MNT: Clean up some lint found by ruff

ci/download_cartopy_maps.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# Copyright (c) 2021 MetPy Developers.
+# Copyright (c) 2021 Siphon Contributors.
 """Explicitly download needed Cartopy maps."""
 from cartopy.io import config, Downloader
 

ci/filter_links.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# Copyright (c) 2021 MetPy Developers.
+# Copyright (c) 2021 Siphon Contributors.
 """Filter links from Sphinx's linkcheck."""
 import json
 import subprocess

ci/gen_versions_json.py

@@ -6,6 +6,6 @@
 
 import glob
 
-with open('versions.json', 'wt') as version_file:
-    version_strings = ','.join('"{}"'.format(d) for d in glob.glob('v*.[0-9]*'))
+with open('versions.json', 'w') as version_file:
+    version_strings = ','.join(f'"{d}"' for d in glob.glob('v*.[0-9]*'))
     version_file.write('{"versions":["latest","dev",' + version_strings + ']}\n')

pyproject.toml

@@ -62,6 +62,27 @@ extras = [
 "Bug Tracker" = "https://github.com/Unidata/siphon/issues"
 "Source Code" = "https://github.com/Unidata/siphon"
 
+[tool.doc8]
+ignore-path = ["docs/build", "docs/api/generated", "docs/_templates", "docs/examples"]
+file-encoding = "utf8"
+max-line-length = 95
+
+[tool.isort]
+line_length = 95
+known_first_party = ["siphon"]
+force_single_line = false
+reverse_relative = true
+use_parentheses = true
+force_sort_within_sections = true
+order_by_type = false
+sort_relative_in_force_sorted_sections = true
+combine_as_imports = true
+combine_star = true
+
+[tool.pytest.ini_options]
+norecursedirs = "build docs .idea"
+doctest_optionflags = "NORMALIZE_WHITESPACE"
+
 [tool.ruff]
 line-length = 95
 exclude = ["docs", "build", "src/siphon/cdmr/*_pb2.py"]

setup.cfg

@@ -7,7 +7,7 @@ max-line-length = 95
 rst-roles = class, data, doc, func, meth, mod
 rst-directives = plot, versionchanged
 known-modules = netcdf4:[netCDF4]
-exclude = docs build
+exclude = docs build src/siphon/cdmr/*_pb2.py
 select = E301 E302 E303 E304 E305 E306 I R
 ignore = F405 W503 RST902 SIM106
 per-file-ignores = src/siphon/_version.py: I900

src/siphon/catalog.py

@@ -11,13 +11,9 @@
 from datetime import datetime
 import logging
 import re
+from urllib.parse import urljoin, urlparse
 import warnings
 import xml.etree.ElementTree as ET  # noqa:N814
-try:
-    from urlparse import urljoin, urlparse
-except ImportError:
-    # Python 3
-    from urllib.parse import urljoin, urlparse
 
 from .http_util import session_manager
 from .metadata import TDSCatalogMetadata
@@ -47,10 +43,7 @@ class DatasetCollection(IndexableMapping):
     def _get_datasets_with_times(self, regex, strptime=None):
         # Set the default regex if we don't have one
         # If strptime is provided, pass the regex group named 'strptime' to strptime
-        if regex is None:
-            regex = self.default_regex
-        else:
-            regex = re.compile(regex)
+        regex = self.default_regex if regex is None else re.compile(regex)
 
         # Loop over the collection looking for keys that match our regex
         found_date = False
@@ -149,7 +142,7 @@ def filter_time_range(self, start, end, regex=None, strptime=None):
         """
         if start > end:
             warnings.warn('The provided start time comes after the end time. No data will '
-                          'be returned.', UserWarning)
+                          'be returned.', UserWarning, stacklevel=2)
         return [item[-1] for item in self._get_datasets_with_times(regex, strptime)
                 if start <= item[0] <= end]
 
@@ -289,8 +282,8 @@ def __init__(self, catalog_url):
         if 'html' in resp.headers['content-type']:
             import warnings
             new_url = self.catalog_url.replace('html', 'xml')
-            warnings.warn('URL {} returned HTML. Changing to: {}'.format(self.catalog_url,
-                                                                         new_url))
+            warnings.warn(f'URL {self.catalog_url} returned HTML. Changing to: {new_url}',
+                          stacklevel=2)
             self.catalog_url = new_url
             resp = self.session.get(self.catalog_url)
             resp.raise_for_status()
@@ -362,9 +355,8 @@ def __del__(self):
 
     def _process_dataset(self, element):
         catalog_url = ''
-        if 'urlPath' in element.attrib:
-            if element.attrib['urlPath'] == 'latest.xml':
-                catalog_url = self.catalog_url
+        if 'urlPath' in element.attrib and element.attrib['urlPath'] == 'latest.xml':
+            catalog_url = self.catalog_url
 
         ds = Dataset(element, catalog_url=catalog_url)
         self.datasets[ds.name] = ds
@@ -524,24 +516,21 @@ def resolve_url(self, catalog_url):
             resolver_xml = session_manager.urlopen(resolver_url)
             tree = ET.parse(resolver_xml)
             root = tree.getroot()
-            if 'name' in root.attrib:
-                self.catalog_name = root.attrib['name']
-            else:
-                self.catalog_name = 'No name found'
+            self.catalog_name = root.attrib.get('name', 'No name found')
             resolved_url = ''
             found = False
             for child in root.iter():
                 if not found:
                     tag_type = child.tag.split('}')[-1]
-                    if tag_type == 'dataset':
-                        if 'urlPath' in child.attrib:
-                            ds = Dataset(child)
-                            resolved_url = ds.url_path
-                            found = True
+                    if tag_type == 'dataset' and 'urlPath' in child.attrib:
+                        ds = Dataset(child)
+                        resolved_url = ds.url_path
+                        found = True
             if found:
                 return resolved_url
             else:
                 log.warning('no dataset url path found in latest.xml!')
+        return None
 
     def make_access_urls(self, catalog_url, all_services, metadata=None):
         """Make fully qualified urls for the access methods enabled on the dataset.
@@ -611,9 +600,8 @@ def download(self, filename=None):
         """
         if filename is None:
             filename = self.name
-        with self.remote_open() as infile:
-            with open(filename, 'wb') as outfile:
-                outfile.write(infile.read())
+        with self.remote_open() as infile, open(filename, 'wb') as outfile:
+            outfile.write(infile.read())
 
     def remote_open(self, mode='b', encoding='ascii', errors='ignore'):
         """Open the remote dataset for random access.
@@ -727,7 +715,8 @@ def access_with_service(self, service, use_xarray=None):
                     import xarray as xr
                     provider = lambda url: xr.open_dataset(CDMRemoteStore(url))  # noqa: E731
                 except ImportError:
-                    raise ImportError('CdmRemote access needs xarray to be installed.')
+                    raise ImportError('CdmRemote access needs xarray'
+                                      'to be installed.') from None
             else:
                 from .cdmr import Dataset as CDMRDataset
                 provider = CDMRDataset
@@ -737,13 +726,15 @@ def access_with_service(self, service, use_xarray=None):
                     import xarray as xr
                     provider = xr.open_dataset
                 except ImportError:
-                    raise ImportError('xarray needs to be installed if `use_xarray` is True.')
+                    raise ImportError('xarray needs to be installed if '
+                                      '`use_xarray` is True.') from None
             else:
                 try:
                     from netCDF4 import Dataset as NC4Dataset
                     provider = NC4Dataset
                 except ImportError:
-                    raise ImportError('OPENDAP access needs netCDF4-python to be installed.')
+                    raise ImportError('OPENDAP access needs netCDF4-python'
+                                      'to be installed.') from None
         elif service in self.ncss_service_names:
             from .ncss import NCSS
             provider = NCSS
@@ -755,7 +746,7 @@ def access_with_service(self, service, use_xarray=None):
         try:
             return provider(self.access_urls[service])
         except KeyError:
-            raise ValueError(service + ' is not available for this dataset')
+            raise ValueError(service + ' is not available for this dataset') from None
 
     __repr__ = __str__
 

src/siphon/cdmr/coveragedataset.py

@@ -29,7 +29,7 @@ def __init__(self, url):
         """Initialize CoverageDataset from a url pointing to CDMRemoteFeature endpoint."""
         super().__init__()
         warnings.warn('CoverageDataset is in early development, unsupported, and API may '
-                      'change at any time.')
+                      'change at any time.', stacklevel=2)
         self.cdmrf = CDMRemoteFeature(url)
         self.name = 'Unnamed'
         self.lon_lat_domain = None

src/siphon/cdmr/dataset.py

@@ -223,10 +223,9 @@ def _process_indices(self, ind):
         except TypeError:
             ind = [ind]
 
-        # Make sure we don't have too many things to index
-        if len(ind) > self.ndim:
-            # But allow a full slice on a scalar variable
-            if not (self.ndim == 0 and len(ind) == 1 and ind[0] == slice(None)):
+        # Make sure we don't have too many things to index, but allow a full slice on
+        # a scalar variable
+        if len(ind) > self.ndim and (self.ndim != 0 or len(ind) != 1 or ind[0] != slice(None)):
                 raise IndexError('Too many dimensions to index.')
 
         # Expand to a slice/ellipsis for every dimension
@@ -260,15 +259,9 @@ def _process_indices(self, ind):
 
                 # Adjust start and stop to handle negative indexing
                 # and partial support for slicing beyond end.
-                if i.start is None:
-                    start = 0
-                else:
-                    start = self._adjust_index(dim, i.start)
+                start = 0 if i.start is None else self._adjust_index(dim, i.start)
 
-                if i.stop is None:
-                    stop = self.shape[dim]
-                else:
-                    stop = self._adjust_index(dim, i.stop)
+                stop = self.shape[dim] if i.stop is None else self._adjust_index(dim, i.stop)
 
                 # Need to create new slice for adjusted values
                 ind[dim] = slice(start, stop, i.step)

src/siphon/cdmr/ncstream.py

@@ -10,8 +10,7 @@
 
 import numpy as np
 
-from . import cdmrfeature_pb2 as cdmrf
-from . import ncStream_pb2 as stream  # noqa
+from . import cdmrfeature_pb2 as cdmrf, ncStream_pb2 as stream
 
 MAGIC_HEADER = b'\xad\xec\xce\xda'
 MAGIC_DATA = b'\xab\xec\xce\xba'
@@ -58,7 +57,9 @@ def read_ncstream_data(fobj):
         # Handle decompressing the bytes
         if data.compress == stream.DEFLATE:
             bin_data = zlib.decompress(bin_data)
-            assert len(bin_data) == data.uncompressedSize
+            if len(bin_data) != data.uncompressedSize:
+                log.error('Uncompressed size mismatch %d vs. %d', len(bin_data),
+                          data.uncompressedSize)
         elif data.compress != stream.NONE:
             raise NotImplementedError(f'Compression type {data.compress} not implemented!')
 
@@ -140,8 +141,7 @@ def read_messages(fobj, magic_table):
         if func is not None:
             messages.append(func(fobj))
         else:
-            log.error('Unknown magic: ' + str(' '.join(f'{b: 02x}'
-                                                       for b in bytearray(magic))))
+            log.error('Unknown magic: %s', ' '.join(f'{b: 02x}' for b in bytearray(magic)))
 
     return messages
 

src/siphon/cdmr/xarray_support.py

@@ -6,6 +6,7 @@
 from xarray import Variable
 from xarray.backends.common import AbstractDataStore, BackendArray
 from xarray.core import indexing
+
 try:
     from xarray.core.utils import FrozenDict
 except ImportError:

src/siphon/http_util.py

@@ -9,12 +9,8 @@
 from io import BytesIO
 from itertools import chain
 import posixpath
+from urllib.parse import urlencode, urljoin  # noqa: F401
 import warnings
-try:
-    from urllib.parse import urlencode, urljoin  # noqa
-except ImportError:
-    from urllib import urlencode
-    from urlparse import urljoin  # noqa
 
