Folder names are hashes of request to the service, and are used by the sentinelhub-py as caches - if you’d make another (identical) request, by default it would be read from the cached data.

The information about what is in each response.tiff is also stored in the corresponding request.json. It holds the date/time parameter of the particular request.

Here is a short script that will hopefully help you. It targets WmsRequest…

import pandas as pd

import glob

import json

import datetime as dt

from urllib.parse import unquote



def get_request_dt(request_file):

    with open(request_file, 'r') as req:

        request = json.load(req)

        url = unquote(requests'url'])

        time_parameter = =t for t in url.split('&') if t.startswith('TIME=')])0]

        time = time_parameter.split('TIME=')'1].split('/')'0]

        return dt.datetime.strptime(time, "%Y-%m-%dT%H:%M:%S")



if __name__ == "__main__":

    folders = glob.glob('./data/*')

    dates = =get_request_dt(f'{folder}/request.json') for folder in folders]

    df = pd.DataFrame(zip(folders, dates), columns=s'folder', 'datetime'])

    df.sort_values(by='datetime', ascending=True, inplace=True)

    print(df.head().to_markdown())

  

Such script would output something like

|    | folder                                  | datetime            |

|---:|:----------------------------------------|:--------------------|

|  2 | ./data/a986c7940bf6ed992b150b9e42ec13ae | 2017-12-15 07:12:03 |

|  9 | ./data/895642dd7e880aa150881f852409d454 | 2017-12-20 07:12:10 |

| 13 | ./data/d3eecd3d345e2bc98fd2966bd7d261a8 | 2017-12-25 07:12:04 |

| 41 | ./data/c259365d92977c1e0c347ec91d064dd5 | 2017-12-30 07:12:09 |

| 26 | ./data/886e355f522798ec8bbed6b03743f3b1 | 2018-01-04 07:12:06 |

You’ll have to adapt it to work with recursive folders etc, but I hope it will get you going.

Thank you very much for your response! I ended up writing a python script to read in request.json and scrape the date and then rename response.tiff with the date. But your code is pretty nice! i will save it for future documentation of the hashed file names!

Hi @batic

I am downloading weekly Sentinel-2 median mosaic using process-api and sh-py. I want the date range I gave for the median in the downloaded file name. Can you share an example script for this?