Release 3.3.0

CONTRIBUTORS.md

@@ -9,6 +9,7 @@ Code
   * [alechewitt](https://github.com/alechewitt)
   * [camponez](https://github.com/camponez)
   * [Darumin](https://github.com/Darumin)
+  * [davidpirogov](https://github.com/davidpirogov)
   * [dev-iks](https://github.com/dev-iks)
   * [dphildebrandt](https://github.com/dphildebrandt)
   * [dstmar](https://github.com/dstmar)
@@ -45,6 +46,7 @@ Testing
 
 Packaging and Distribution
 --------------------------
+  * [Crozzers](https://github.com/Crozzers)
   * [Diapente](https://github.com/Diapente)
   * [onkelbeh](https://github.com/onkelbeh)
   * [Simone-Zabberoni](https://github.com/Simone-Zabberoni)

Pipfile

@@ -4,6 +4,7 @@ verify_ssl = true
 name = "pypi"
 
 [dev-packages]
+Babel = ">=2.9.1"
 coverage = "*"
 coveralls = "*"
 Jinja2 = "*"

dev-requirements.txt

@@ -11,3 +11,4 @@ tox
 tox-travis
 virtualenv
 twine
+urllib3>=1.26.5

pyowm/agroapi10/agro_manager.py

@@ -6,7 +6,8 @@
 from pyowm.agroapi10.polygon import Polygon, GeoPolygon
 from pyowm.agroapi10.search import SatelliteImagerySearchResultSet
 from pyowm.agroapi10.soil import Soil
-from pyowm.agroapi10.uris import ROOT_AGRO_API, POLYGONS_URI, NAMED_POLYGON_URI, SOIL_URI, SATELLITE_IMAGERY_SEARCH_URI
+from pyowm.agroapi10.uris import ROOT_AGRO_API, ROOT_DOWNLOAD_PNG_API, ROOT_DOWNLOAD_GEOTIFF_API, POLYGONS_URI, \
+    NAMED_POLYGON_URI, SOIL_URI, SATELLITE_IMAGERY_SEARCH_URI
 from pyowm.commons.http_client import HttpClient
 from pyowm.commons.image import Image
 from pyowm.commons.tile import Tile
@@ -33,6 +34,8 @@ def __init__(self, API_key, config):
         self.API_key = API_key
         assert isinstance(config, dict)
         self.http_client = HttpClient(API_key, config, ROOT_AGRO_API)
+        self.geotiff_downloader_http_client = HttpClient(self.API_key, config, ROOT_DOWNLOAD_GEOTIFF_API)
+        self.png_downloader_http_client = HttpClient(self.API_key, config, ROOT_DOWNLOAD_PNG_API)
 
     def agro_api_version(self):
         return AGRO_API_VERSION
@@ -279,14 +282,14 @@ def download_satellite_image(self, metaimage, x=None, y=None, zoom=None, palette
         # polygon PNG
         if isinstance(metaimage, MetaPNGImage):
             prepared_url = metaimage.url
-            status, data = self.http_client.get_png(
+            status, data = self.png_downloader_http_client.get_png(
                 prepared_url, params=params)
             img = Image(data, metaimage.image_type)
             return SatelliteImage(metaimage, img, downloaded_on=timestamps.now(timeformat='unix'), palette=palette)
         # GeoTIF
         elif isinstance(metaimage, MetaGeoTiffImage):
             prepared_url = metaimage.url
-            status, data = self.http_client.get_geotiff(
+            status, data = self.geotiff_downloader_http_client.get_geotiff(
                 prepared_url, params=params)
             img = Image(data, metaimage.image_type)
             return SatelliteImage(metaimage, img, downloaded_on=timestamps.now(timeformat='unix'), palette=palette)

pyowm/agroapi10/uris.py

@@ -2,6 +2,8 @@
 # -*- coding: utf-8 -*-
 
 ROOT_AGRO_API = 'agromonitoring.com/agro/1.0'
+ROOT_DOWNLOAD_PNG_API = 'agromonitoring.com/image/1.0'
+ROOT_DOWNLOAD_GEOTIFF_API = 'agromonitoring.com/data/1.0'
 
