-
Notifications
You must be signed in to change notification settings - Fork 230
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Build a rmgmolecule
Subpackage in Place, fix rmg
conda recipe, initial SimulatorInterface
Efforts
#2539
Build a rmgmolecule
Subpackage in Place, fix rmg
conda recipe, initial SimulatorInterface
Efforts
#2539
Conversation
This comment was marked as off-topic.
This comment was marked as off-topic.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks promising.
When would we publish new conda binaries of the molecule package?
I suggest we try building it on every push to detect errors, but presumably only want to publish a new "release" whenever something significant changes that RMS/pynta/etc. or another package is depending on? Would this be done by just incrementing a minor version number in a yml file or something? We might wish to do it more often than a full RMG release.
4ced2b0
to
bd8f7c3
Compare
rmgmolecule
Subpackage in Placermgmolecule
Subpackage in Place, fic rmg
conda recipe
rmgmolecule
Subpackage in Place, fic rmg
conda recipermgmolecule
Subpackage in Place, fix rmg
conda recipe
Building RMG binary with Julia included is proving to be annoying. Currently attempting two fixes: (1) relax version requirements on RMS to try and get the environment to solve and precompile and (2) pin packages that are compatible. Alternative to these alternatives is to ship the binary with only the basic RMG python functionality, and then add installation steps for installing RMS from the command line (this might be a really good idea). Would need to change tests to use Python instead of python-jl. |
79fcb93
to
fc706bc
Compare
Julia is attempting to install the |
3768bb4
to
5e9bd6c
Compare
rmgmolecule
Subpackage in Place, fix rmg
conda recipermgmolecule
Subpackage in Place, fix rmg
conda recipe, initial SimulatorInterface
Efforts
This comment was marked as resolved.
This comment was marked as resolved.
Have to use |
|
1e9d84c
to
af201e0
Compare
squashed debugging commits and rebased onto main |
This PR should fix the linked issue when cleaning the environment file. |
- small updates to rmg-py recipe, more coming later - removed conda_build.yml action since it will now be done in a private repo - added a recipe for rmgmolecule - add warnings on molecule when running from rmgmolecule that some functionality is not available
changes required to make the RMG-Py conda build work again. detailed description: - recipe now contains the appropriate requirements - removed macos travis-specific thing - added comments in build.sh for historical preservation of my failed efforts - other changes are the initial steps toward making the conda build work, future commits will make the build tests pass the ci runs on a self-hosted runner in the green group offices. required a specific version of conda build and libmamba to be installed (see PR of this commit for more details), currently fails tests because rmg tries to import Julia during testing. will now need to 'bury' the Julia import and make it optional, including docs updates, etc.
- the import for the Julia dependencies is now properly buried, s.t. it will not interfere with running Arkane w/o Julia installed - inside main, input, and model there is a new argument requires_rms. This argument is passed to prune, add_species*, and other functions that use those functions to avoid moving things to RMS when we either (1) are not using an RMS reactor or (2) don't even have Julia dependencies installed at all
will also need to add to the docs for the installation instructions for installation using conda binaries and installation from source
missed some rms-dependent calls in previous commits
dc3ce92
to
7de8a3a
Compare
This comment was marked as outdated.
This comment was marked as outdated.
This pull request is being automatically marked as stale because it has not received any interaction in the last 90 days. Please leave a comment if this is still a relevant pull request, otherwise it will automatically be closed in 30 days. |
WIP |
Splitting this into three PRs:
|
Summary
This PR is attempting to do 3 things:
rmgmolecule
, which contains just the code contained inrmgpy/molecule
, so that other softwares can use it without having to install all of RMG-Py.SimulatorInterface
, which will allow users to install RMG without installing RMS and its additional requirements.SimulatorInterface
class which will allow users to 'plug-in' whatever simulators they want, but for the time being the solution is going to look more like "if you want RMS, follow these additional installation instructions after installing RMG" (which are going to be the dev installation instructions for RMS as given in RMG)This was originally small in scope but these sub-goals have required one another to proceed.
TODO
rmgmolecule
can be built on windows, and or more version of python/numpy/etc., since it now has lighter dependenciesreactors.py
tojulia_reactors.py
, bury the import statement and add a runtime warning when attempting to import it that says RMS must be installed, and then update the docs to explain how to install RMS