find_cmip6_model

find_cmip6_model.py

#!/usr/bin/env python3

import glob
import sys
import re

def return_datasets(activity="*",
                    institute="*",
                    dataset="*",
                    exp="*",
                    ensemble="*",
                    mip="*",
                    short_name="*",
                    grid="*",
                    version="*"):
    
    rootpath_cmip6=["/g/data/oi10/replicas/CMIP6/",  "/g/data/fs38/publications/CMIP6/", "/g/data/xp65/public/apps/esmvaltool/replicas/CMIP6/"]

    results = []
 
    for path in rootpath_cmip6:
        query = path + f'{activity}/{institute}/{dataset}/{exp}/{ensemble}/{mip}/{short_name}/{grid}/{version}'
        results = glob.glob(query)
        for result in results:
            print(extract_and_parse_path(result, path))


def extract_and_parse_path(full_path, root_value):
    # Ensure the root ends with a slash for accurate comparison
    if not root_value.endswith('/'):
        root_value += '/'
    
    # Check if the full path starts with the given root value
    if not full_path.startswith(root_value):
        raise ValueError("The provided root value does not match the beginning of the full path.")
    
    # Remove the root part from the path
    remaining_path = full_path[len(root_value):]
    
    # Define the pattern to match "{val1}/{val2}/{val3}"
    pattern = r'^(?P<activity>[^/]+)/(?P<institute>[^/]+)/(?P<dataset>[^/]+)/(?P<exp>[^/]+)/(?P<ensemble>[^/]+)/(?P<mip>[^/]+)/(?P<shortname>[^/]+)/(?P<grid>[^/]+)/(?P<version>[^/]+)$'
    
    # Match the remaining path against the pattern
    match = re.match(pattern, remaining_path)
    
    # Extract values
    activity = match.group('activity')
    institute = match.group('institute')
    dataset = match.group('dataset')
    exp = match.group('exp')
    
    ensemble = match.group('ensemble')
    mip = match.group('mip')
    shortname = match.group('shortname')
    grid = match.group('grid')
    version = match.group('version')

    
    #result = {
    #    'activity': activity,
    #    'institute': institute,
    #    'dataset': dataset,
    #    'exp': exp,
    #    'ensemble': ensemble,
    #    'mip': mip,
    #    'shortname': shortname,
    #    'grid': grid,
    #    'version': version
    #}
    
    result = {
        'dataset': dataset,
        'institute': institute,
    }

    return {key: val for key, val in result.items() if val != " "}

if __name__ == "__main__":
    args = sys.argv[1:]  # Exclude the script name
    kwargs = {}

    for arg in args:
        key, value = arg.split('=')
        kwargs[key] = value

    return_datasets(**kwargs)

#return_datasets(grid="gn", ensemble="r1i1p1f1", dataset="NorESM2-MM", institute="NCC")
#return_datasets(ensemble="r1i1p1f1", dataset="ACCESS-ESM1-5", institute="CSIRO")