 # Polygons API subset
 POLYGONS_URI = 'polygons'

pyowm/commons/cityidregistry.py

@@ -2,62 +2,87 @@
 # -*- coding: utf-8 -*-
 
 import bz2
+import sqlite3
+import tempfile
 from pkg_resources import resource_filename
 from pyowm.weatherapi25.location import Location
 
-
-CITY_ID_FILES_PATH = 'cityids/%03d-%03d.txt.bz2'
+CITY_ID_DB_PATH = 'cityids/cities.db.bz2'
 
 
 class CityIDRegistry:
 
     MATCHINGS = {
-        'exact': lambda city_name, toponym: city_name == toponym,
-        'nocase': lambda city_name, toponym: city_name.lower() == toponym.lower(),
-        'like': lambda city_name, toponym: city_name.lower() in toponym.lower(),
-        'startswith': lambda city_name, toponym: toponym.lower().startswith(city_name.lower())
+        'exact': "SELECT city_id, name, country, state, lat, lon FROM city WHERE name=?",
+        'like': r"SELECT city_id, name, country, state, lat, lon FROM city WHERE name LIKE ?"
     }
 
-    def __init__(self, filepath_regex):
-        """
-        Initialise a registry that can be used to lookup info about cities.
-
-        :param filepath_regex: Python format string that gives the path of the files
-               that store the city IDs information.
-               Eg: ``folder1/folder2/%02d-%02d.txt``
-        :type filepath_regex: str
-        :returns: a *CityIDRegistry* instance
-
-        """
-        self._filepath_regex = filepath_regex
+    def __init__(self, sqlite_db_path: str):
+        self.connection = self.__decompress_db_to_memory(sqlite_db_path)
 
     @classmethod
     def get_instance(cls):
         """
         Factory method returning the default city ID registry
         :return: a `CityIDRegistry` instance
         """
-        return CityIDRegistry(CITY_ID_FILES_PATH)
+        return CityIDRegistry(CITY_ID_DB_PATH)
 
-    def ids_for(self, city_name, country=None, matching='nocase'):
+    def __decompress_db_to_memory(self, sqlite_db_path: str):
         """
-        Returns a list of tuples in the form (long, str, str) corresponding to
-        the int IDs and relative toponyms and 2-chars country of the cities
-        matching the provided city name.
-        The rule for identifying matchings is according to the provided
-        `matching` parameter value.
+        Decompresses to memory the SQLite database at the provided path
+        :param sqlite_db_path: str
+        :return: None
+        """
+        # https://stackoverflow.com/questions/3850022/how-to-load-existing-db-file-to-memory-in-python-sqlite3
+        # https://stackoverflow.com/questions/32681761/how-can-i-attach-an-in-memory-sqlite-database-in-python
+        # https://pymotw.com/2/bz2/
+
+        # read and uncompress data from compressed DB
+        res_name = resource_filename(__name__, sqlite_db_path)
+        bz2_db = bz2.BZ2File(res_name)
+        decompressed_data = bz2_db.read()
+
+        # dump decompressed data to a temp DB
+        with tempfile.NamedTemporaryFile(mode='wb') as tmpf:
+            tmpf.write(decompressed_data)
+            tmpf_name = tmpf.name
+
+            # read temp DB to memory and return handle
+            src_conn = sqlite3.connect(tmpf_name)
+            dest_conn = sqlite3.connect(':memory:')
+            src_conn.backup(dest_conn)
+            src_conn.close()
+            return dest_conn
+
+    def __query(self, sql_query: str, *args):
+        """
+        Queries the DB with the specified SQL query
+        :param sql_query: str
+        :return: list of tuples
+        """
+        cursor = self.connection.cursor()
+        try:
+            return cursor.execute(sql_query, args).fetchall()
+        finally:
+            cursor.close()
+
+    def ids_for(self, city_name, country=None, state=None, matching='like'):
+        """
+        Returns a list of tuples in the form (city_id, name, country, state, lat, lon )
+        The rule for querying follows the provided `matching` parameter value.
         If `country` is provided, the search is restricted to the cities of
-        the specified country.
+        the specified country, and an even stricter search when `state` is provided as well
+        :param city_name: the string toponym of the city to search
         :param country: two character str representing the country where to
         search for the city. Defaults to `None`, which means: search in all
         countries.
-        :param matching: str. Default is `nocase`. Possible values:
-        `exact` - literal, case-sensitive matching,
-        `nocase` - literal, case-insensitive matching,
+        :param state: two character str representing the state where to
+        search for the city. Defaults to `None`. When not `None` also `state` must be specified
+        :param matching: str. Default is `like`. Possible values:
+        `exact` - literal, case-sensitive matching
         `like` - matches cities whose name contains, as a substring, the string
         fed to the function, case-insensitive,
-        `startswith` - matches cities whose names start with the string fed
-        to the function, case-insensitive.
         :raises ValueError if the value for `matching` is unknown
         :return: list of tuples
         """
@@ -68,43 +93,49 @@ def ids_for(self, city_name, country=None, matching='nocase'):
                              "allowed values are %s" % ", ".join(self.MATCHINGS))
         if country is not None and len(country) != 2:
             raise ValueError("Country must be a 2-char string")
-        splits = self._filter_matching_lines(city_name, country, matching)
-        return [(int(item[1]), item[0], item[4]) for item in splits]
+        if state is not None and country is None:
+            raise ValueError("A country must be specified whenever a state is specified too")
+
+        q = self.MATCHINGS[matching]
+        if matching == 'exact':
+            params = [city_name]
+        else:
+            params = ['%' + city_name + '%']
+
+        if country is not None:
+            q = q + ' AND country=?'
+            params.append(country)
+
+        if state is not None:
+            q = q + ' AND state=?'
+            params.append(state)
+
+        rows = self.__query(q, *params)
+        return rows
 
-    def locations_for(self, city_name, country=None, matching='nocase'):
+    def locations_for(self, city_name, country=None, state=None, matching='like'):
         """
-        Returns a list of Location objects corresponding to
-        the int IDs and relative toponyms and 2-chars country of the cities
-        matching the provided city name.
-        The rule for identifying matchings is according to the provided
-        `matching` parameter value.
+        Returns a list of `Location` objects
+        The rule for querying follows the provided `matching` parameter value.
         If `country` is provided, the search is restricted to the cities of
-        the specified country.
+        the specified country, and an even stricter search when `state` is provided as well
+        :param city_name: the string toponym of the city to search
         :param country: two character str representing the country where to
         search for the city. Defaults to `None`, which means: search in all
         countries.
-        :param matching: str. Default is `nocase`. Possible values:
-        `exact` - literal, case-sensitive matching,
-        `nocase` - literal, case-insensitive matching,
+        :param state: two character str representing the state where to
+        search for the city. Defaults to `None`. When not `None` also `state` must be specified
+        :param matching: str. Default is `like`. Possible values:
+        `exact` - literal, case-sensitive matching
         `like` - matches cities whose name contains, as a substring, the string
         fed to the function, case-insensitive,
-        `startswith` - matches cities whose names start with the string fed
-        to the function, case-insensitive.
         :raises ValueError if the value for `matching` is unknown
-        :return: list of `weatherapi25.location.Location` objects
+        :return: list of `Location` objects
         """
-        if not city_name:
-            return []
-        if matching not in self.MATCHINGS:
-            raise ValueError("Unknown type of matching: "
-                             "allowed values are %s" % ", ".join(self.MATCHINGS))
-        if country is not None and len(country) != 2:
-            raise ValueError("Country must be a 2-char string")
-        splits = self._filter_matching_lines(city_name, country, matching)
-        return [Location(item[0], float(item[3]), float(item[2]),
-                         int(item[1]), item[4]) for item in splits]
+        items = self.ids_for(city_name, country=country, state=state, matching=matching)
+        return [Location(item[1], item[5], item[4], item[0], country=item[2]) for item in items]
 
