Not all of the following are GIFT-Grab issues. Some of them are related to the external dependencies of GIFT-Grab. However here we document all as they may be useful when getting started with GIFT-Grab.
In case GIFT-Grab is built with a custom CMAKE_INSTALL_PREFIX
that is not a standard system path, ${CMAKE_INSTALL_PREFIX}/lib/giftgrab
needs to be included in CMake's CMAKE_MODULE_PATH
for FIND_PACKAGE(GiftGrab)
to work.
If you encounter an option not recognized
error while pip install
ling GIFT-Grab, run:
pip install pytest
andpip install PyYAML
before runningpip install gift-grab [...]
.
Please see the note in the libVLC installation instructions about capturing network streams in the I420 colour space.
The frame rate of VP9-encoded WebM files is set incorrectly on Ubuntu 16.04 LTS.
On Ubuntu 16.04 LTS, the first few (~1-3) attempts to connect to the DVI port of an Epiphan DVI2PCIe Duo, after connecting and turning on a video source, fail. Subsequent attempts work as expected.
- Encountered on Debian 9 when compiling Boost.
CPLUS_INCLUDE_PATH
might need to be set to include the directory wherepyconfig.h
is located.
- On Debian 9 CMake was reported not to pick an
export
'edBOOST_ROOT
. - Specifying
BOOST_ROOT
to CMake as-D BOOST_ROOT=/path/to/your/boost
solves this problem.
If after successfully building / installing GIFT-Grab, you encounter an error similar to
ImportError: /home/stefan/src/GIFT-Grab/pygiftgrab.so: undefined symbol: _ZN5boost6python6detail11init_moduleEPKcPFvv
this probably means that your Boost was compiled and linked against Python 3.
Please make sure:
- you are using Python 2
- your Boost is built and linked against Python 2 libs
If (although you've successfully installed Boost) you encounter an error similar to
Could not find the following Boost libraries:
boost_numpy
Some (but not all) of the required Boost libraries were found. You may
need to install these additional Boost libraries. Alternatively, set
BOOST_LIBRARYDIR to the directory containing Boost libraries or BOOST_ROOT
to the location of Boost.
this means that your Boost was built with Python support, but does not include NumPy support.
Boost seems not to build the boost_numpy
library if NumPy is not installed on the system when building Boost.
This issue can be resolved by:
- installing NumPy
- building and installing Boost from scratch
- This issue seems to be related to custom
CXXFLAGS
. - A possible workaround is to
unset CXXFLAGS
before installing GIFT-Grab withpip
.