pycoustic 0.1.9__tar.gz → 0.1.10__tar.gz

{pycoustic-0.1.9 → pycoustic-0.1.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: pycoustic
-Version: 0.1.9
+Version: 0.1.10
 Summary: 
 Author: thumpercastle
 Author-email: tony.ryb@gmail.com

{pycoustic-0.1.9 → pycoustic-0.1.10}/pycoustic/survey.py

@@ -1,6 +1,6 @@
 import pandas as pd
 import numpy as np
-from .weather import WeatherHistory
+# from .weather import WeatherHistory
 
 
 DECIMALS=1
@@ -193,17 +193,17 @@ class Survey:
             pos_summary = []
             # Daytime
             period_headers = ["Daytime"]
-            days = log.get_modal(data=log._get_period(data=log.as_interval(t=day_t), period="days"), by_date=by_date, cols=cols)
+            days = log.get_modal(data=log.get_period(data=log.as_interval(t=day_t), period="days"), by_date=by_date, cols=cols)
             days.sort_index(inplace=True)
             pos_summary.append(days)
             # Evening
             if log.is_evening():
                 period_headers.append("Evening")
-                evenings = log.get_modal(data=log._get_period(data=log.as_interval(t=evening_t), period="evenings"), by_date=by_date, cols=cols)
+                evenings = log.get_modal(data=log.get_period(data=log.as_interval(t=evening_t), period="evenings"), by_date=by_date, cols=cols)
                 evenings.sort_index(inplace=True)
                 pos_summary.append(evenings)
             # Night time
-            nights = log.get_modal(data=log._get_period(data=log.as_interval(t=night_t), period="nights"), by_date=by_date, cols=cols)
+            nights = log.get_modal(data=log.get_period(data=log.as_interval(t=night_t), period="nights"), by_date=by_date, cols=cols)
             nights.sort_index(inplace=True)
             pos_summary.append(nights)
             period_headers.append("Night-time")
@@ -234,17 +234,17 @@ class Survey:
             pos_summary = []
             # Daytime
             period_headers = ["Daytime"]
-            days = log.counts(data=log._get_period(data=log.as_interval(t=day_t), period="days"), cols=cols)
+            days = log.counts(data=log.get_period(data=log.as_interval(t=day_t), period="days"), cols=cols)
             days.sort_index(inplace=True)
             pos_summary.append(days)
             # Evening
             if log.is_evening():
                 period_headers.append("Evening")
-                evenings = log.counts(data=log._get_period(data=log.as_interval(t=evening_t), period="evenings"), cols=cols)
+                evenings = log.counts(data=log.get_period(data=log.as_interval(t=evening_t), period="evenings"), cols=cols)
                 evenings.sort_index(inplace=True)
                 pos_summary.append(evenings)
             # Night time
-            nights = log.counts(data=log._get_period(data=log.as_interval(t=night_t), period="nights"), cols=cols)
+            nights = log.counts(data=log.get_period(data=log.as_interval(t=night_t), period="nights"), cols=cols)
             nights.sort_index(inplace=True)
             pos_summary.append(nights)
             period_headers.append("Night-time")
@@ -276,7 +276,7 @@ class Survey:
         for key in self._logs.keys():
             log = self._logs[key]
             combined_list = []
-            maxes = log.get_nth_high_low(n=n, data=log._get_period(data=log.as_interval(t=t), period=period))[["Lmax", "Time"]]
+            maxes = log.get_nth_high_low(n=n, data=log.get_period(data=log.as_interval(t=t), period=period))[["Lmax", "Time"]]
             maxes.sort_index(inplace=True)
             combined_list.append(maxes)
             summary = pd.concat(objs=combined_list, axis=1)
@@ -299,15 +299,15 @@ class Survey:
         for key in self._logs.keys():
             log = self._logs[key]
             # Day
-            days = log._get_period(data=log.get_antilogs(), period="days")
+            days = log.get_period(data=log.get_antilogs(), period="days")
             days = days[leq_cols].apply(lambda x: np.round(10*np.log10(np.mean(x)), DECIMALS))
             # Night-time
-            nights = log._get_period(data=log.get_antilogs(), period="nights")
+            nights = log.get_period(data=log.get_antilogs(), period="nights")
             nights = nights[leq_cols].apply(lambda x: np.round(10*np.log10(np.mean(x)), DECIMALS))
             df = pd.DataFrame
             # Evening
             if log.is_evening():
-                evenings = log._get_period(data=log.get_antilogs(), period="evenings")
+                evenings = log.get_period(data=log.get_antilogs(), period="evenings")
                 evenings = evenings[leq_cols].apply(lambda x: np.round(10 * np.log10(np.mean(x)), DECIMALS))
                 df = pd.concat([days, evenings, nights], axis=1, keys=["Daytime", "Evening", "Night-time"])
             else:
@@ -344,6 +344,101 @@ class Survey:
                             index=["Min", "Max", "Mean"]).drop(columns=["dt"]).round(decimals=1)
 
