6. pyOver Documentation

This is the documentation for pyOver/pyover, which provides a Python module and scripts for interacting with OVERFLOW.

The basic usage for this module is to create a file called pyOver.json and use the script pyover. In addition to this control file for pyOver, several other OVERFLOW input files (or, more accurately, input template files) must be provided.

  • OVERFLOW namelist, overflow.inp

  • Mesh files, usually in the common/ folder

  • There must also be a run matrix, which can either be specified within pyOver.json or within a separate file

All of these files can have different names for cases when that helps in the organization of runs, but these are the default file names. If all these files exist in the same folder, the following is the basic command that will start up to 10 cases.

$ pyover

The following will start up to N cases.

$ pyover -n $N

This will start no cases but show you the status of all the runs.

$ pyover -c

Changing the flag to -cj will also show PBS job numbers for cases where a current job is either running or in the queue. Supposing that the file is not called pyOver.json but run.json, the -f flag can be used. If at least some cases have partially run, the command to update or create the data book (i.e. analyze force and moment coefficients and statistics as specified in the control file), the --aero flag will lead to that action.

$ pyover -f run.json --aero

The last option to be highlighted on the front page is the automated report generation capability. This command generates a multipage PDF based on settings in the “Report” section of run.json that usually gives a summary of each case in the run directory on its own page.

$ pyover -f run.json --report

There are many more flags to pyover, including options to compress or archive folders, plot iterative histories, and constrain to only a subset of the run matrix.

Controlling runs with pyOver really focuses on the pyOver.json file. This file is divided into sections, some of which are mandatory.