the poky-dizzy distribution, the do_rootfs task for
core-image-ros-world fails with:
ERROR: Unable to install packages. Command '/[...]/build/tmp/sysroots/x86_64-linux/usr/bin/smart --quiet --data-dir=/[...]/core-image-ros-world/1.0-r0/rootfs/var/lib/smart install -y run-postinsts@all packagegroup-core-boot@beaglebone packagegroup-ros-world@all' returned 1:
error: Can't install ar-track-alvar-0.4.1-r0@cortexa8hf_vfp_neon: no package provides libmedianFilter.so
Build Configuration: poky 1.7.1;
poky: "dizzy:ec75238f6cc2d2d8d40e0268f6d2acc070cbe9a4";
meta-openembedded: "dizzy:9efaed99125b1c4324663d9a1b2d3319c74e7278"
To resolve this problem, this commit updates ar-track-alvar to the
latest Hydro version 0.4.2. Unfortunately, there is no archive file
for version 0.4.2, so the recipe uses the git repository with the
commit intended to mark version 0.4.2 to fetch the source code.
Due to the update, this commit also removes the upstream-accepted
patch file from this repository here.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
During bitbaking core-image-ros-world for v0.2-rc1 release testing,
compiling libpoco for the qemuarm64 machine failed with:
In file included from [...]/poco-poco-1.5.3-release/Foundation/src/diy-fp.h:31:0,
from [...]/poco-poco-1.5.3-release/Foundation/src/diy-fp.cc:29,
from [...]/poco-poco-1.5.3-release/Foundation/src/NumericString.cpp:23:
[...]/poco-poco-1.5.3-release/Foundation/src/utils.h:72:2: error: #error Target architecture was not detected as supported by Double-Conversion.
This issue has been already been reported in the libpoco github issue
tracker [1] and has been resolved with a simple patch [2] in the libpoco
repository and libpoco releases since 1.5.4. Hence, this commit simply
adds this patch to the current libpoco recipe.
To address the libpoco issue, I also considered to update libpoco to
version 1.6.0. However, this was not possible as version 1.6.0 requires
CMake >= 3.0.0 and this would require updating cmake in
OpenEmbedded-Core, which has major impact on all layers. Also, updating
libpoco to 1.5.4 lead to a problem with the OpenEmbedded-Core-provided
pcre 1.5.36 and the POCO_UNBUNDLED setting during compilation:
In file included from [...]/poco-poco-1.5.4-release/Foundation/src/RegularExpression.cpp:21:0:
[...]/usr/include/pcre.h:325:26: error: conflicting declaration 'typedef struct real_pcre pcre'
In file included from [...]/poco-poco-1.5.4-release/Foundation/src/RegularExpression.cpp:17:0:
[...]/poco-poco-1.5.4-release/Foundation/include/Poco/RegularExpression.h:37:34: note: previous declaration as 'typedef struct real_pcre8_or_16 pcre'
This issue is probably caused by the commit 'PCRE 8.35.0 Update' [3],
which defines types from pcre 8.35 that are incompatible to pcre 8.36.
[1] https://github.com/pocoproject/poco/issues/508
[2] 9258e482d7.patch
[3] 010f7a5370.patch
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
When compiling python-rosdep before python-nose has been built,
'bitbake python-rosdep' fails with:
| distutils.errors.DistutilsError: Could not find suitable distribution for Requirement.parse('nose>=1.0')
| ERROR: python setup.py build_ext execution failed.
| WARNING: /[...]/build/tmp-glibc/work/i586-oe-linux/python-rosdep/0.11.0-r0/temp/run.do_compile.14813:1 exit 1 from
| exit 1
| ERROR: Function failed: do_compile (log file is located at /[...]/build/tmp-glibc/work/i586-oe-linux/python-rosdep/0.11.0-r0/temp/log.do_compile.14813)
ERROR: Task 6 (/[...]/meta-ros/recipes-devtools/python/python-rosdep_0.11.0.bb, do_compile) failed with exit code '1'
In a recent commit in the rosdep repository [1], the dependency for
the rosdep setup on python's nose package is added. Hence, we must
also add this dependency in the rosdep recipe, which then resolves
the reported failure.
[1] fc8b56be07
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
On the meta-ros issue tracker, the issue #291 reports a
linking problem with rosconsole if ROS Hydro is also installed
on the host system. Kristof investigated in that, and published
a patch [1] on September 27th, 2014, which resolved the rosconsole
issue, but showed issues with urdfdom.
After reading through the discussion of issue #291, I started
initial testing with Kristof's patch. After testing Kristof's
patch, I also investigated the urdfdom problem, and came up with
the solution to revert the part of his patch, which moved
`-DCMAKE_INSTALL_PREFIX:PATH='${ros_prefix}'` to ros.bbclass.
Initially, I believed that the issue that was addressed with
the second part of Kristof's patch, has been resolved with
commit 7e2eb25e51. However,
the issue remains, but is only reproducible with the Ubuntu
saucy distribution.
On my first local setup, I could reproduce the issue #291 with
rosconsole on commit 47eab4263c.
After applying this commit, the issue with rosconsole did not
occur anymore on a clean fresh build.
`bitbake packagegroup-ros-world` did not show any other further
issues.
On my second local setup, on a newly-installed Ubuntu 12.04
(precise) system, I checked that the proposed commit resolves
some linking problems to boost, with some latest
OpenEmbedded-Core repository and the poky-dizzy distribution.
A detailed report of the investigation of this second local
setup is at my Github Gist [2].
In the first review of the pull request #318, Kristof noticed
that on an Ubuntu 13.10 (saucy), an issue with message-filters
still occurs. I could not reproduce this and other reported
errors on the Ubuntu 12.04 system, so I believe certain errors
only appear on certain Ubuntu distributions, which makes them
difficult to pinpoint. Therefore, we decided to defer the
resolution of this problem with Ubuntu 13.10.
To adjust to the concurrent work in pull request #319, during
the rebasing, the patch has been moved from the catkin directory
to the files directory.
Kristof remains the author of the applied patch, as I have not
modified the patch. I have put myself as this author's commit,
as I take the responsibility of the modifications compared to
Kristof's original work and I have tested this commit in my
test setting.
[1] 9ff76ffb7a
[2] https://gist.github.com/bulwahn/a8d5b7c27550b399f866
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
catkin: move to files directory (fixup)
Lukas Bulwahn