Open tvelden opened 12 years ago
The README is now completed and located within /batch-processing. It should have all the information needed to explain how these batch-processes should be used.
I would suggest to mention early on in the README that the script that sets up the core directory structure reads the parameter file; so in the work flow specifying the parameters is probably the first step
the make_core_dirs.sh script actually doesn't require that the parameter file be setup first. This script only creates the subdirectories "data" and "runs" pretty much. The script make_dirs_for_timeslice.sh is what sets up the directory based on the parameter file and this is script is mainly intended to be called from the batch-processing scripts that are called in a run anyway. I wasn't sure how much this particular README file should go into the details of how the runs and parameter file work in general, so I guess that's not entirely clear from this one README, but if this should be intended to be the main README file of our project, I will make this much more clear or refer to another README file that could go in the parameters folder describing the purpose of the parameters-global.txt file in greater detail.
The way I read the script it does access the parameter field and if I specify e.g. FIELD = field3 in the parameter file, the script does create a directory nwa-field3 to hold the runs and data subdirectories. If I remove the parameter file it creates a directory nwa- in the communities directory itself. So to me looks as if the parameter fiel needs to be in place at this point, if only for one parameter value.
I suggest this particular README file need only give anyone on the team minimal how-to guide for a typical work flow that starts from scratch. Eventually, before a public release we may want to produce a more complete documentation.
ah yes, I see now. I will take care of that today.
Please add brief documentation of typical work flow: order of scripts to run, initial directory structure to set up.