Build Script Option for Different version of WWM #4
Comments
|
Hi @jamal919 , I believe you can do it easily. In the src/Makefile, you just change: or to whatever you like. |
|
Thanks @huyquangt, I understood the change you suggested. The problem is WWM.3012 bundled with current releases does not work directly with this change, because the files to be compiled are not same for the two releases of WWM. |
|
Drifting here a bit on the philosophical side... The WWM branch you are referring to is not yet official, correct? The SCHISM master branch should only support official releases of the module. If you are interested in support for other branches in other modules (which are to be treated as forks in git) what you could do is fork the SCHISM repo and modify it so this branch is supported. In the future, when the module branch is integrated into the main repo, then we can merge your branch to the master to support officially these new features. In general, the master branch should only support what is tested. Everything else is up to us, the developers, to work out on our forks so we can prepare the modifications needed for future merging. @jamal919 @huyquangt @josephzhang8 @water-e Please take note of this thread and comment, because this is precisely the kind of situation that needs to be communicated so we can all have effective collaboration. Please post your thoughts and opinions. |
|
@jamal919 |
|
@jreniel, thanks for your thoughts. I can not agree more with your comments. And yes, as far as I have understood the La Rochelle version is not official, but added as a separate branch to facilitate the continuation of the development they are pursusing by @josephzhang8. Ideal scenario is of course to have a single implementation of a single module, not two as it appear now. Since the user/developer community of WWM-SCHISM can safely be assumed to be smaller than the SCHISM-Core this issue will probably persists for some more time. Any idea or roadmap for unification is very much appreciated, of course. |
|
@jamal919 I'm glad to know that we are on agreement. Because we're just starting up the git, and I am a newcomer to the SCHISM community, I was a bit afraid of what would be the response. But I am glad that it seems the ship is steering well. But back to the main discussion (the following are just ideas, don't take them as absolutes)... First I have to ask, are there any plans to integrate the La Rochelle branch into the trunk at some point? If the answer is yes, then I suppose compilation steps will change, so this is a semi major version change of schism, and you may start tracking this via a feature branch on the main repo (if collaborative tracking is expected), or a fork (if you're flying this one solo). Then when the the La Rochelle branch gets merged, we can do a PR to a test branch and finally merge the test branch to the master, along with a semi major version number increase and new tag. If the La Rochelle branch is expected to diverge, then perhaps the schism master should not consider it at all. Any schism version interested in using the La Rochelle version should fork schism and name it schism_la_rochelle or something else. But the point is that La Rochelle, as of now, is a fork, until it becomes part of the trunk. That being said, I would love to see all these modules be brought to git. We can work with SVN trunks, but git based development would be beneficial I think for everyone. In any case, I'm available to help out with those migrations as well if the parties involved are interested. |
|
Closing issue since it seems we are on strategic agreement. |
* update notes * veg_icm frame * boxes left for light/inundation stress + init * backup before adding dry_sediment_diagenesis * sediment diagenesis under dry condition * backup for optional mortality process * veg light interactions + renew sav whole volume unit * round for check; back up for dynamic linkage * non-dynamic feedback set for multi-veg condition * backup before going through diff and adding debugging outputs/msg * backup test 1 * backup before removing mht etc; replace morris formula with two linear equaitons since it is universal for all locations * benchmark #2, to add veg-related process on nutrient dynamics; leave full dynamics loop apart after benchmark #3 with both veg+sav * benchmark#3, to fix temperal variation + spatial distribution; fixed SAV unit bug for nutrient dynamic processes + allow co-occupation * reset CMakeLists in WWMIII_Sept2020 * backup for benchmark #3; to modify parameters + add full feedback loop * benchmark #4, renewed nutrient recycle and tested equilibrium * add vegetation production; modify hcan calc Co-authored-by: Nicole Cai <ncai@vims.edu>
Since the sourcecode of SCHISM has been moved from SVN to Git, the several other branches that were made in the course of last two months has not been shifted here in Git. One of the possible consequence of this is probably the development branches of different modules - WWM is one of such example.
One version of WWM that I am particularly interested in is the La Rochelle branch. Few months ago, the La Rochelle branch was added officially to a separate SVN branch, which continued developement from the original WWM.3012 (with same folder name). Another version is the official version (with folder name WWM).
The two code folder for one module may seem problematic to people and I hope that in the future these two codes get merged to one place implementing on something that the user community agrees and thoroughly tested. Until such time comes, is it possible to add an option in the build script to define which WWM version is to be built? The two version I have in mind are La Rochelle Branch (WWM.3012) and the Trunk (WWM).
The text was updated successfully, but these errors were encountered: