Currently IQTREE2 is installed and built in the devcontainer when the docker container is built, just so we can build a set of linux *.a static libraries for the purposes of getting the library working on linux.
This is not the best place long term.
We want the process of cloning this repo to an OS and running
cd ~/repos/iqtree2
rm -rf build
mkdir build
cd build
cmake -DBUILD_LIB=ON ..
make -j
Then to copy the resulting static libraries (.a files in linux) to /pyiqtree/libiqtree/OS {linux/mac/windows}
Then build a relocatable shared library (.so/.dynlib/.dll) around that static library, and copy it to /pyiqtree/libiqtree
Then build the python project for that OS
My instinct is to have a linked submodule in the pyiqtree2 repository that points to the branch of the IQTREE2 repo that implements the library -libiqtree until such time as that code is merged with their main branch. Still need to test if we can specify specific branches in git submodules
Currently IQTREE2 is installed and built in the devcontainer when the docker container is built, just so we can build a set of linux *.a static libraries for the purposes of getting the library working on linux.
This is not the best place long term.
We want the process of cloning this repo to an OS and running
to build IQTREE using the libiqtree flag
Then to copy the resulting static libraries (.a files in linux) to /pyiqtree/libiqtree/OS {linux/mac/windows} Then build a relocatable shared library (.so/.dynlib/.dll) around that static library, and copy it to /pyiqtree/libiqtree Then build the python project for that OS
My instinct is to have a linked submodule in the pyiqtree2 repository that points to the branch of the IQTREE2 repo that implements the library -libiqtree until such time as that code is merged with their main branch. Still need to test if we can specify specific branches in git submodules