 
+
+appid = ""
+# with open("tests/openweather_app_id.txt") as f:
+#     appid = f.readlines()[0]
+
+w_dict = {
+    "start": "2022-09-16 12:00:00",
+    "end": "2022-09-17 18:00:00",
+    "interval": 6,
+    "api_key": appid,
+    "country": "GB",
+    "postcode": "WC1",
+    "tz": "GB"
+}
+
+
+def test_weather_obj(weather_test_dict):
+    hist = WeatherHistory(start=w_dict["start"], end=w_dict["end"], interval=w_dict["interval"],
+                          api_key=w_dict["api_key"], country=w_dict["country"], postcode=w_dict["postcode"],
+                          tz=w_dict["tz"])
+    hist.compute_weather_history()
+    return hist
+
+#TODO: Make this take the start and end times of a Survey object.
+#TODO: Implement post codes instead of coordinates
+#TODO: Implement the WeatherHistory as methods within Survey.
+class WeatherHistory:
+    def __init__(self):
+        return
+
+    def reinit(self, start=None, end=None, interval=6, api_key="", country="GB", postcode="WC1", tz="",
+                 units="metric"):
+        if api_key==None:
+            raise ValueError("API key is missing")
+        if type(start) == str:
+            self._start = dt.datetime.strptime(start, "%Y-%m-%d %H:%M:%S")
+        else:
+            self._start = start
+        if type(end) == str:
+            self._end = dt.datetime.strptime(end, "%Y-%m-%d %H:%M:%S")
+        else:
+            self._end = end
+        self._interval = interval
+        self._api_key = str(api_key)
+        self._lat, self._lon = self.get_latlon(api_key=api_key, country=country, postcode=postcode)
+        self._hist = None
+        self._units = units
+
+    def get_latlon(self, api_key="", country="GB", postcode=""):
+        query = str("http://api.openweathermap.org/geo/1.0/zip?zip=" + postcode + "," + country + "&appid=" + api_key)
+        resp = requests.get(query)
+        return resp.json()["lat"], resp.json()["lon"]
+
+    def _construct_api_call(self, timestamp):
+        base = "https://api.openweathermap.org/data/3.0/onecall/timemachine?"
+        query = str(base + "lat=" + str(self._lat) + "&" + "lon=" + str(self._lon) + "&" + "units=" + self._units + \
+                    "&" + "dt=" + str(timestamp) + "&" + "appid=" + self._api_key)
+        return query
+
+    def _construct_timestamps(self):
+        next_time = (self._start + dt.timedelta(hours=self._interval))
+        timestamps = [int(self._start.timestamp())]
+        while next_time < self._end:
+            timestamps.append(int(next_time.timestamp()))
+            next_time += dt.timedelta(hours=self._interval)
+        return timestamps
+
+    def _make_and_parse_api_call(self, query):
+        response = requests.get(query)
+        # This drops some unwanted cols like lat, lon, timezone and tz offset.
+        resp_dict = response.json()["data"][0]
+        del resp_dict["weather"]    # delete weather key as not useful.
+        # TODO: parse 'weather' nested dict.
+        return resp_dict
+
+    def compute_weather_history(self, drop_cols):
+        # construct timestamps
+        timestamps = self._construct_timestamps()
+        # make calls to API
+        responses = []
+        for ts in timestamps:
+            query = self._construct_api_call(timestamp=ts)
+            response_dict = self._make_and_parse_api_call(query=query)
+            responses.append(pd.Series(response_dict))
+        df = pd.concat(responses, axis=1).transpose()
+        for col in ["dt", "sunrise", "sunset"]:
+            df[col] = df[col].apply(lambda x: dt.datetime.fromtimestamp(int(x)))  # convert timestamp into datetime
+        df.drop(columns=drop_cols, inplace=True)
+        return df
+
+    def get_weather_history(self):
+        return self._hist
+
+
+
 # TODO: Fix this bug in weatherhist
 # survey.weather(api_key=r"eef3f749e018627b70c2ead1475a1a32", postcode="HA8")
 #                     dt   temp pressure humidity clouds wind_speed wind_deg  \
@@ -429,4 +524,10 @@ class Survey:
 #   File "C:\Users\tonyr\PycharmProjects\pycoustic\.venv2\Lib\site-packages\numpy\_core\_methods.py", line 48, in _amin
 #     return umr_minimum(a, axis, None, out, keepdims, initial, where)
 #            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-# TypeError: '<=' not supported between instances of 'dict' and 'dict'
+# TypeError: '<=' not supported between instances of 'dict' and 'dict'
+
+
+#TODO: Fix this error:
+#
+# C:\Users\tonyr\PycharmProjects\pycoustic\pycoustic\survey.py:316: FutureWarning:
+# The behavior of pd.concat with len(keys) != len(objs) is deprecated. In a future version this will raise instead of truncating to the smaller of the two sequences

{pycoustic-0.1.9 → pycoustic-0.1.10}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pycoustic"
-version = "0.1.9"
+version = "0.1.10"
 description = ""
 authors = ["thumpercastle <tony.ryb@gmail.com>"]
 readme = "README.md"