 import requests
 
@@ -324,7 +320,7 @@ def time_range(self, start, end):
         """
         if start > end:
             warnings.warn('The provided start time comes after the end time. No data will '
-                          'be returned.', UserWarning)
+                          'be returned.', UserWarning, stacklevel=2)
         self._set_query(self.time_query, time_start=self._format_time(start),
                         time_end=self._format_time(end))
         return self

src/siphon/metadata.py

@@ -166,10 +166,7 @@ def handle_dataType(self, element):  # noqa
 class _ComplexTypes:
     @staticmethod
     def _get_tag_name(element):
-        if '}' in element.tag:
-            element_name = element.tag.split('}')[-1]
-        else:
-            element_name = element.tag
+        element_name = element.tag.split('}')[-1] if '}' in element.tag else element.tag
         return element_name
 
     @staticmethod
@@ -468,10 +465,7 @@ def __init__(self, element, metadata_in=None):
         inherited = False
         if 'inherited' in element.attrib:
             inherited = element.attrib['inherited']
-            if inherited == 'true':
-                inherited = True
-            else:
-                inherited = False
+            inherited = inherited == 'true'
 
         if metadata_in and (inherited or self._is_external_metadata_doc(element)):
             # only inherit metadata passed in if the new metadata
@@ -491,10 +485,7 @@ def __init__(self, element, metadata_in=None):
 
     @staticmethod
     def _get_tag_name(element):
-        if '}' in element.tag:
-            element_name = element.tag.split('}')[-1]
-        else:
-            element_name = element.tag
+        element_name = element.tag.split('}')[-1] if '}' in element.tag else element.tag
         return element_name
 
     @staticmethod
@@ -512,6 +503,7 @@ def _get_handler(self, handler_name):
             return getattr(self._st, handler_name)
         else:
             log.warning('cannot find handler for element %s', handler_name)
+        return None
 
     def _parse_element(self, element):
 
@@ -572,10 +564,7 @@ def _parse_documentation(self, element):
         md = self.metadata
         md.setdefault('documentation', {})
         if known or plain_doc:
-            if known:
-                doc_type = element.attrib['type']
-            else:
-                doc_type = 'generic'
+            doc_type = element.attrib['type'] if known else 'generic'
             md['documentation'].setdefault(doc_type, []).append(element.text)
         elif xlink_href_attr in element.attrib:
             title = element.attrib[xlink_title_attr]

src/siphon/ncss.py

@@ -347,7 +347,7 @@ def combine_xml_points(seq, units, handle_units):
 
 def parse_xml_dataset(elem, handle_units):
     """Create a netCDF-like dataset from XML data."""
-    points, units = zip(*[parse_xml_point(p) for p in elem.findall('point')])
+    points, units = zip(*[parse_xml_point(p) for p in elem.findall('point')], strict=False)
     # Group points by the contents of each point
     datasets = {}
     for p in points:
@@ -359,9 +359,10 @@ def parse_xml_dataset(elem, handle_units):
 
 # Handling of netCDF 3/4 from NCSS
 try:
-    from netCDF4 import Dataset
     from tempfile import NamedTemporaryFile
 
+    from netCDF4 import Dataset
+
     @response_handlers.register('application/x-netcdf')
     @response_handlers.register('application/x-netcdf4')
     def read_netcdf(data, handle_units):  # pylint:disable=unused-argument
@@ -381,7 +382,7 @@ def read_netcdf(data, handle_units):  # pylint:disable=unused-argument
 except ImportError:
     import warnings
     warnings.warn('netCDF4 module not installed. '
-                  'Will be unable to handle NetCDF returns from NCSS.')
+                  'Will be unable to handle NetCDF returns from NCSS.', stacklevel=2)
 
 
 def deletetempfile(fname):
@@ -395,7 +396,7 @@ def deletetempfile(fname):
         import warnings
         warnings.warn('temporary netcdf dataset file not deleted. '
                       'to delete temporary dataset file in the future '
-                      'be sure to use dataset.close() when finished.')
+                      'be sure to use dataset.close() when finished.', stacklevel=2)
 
 
 # Parsing of CSV data returned from NCSS