Full reference can be found on: http://nve-wiki.nve.no/index.php/Web_Chart_Service
In [11]:
# Required imports
import requests
from urllib.parse import urljoin, urlparse
import json
import re
import time
import pandas as pd
In [12]:
%matplotlib inline
GetChart
Parameter
Påkrevd/Valgfri
Standardverdi
Kommentar
ver=
req=getchart P getchart Tjeneste.
lang=
efmt=
chs=
dpi=
chf=
chtt=
In [13]:
url = "http://h-web01.nve.no/chartserver/ShowData.aspx?req=getchart&ver=1.0&vfmt=json&time=20130504T0000;20130506T0000&chs=10x10&lang=no&chlf=none&chsl=0;+0&chd=ds=htsre,da=29,id=25100.0,rt=1,cht=col,mth=sum,timeo=6:0|ds=htsry,id=metx[25100;0].6001,mth=sum,rt=1,cht=col&nocache=0.20784913911484182"
print(urlparse(url))
In [14]:
def chartserver_query(base, mth, ver, fmt, tm, lang, data_req):
return '{0}?{1}&{2}&{3}&{4}&{5}&{6}'.format(base, mth, ver, fmt, tm, lang, data_req)
url_base = 'http://h-web01.nve.no/chartserver/ShowData.aspx'
cs_mth = 'req=getchart'
cs_ver = 'ver=1.0'
cs_fmt = 'vfmt=json'
cs_lang = 'lang=no'
#&chd=ds=htsre,da=29,id=25100.0,rt=1,cht=col,mth=sum,timeo=6:0|
# ds=htsry,id=metx[25100;0].6001,mth=sum,rt=1,cht=col
#&nocache=0.20784913911484182"
In [15]:
rt = "1:00" # hourly time resolution
| Parameter | Code |
|---|---|
| Precipitation | 0 |
| Wind direction | 14 |
| Wind speed (10m) | 16 |
| Air temperature (2m) | 17 |
| Snow depth | 2002 |
| Surface temperature | 2040 |
In [16]:
def get_hourly(station_id, parameter_id, mth='inst'):
"""
Allowed methods (mth): inst, mean, min, max, sum
"""
return "ds=htsre,id={0}.{1},rt=1:00,mth={2}".format(station_id, parameter_id, mth) #,da=29, timeo=6:0
#station_id = 18500 # Bjørnholt
#station_id = 18700 # Blindern
station_id = 54710 # Filefjell
#station_id = 25830 # Finsevatn
# Hourly precipitation
ds_precip = get_hourly(station_id, 0, 'sum')
print(ds_precip)
# Hourly temperature
ds_temp = get_hourly(station_id, 17)
# Hourly wind speed
ds_wind = get_hourly(station_id, 16, 'max')
# Hourly snow depth
ds_snow = get_hourly(station_id, 2002)
In [17]:
def create_data_request(req_list):
count = 0
max = len(req_list)
req = "chd="
for r in req_list:
count += 1
req += r
if count<max:
req +='|'
return req
data_req = create_data_request([ds_precip, ds_temp, ds_wind, ds_snow])
print(data_req)
In [18]:
cs_time = 'time=20160127T0000;20160209T0000'
In [19]:
url = chartserver_query(url_base, cs_mth, cs_ver, cs_fmt, cs_time, cs_lang, data_req)
print(url)
In [20]:
resp = requests.get(url)
data = json.loads(resp.text)
In [ ]:
In [21]:
#print(json.dumps(data, indent=2)) # formated printing of a json object
In [ ]:
In [ ]:
{ "LegendText": "FILEFJELL - KYRKJEST\u00d8LANE (54710), Sn\u00f8dybde (cm)", "Statistics": [], "SeriesPoints": [ { "Value": 78, "Key": "/Date(1454544000000)/", "CorrectionMark": 0 }, }
In [ ]:
In [ ]:
In [22]:
def convert_timestr(s, as_string=True):
"""
:param s: time string returned by chartserver query, e.g. /Date(1457485200000)/
:return:
"""
# regular expression to extract the numeric values in the json date string
regex = re.compile("\d+")
# extract numerical value, make it a float, and divide by 1000 to get seconds since 1.1.1970
ds = time.gmtime(float(regex.search(s).group())*0.001)
if as_string:
ds = time.strftime("%Y-%m-%dT%H:%M:%S", ds)
return ds
#df['Key'] = [convert_timestr(s) for s in df['Key']]
#print(df)
# s = re.findall('\(.*?\)', df['Key'][0])
# s = time.gmtime(df['Key'][0])
#print('Done')
In [23]:
# parse data as pandas time-series
ts_list = []
for i in range(len(data)):
val = [v['Value'] for v in data[i]['SeriesPoints']]
d = [convert_timestr(v['Key']) for v in data[i]['SeriesPoints']]
ts_list.append(pd.Series(val, index=d, name=data[i]['LegendText']))
df = pd.concat(ts_list, axis=1)
# print(df)
# make the date the new index
# df.index = df['Key']
# remove the duplicate Key column
# del df['Key']
In [24]:
df.plot(subplots='True', figsize=(14, 9))
Out[24]:
In [25]:
import sqlite3
In [26]:
db_name = 'filefjell.db'
# Establish connection and cursor
conn = sqlite3.connect(db_name)
#ur = conn.cursor()
df.to_sql(name='FILEFJELL', con=conn)
In [27]:
conn.close()
In [ ]: