Update WyomingUpperAir client

README.rst

@@ -49,7 +49,7 @@ still evolving (we won't break things just for fun, but many things are still ch
 work through design issues). Also, for a version `0.x.y`, we change `x` when we
 release new features, and `y` when we make a release with only bug fixes.
 
-We support Python >= 3.7.
+We support Python >= 3.11.
 
 Important Links
 ---------------
@@ -64,11 +64,10 @@ Important Links
 Dependencies
 ------------
 
-- requests>=1.2
-- numpy>=1.8
-- protobuf>=3.0.0a3
-- beautifulsoup4>=4.6
-- pandas
+- numpy>=1.25.0
+- pandas>=2.1.0
+- protobuf>=4.24.3
+- requests>=2.30.0
 
 Developer Dependencies
 ----------------------

ci/Prerelease

@@ -1,4 +1,3 @@
-beautifulsoup4>=0.0.dev0
 numpy>=0.0.dev0
 pandas>=0.0.dev0
 protobuf>=0.0.dev0

ci/requirements.txt

@@ -1,4 +1,3 @@
-beautifulsoup4==4.15.0
 numpy==2.4.6
 pandas==3.0.3
 protobuf==7.35.1

docs/installguide.rst

@@ -11,7 +11,7 @@ years. For Python itself, that means supporting the last two minor releases.
 Siphon currently supports the following versions of required dependencies:
 
 .. literalinclude:: ../pyproject.toml
-   :start-at: beautifulsoup4
+   :start-at: numpy
    :end-at: requests
 
 Installation Instructions for NumPy can be found at:

pyproject.toml

@@ -25,7 +25,6 @@ classifiers = [
 ]
 requires-python = ">=3.11"
 dependencies = [
-    "beautifulsoup4>=4.9.1",
     "numpy>=1.25.0",
     "pandas>=2.1.0",
     "protobuf>=4.24.3",

src/siphon/simplewebservice/wyoming.py

@@ -3,29 +3,23 @@
 # SPDX-License-Identifier: BSD-3-Clause
 """Read upper air data from the Wyoming archives."""
 
-from datetime import datetime
 from io import StringIO
 import warnings
 
-from bs4 import BeautifulSoup
 import numpy as np
 import pandas as pd
+import requests
 
 from .._tools import get_wind_components
 from ..http_util import HTTPEndPoint
 
 
-def _safe_float(s):
-    """Convert to float, handling ****** as a string for missing."""
-    return pd.NA if all(c == '*' for c in s) or s == '-9999.0' else float(s)
-
-
 class WyomingUpperAir(HTTPEndPoint):
     """Download and parse data from the University of Wyoming's upper air archive."""
 
     def __init__(self):
         """Set up endpoint."""
-        super().__init__('http://weather.uwyo.edu/cgi-bin/sounding')
+        super().__init__('http://weather.uwyo.edu/wsgi')
 
     @classmethod
     def request_data(cls, time, site_id, recalc=False, **kwargs):
@@ -79,41 +73,26 @@ def _get_data(self, time, site_id, recalc=False):
 
         """
         raw_data = self._get_data_raw(time, site_id, recalc=recalc)
-        soup = BeautifulSoup(raw_data, 'html.parser')
-        tabular_data = StringIO(soup.find_all('pre')[0].contents[0])
-        col_names = ['pressure', 'height', 'temperature', 'dewpoint', 'direction', 'speed']
-        df = pd.read_fwf(tabular_data, widths=[7] * 8, skiprows=5,
-                         usecols=[0, 1, 2, 3, 6, 7], names=col_names)
 
-        df['u_wind'], df['v_wind'] = get_wind_components(df['speed'],
-                                                         np.deg2rad(df['direction']))
+        # Parse CSV. skipinitialspace is used to handle missing data being encoded by just
+        # having spaces in the field
+        df = pd.read_csv(StringIO(raw_data), parse_dates=['time'],
+                         date_format='%Y-%m-%d %H:%M:%S', skipinitialspace=True,
+                         na_values={'latitude':'-99.9900', 'longitude':'-99.9900'})
+        df = df.rename(columns=lambda c: c.split('_')[0])
+        df = df.rename(columns=lambda c: c.replace(' ', '_'))
+        df = df.rename(columns={'geopotential_height': 'height',
+                                'dew_point_temperature': 'dewpoint',
+                                'wind_direction': 'direction', 'wind_speed': 'speed'})
 
         # Drop any rows with all NaN values for T, Td, winds
-        df = df.dropna(subset=('temperature', 'dewpoint', 'direction', 'speed',
-                               'u_wind', 'v_wind'), how='all').reset_index(drop=True)
-
-        # Parse metadata
-        meta_data = soup.find_all('pre')[1].contents[0]
-        lines = meta_data.splitlines()
-
-        # Convert values after table into key, value pairs using the name to the left of the :
-        post_values = dict(tuple(map(str.strip, line.split(': '))) for line in lines[1:])
-
-        station = post_values.get('Station identifier', '')
-        station_number = int(post_values['Station number'])
-        sounding_time = datetime.strptime(post_values['Observation time'], '%y%m%d/%H%M')
-        latitude = _safe_float(post_values.get('Station latitude', '******'))
-        longitude = _safe_float(post_values.get('Station longitude', '******'))
-        elevation = _safe_float(post_values.get('Station elevation', '******'))
-        pw = float(post_values['Precipitable water [mm] for entire sounding'])
-
-        df['station'] = station
-        df['station_number'] = station_number
-        df['time'] = sounding_time
-        df['latitude'] = latitude
-        df['longitude'] = longitude
-        df['elevation'] = elevation
-        df['pw'] = pw
+        df = df.dropna(subset=('temperature', 'dewpoint', 'direction', 'speed'),
+                       how='all').reset_index(drop=True)
+
+        df['u_wind'], df['v_wind'] = get_wind_components(df['speed'],
+                                                         np.deg2rad(df['direction']))
+        df['station'] = site_id
+        df['height'] = df['height'].astype('float')
 
         # Add unit dictionary
         with warnings.catch_warnings():
@@ -124,16 +103,13 @@ def _get_data(self, time, site_id, recalc=False):
                         'temperature': 'degC',
                         'dewpoint': 'degC',
                         'direction': 'degrees',
-                        'speed': 'knot',
-                        'u_wind': 'knot',
-                        'v_wind': 'knot',
+                        'speed': 'm/s',
+                        'u_wind': 'm/s',
+                        'v_wind': 'm/s',
                         'station': None,
-                        'station_number': None,
                         'time': None,
                         'latitude': 'degrees',
-                        'longitude': 'degrees',
-                        'elevation': 'meter',
-                        'pw': 'millimeter'}
+                        'longitude': 'degrees'}
 
         return df
 
@@ -155,17 +131,11 @@ def _get_data_raw(self, time, site_id, recalc=False):
             text of the server response
 
         """
-        path = ('?region=naconf&TYPE=TEXT%3ALIST'
-                f'&YEAR={time:%Y}&MONTH={time:%m}&FROM={time:%d%H}&TO={time:%d%H}'
-                f'&STNM={site_id}')
+        path = f'sounding?type=TEXT%3ACSV&datetime={time:%Y-%m-%d %H:%M:%S}&id={site_id}'
         if recalc:
             path += '&REPLOT=1'
 
-        resp = self.get_path(path)
-        # See if the return is valid, but has no data
-        if resp.text.find("Can't") != -1:
-            raise ValueError(
-                f'No data available for {time:%Y-%m-%d %HZ} '
-                f'for station {site_id}.')
-
-        return resp.text
+        try:
+            return self.get_path(path).text
+        except requests.HTTPError as e:
+            raise ValueError from e