Asynchronously download information about all stations worldwise and write them to a JSON file.

get_all_stations.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Requires Python >= 3.5, aiohttp, and pandas.

$ pip install aiohttp pandas

Upon running it will create a file 'networks_stations.json' which contains all
stations it could find across FDSN web service capable data centers. For each
station it contains the epochs and coordinates.
"""
import asyncio
from aiohttp import ClientSession
import io
import json
import pandas as pd


# All known nodes.
NODES = ["http://eida.bgr.de",
         "http://eida.ethz.ch",
         "http://service.geonet.org.nz",
         "https://geofon.gfz-potsdam.de",
         "http://webservices.rm.ingv.it",
         "http://eida.ipgp.fr",
         "http://eida.koeri.boun.edu.tr",
         "http://erde.geophysik.uni-muenchen.de",
         "http://service.ncedc.org",
         "http://eida-sc3.infp.ro",
         "http://eida.gein.noa.gr",
         "http://ws.resif.fr",
         "https://service.scedc.caltech.edu",
         "http://sismo.iag.usp.br",
         "http://www.orfeus-eu.org",
         "http://service.iris.edu"]


async def fetch(url, session):
    async with session.get(url) as response:
        if response.status != 200:
            print("'%s' return code %i - will be ignored." %
                  (url, response.status))
            return None
        r = await response.read()
        return (url, r)


async def run():
    tasks = []
    async with ClientSession() as session:
        for url in NODES:
            _url = url + "/fdsnws/station/1/query?level=station&format=text"
            task = asyncio.ensure_future(fetch(_url, session))
            tasks.append(task)

        responses = await asyncio.gather(*tasks)
        return responses


loop = asyncio.get_event_loop()
future = asyncio.ensure_future(run())
inventories = [_i for _i in loop.run_until_complete(future) if _i]
print("Downloaded %i inventories." % len(inventories))


_inv = []
for _i in inventories:
    with io.BytesIO(_i[1]) as buf:
        buf.seek(0, 0)
        try:
            csv = pd.read_csv(buf, sep="|")
            csv.columns = ["network", "station", "latitude", "longitude",
                           "elevation", "site_name", "starttime", "endtime"]
            _inv.append(csv)
        except:
            print("Failed to parse result from '%s'. Invalid CSV file?" %
                  _i[0])


inv = pd.concat(_inv)
result = {}
# Get unique stations and get all epochs.
for key, value in inv.sort_values(by=["network", "station"]).groupby(
        ["network", "station"]):
    # Safe-guard.
    assert key not in result
    del value["network"]
    del value["station"]
    del value["site_name"]
    value.drop_duplicates()
    result["%s.%s" % key] = list(value.T.to_dict().values())

with open("networks_stations.json", "wt") as fh:
    json.dump(result, fh, indent=4)

print("Written to 'networks_stations.json'.")