generate_s_index

generate_s_index.py: generate the s_index.json file for an Experiment

usage: generate_s_index.py [ -o outfile ] SCRIPT [...]

where:

• SCRIPT ... is a list of JSON scripts, or
• SCRIPT is a directory containing JSON scripts (if a directory, the directory is scanned for all its EXOSIMS scripts)
• -o allows naming a specific output .json file

Only .json files that look like EXOSIMS input scripts are indexed.

The output file (-o) typically would not be given. If not, its path is inferred from the directory of the first given SCRIPT (or from the script-directory, if that style of argument is used).

The full output filename will be the script directory name as determined above, plus s_index.json.

Typical usage:

• Usually best: util/generate_s_index.py Scripts/ExampleExp.exp
• Also OK: util/generate_s_index.py Scripts/ExampleExp.exp/*.json

NOTE: This is a "primitive" indexer that does not do anything clever about the actual Experiment parameters. The s_index.json file lists all viable EXOSIMS scripts, but the index does not break out any specific parameter values.

That is, if iterating over telescope diameter and contrast, or scheduler coefficients, symbolic names of these parameters will be entered in the s_index.json file generated by other tools, like json-xform.py. Such index files will look like this:

{
  "diam": 5,
  "contrast": 1e-10,
  "metric": "A",
  "IWA": 2.0,
  "script_name": "H5_C1e-10_baseA_IWA2.0.json",
  "run_name": "H5_C1e-10_baseA_IWA2.0"
},

An entry produced by the present routine just has an arbitrary script number:

{
  "index": 11,
  "script_name": "H5_C1e-10_baseA_IWA2.0.json",
  "run_name": "H5_C1e-10_baseA_IWA2.0"
},

Thus, this is a backup mechanism to allow Sandbox tools to work without complaint. The downstream effect (not catastrophic) is that Experiment-wide tables like reduce-yield-plus.csv will not have row-by-row parameter information other than the textual Ensemble name (e.g., H5_C1e-10_baseA_IWA2.0).