-    def geopoints_for(self, city_name, country=None, matching='nocase'):
+    def geopoints_for(self, city_name, country=None, state=None, matching='like'):
         """
         Returns a list of ``pyowm.utils.geo.Point`` objects corresponding to
         the int IDs and relative toponyms and 2-chars country of the cities
@@ -113,114 +144,18 @@ def geopoints_for(self, city_name, country=None, matching='nocase'):
         `matching` parameter value.
         If `country` is provided, the search is restricted to the cities of
         the specified country.
+        :param city_name: the string toponym of the city to search
         :param country: two character str representing the country where to
         search for the city. Defaults to `None`, which means: search in all
         countries.
+        :param state: two character str representing the state where to
+        search for the city. Defaults to `None`. When not `None` also `state` must be specified
         :param matching: str. Default is `nocase`. Possible values:
-        `exact` - literal, case-sensitive matching,
-        `nocase` - literal, case-insensitive matching,
+        `exact` - literal, case-sensitive matching
         `like` - matches cities whose name contains, as a substring, the string
         fed to the function, case-insensitive,
-        `startswith` - matches cities whose names start with the string fed
-        to the function, case-insensitive.
         :raises ValueError if the value for `matching` is unknown
         :return: list of `pyowm.utils.geo.Point` objects
         """
-        locations = self.locations_for(city_name, country, matching=matching)
+        locations = self.locations_for(city_name, country=country, state=state, matching=matching)
         return [loc.to_geopoint() for loc in locations]
-
-    # helper functions
-
-    def _filter_matching_lines(self, city_name, country, matching):
-        """
-        Returns an iterable whose items are the lists of split tokens of every
-        text line matched against the city ID files according to the provided
-        combination of city_name, country and matching style
-        :param city_name: str
-        :param country: str or `None`
-        :param matching: str
-        :return: list of lists
-        """
-        result = []
-
-        # find the right file to scan and extract its lines. Upon "like"
-        # matchings, just read all files
-        if matching == 'like':
-            lines = [l.strip() for l in self._get_all_lines()]
-        else:
-            filename = self._assess_subfile_from(city_name)
-            lines = [l.strip() for l in self._get_lines(filename)]
-
-        # look for toponyms matching the specified city_name and according to
-        # the specified matching style
-        for line in lines:
-            tokens = line.split(",")
-            # sometimes city names have one or more inner commas
-            if len(tokens) > 5:
-                tokens = [','.join(tokens[:-4]), *tokens[-4:]]
-            # check country
-            if country is not None and tokens[4] != country:
-                continue
-
-            # check city_name
-            if self._city_name_matches(city_name, tokens[0], matching):
-                result.append(tokens)
-
-        return result
-
-    def _city_name_matches(self, city_name, toponym, matching):
-        comparison_function = self.MATCHINGS[matching]
-        return comparison_function(city_name, toponym)
-
-    def _lookup_line_by_city_name(self, city_name):
-        filename = self._assess_subfile_from(city_name)
-        lines = self._get_lines(filename)
-        return self._match_line(city_name, lines)
-
-    def _assess_subfile_from(self, city_name):
-        c = ord(city_name.lower()[0])
-        if c < 97:  # not a letter
-            raise ValueError('Error: city name must start with a letter')
-        elif c in range(97, 103):  # from a to f
-            return self._filepath_regex % (97, 102)
-        elif c in range(103, 109):  # from g to l
-            return self._filepath_regex % (103, 108)
-        elif c in range(109, 115):  # from m to r
-            return self._filepath_regex % (109, 114)
-        elif c in range(115, 123):  # from s to z
-            return self._filepath_regex % (115, 122)
-        else:
-            raise ValueError('Error: city name must start with a letter')
-
-    def _get_lines(self, filename):
-        res_name = resource_filename(__name__, filename)
-        with bz2.open(res_name, mode='rb') as fh:
-            lines = fh.readlines()
-            if type(lines[0]) is bytes:
-                lines = map(lambda l: l.decode("utf-8"), lines)
-            return lines
-
-    def _get_all_lines(self):
-        all_lines = []
-        for city_name in ['a', 'g', 'm', 's']:  # all available city ID files
-            filename = self._assess_subfile_from(city_name)
-            all_lines.extend(self._get_lines(filename))
-        return all_lines
-
-    def _match_line(self, city_name, lines):
-        """
-        The lookup is case insensitive and returns the first matching line,
-        stripped.
-        :param city_name: str
-        :param lines: list of str
-        :return: str
-        """
-        for line in lines:
-            toponym = line.split(',')[0]
-            if toponym.lower() == city_name.lower():
-                return line.strip()
-        return None
-
-    def __repr__(self):
-        return "<%s.%s - filepath_regex=%s>" % (__name__, \
-          self.__class__.__name__, self._filepath_regex)