MAINT: More work on broken APIs

Author: bashtage

pandas_datareader/bankofcanada.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals
 
 from pandas_datareader.base import _BaseReader
-from pandas_datareader.compat import string_types
 
 
 class BankOfCanadaReader(_BaseReader):
@@ -16,7 +15,7 @@ class BankOfCanadaReader(_BaseReader):
     @property
     def url(self):
         """API URL"""
-        if not isinstance(self.symbols, string_types):
+        if not isinstance(self.symbols, str):
             raise ValueError("data name must be string")
 
         return "{0}/{1}/csv".format(self._URL, self.symbols)

pandas_datareader/base.py

@@ -1,4 +1,5 @@
 import datetime
+from io import StringIO
 import time
 from urllib.parse import urlencode
 import warnings
@@ -13,13 +14,6 @@
     _init_session,
     _sanitize_dates,
 )
-from pandas_datareader.compat import (
-    PANDAS_0230,
-    StringIO,
-    binary_type,
-    bytes_to_str,
-    string_types,
-)
 
 
 class _BaseReader(object):
@@ -124,8 +118,8 @@ def _read_url_as_StringIO(self, url, params=None):
                 "{} request returned no data; check URL for invalid "
                 "inputs: {}".format(service, self.url)
             )
-        if isinstance(text, binary_type):
-            out.write(bytes_to_str(text))
+        if isinstance(text, bytes):
+            out.write(text.decode("utf-8"))
         else:
             out.write(text)
         out.seek(0)
@@ -248,7 +242,7 @@ def _get_params(self, *args, **kwargs):
     def read(self):
         """Read data"""
         # If a single symbol, (e.g., 'GOOG')
-        if isinstance(self.symbols, (string_types, int)):
+        if isinstance(self.symbols, (str, int)):
             df = self._read_one_data(self.url, params=self._get_params(self.symbols))
         # Or multiple symbols, (e.g., ['GOOG', 'AAPL', 'MSFT'])
         elif isinstance(self.symbols, DataFrame):
@@ -280,11 +274,8 @@ def _dl_mult_symbols(self, symbols):
                 df_na[:] = np.nan
                 for sym in failed:
                     stocks[sym] = df_na
-            if PANDAS_0230:
                 result = concat(stocks, sort=True).unstack(level=0)
-            else:
-                result = concat(stocks).unstack(level=0)
-            result.columns.names = ["Attributes", "Symbols"]
+                result.columns.names = ["Attributes", "Symbols"]
             return result
         except AttributeError as exc:
             # cannot construct a panel with just 1D nans indicating no data

pandas_datareader/compat/__init__.py

@@ -7,27 +7,13 @@
 from pandas.io import common as com
 from pandas.testing import assert_frame_equal
 
-PANDAS_VERSION = version.parse(pd.__version__)
-
-PANDAS_0210 = PANDAS_VERSION >= version.parse("0.21.0")
-PANDAS_0220 = PANDAS_VERSION >= version.parse("0.22.0")
-PANDAS_0230 = PANDAS_VERSION >= version.parse("0.23.0")
-
 __all__ = [
     "HTTPError",
     "StringIO",
-    "PANDAS_0210",
-    "PANDAS_0220",
-    "PANDAS_0230",
     "get_filepath_or_buffer",
-    "str_to_bytes",
-    "string_types",
     "assert_frame_equal",
     "is_list_like",
     "is_number",
-    "lmap",
-    "lrange",
-    "concat",
     "reduce",
 ]
 
@@ -47,34 +33,3 @@ def get_filepath_or_buffer(filepath_or_buffer, encoding=None, compression=None):
             filepath_or_buffer, encoding=encoding, compression=None
         )
         return tmp
-
-
-string_types = (str,)
-binary_type = bytes
-
-
-def str_to_bytes(s, encoding=None):
-    if isinstance(s, bytes):
-        return s
-    return s.encode(encoding or "ascii")
-
-
-def bytes_to_str(b, encoding=None):
-    return b.decode(encoding or "utf-8")
-
-
-def lmap(*args, **kwargs):
-    return list(map(*args, **kwargs))
-
-
-def lrange(*args, **kwargs):
-    return list(range(*args, **kwargs))
-
-
-def concat(*args, **kwargs):
-    """
-    Shim to wokr around sort keyword
-    """
-    if not PANDAS_0230 and "sort" in kwargs:
-        del kwargs["sort"]
-    return pd.concat(*args, **kwargs)

pandas_datareader/enigma.py

@@ -3,7 +3,7 @@
 
 import pandas as pd
 
-from pandas_datareader.base import _BaseReader, string_types
+from pandas_datareader.base import _BaseReader
 from pandas_datareader.compat import StringIO
 from pandas_datareader.exceptions import DEP_ERROR_MSG, ImmediateDeprecationError
 
@@ -71,7 +71,7 @@ def __init__(
             self._api_key = api_key
 
         self._dataset_id = dataset_id
-        if not isinstance(self._dataset_id, string_types):
+        if not isinstance(self._dataset_id, str):
             raise ValueError(
                 "The Enigma dataset_id must be a string (ex: "
                 "'bedaf052-5fcd-4758-8d27-048ce8746c6a')"

pandas_datareader/eurostat.py

@@ -3,7 +3,6 @@
 import pandas as pd
 
 from pandas_datareader.base import _BaseReader
-from pandas_datareader.compat import string_types
 from pandas_datareader.io.sdmx import _read_sdmx_dsd, read_sdmx
 
 
@@ -15,7 +14,7 @@ class EurostatReader(_BaseReader):
     @property
     def url(self):
         """API URL"""
-        if not isinstance(self.symbols, string_types):
+        if not isinstance(self.symbols, str):
             raise ValueError("data name must be string")
 
         q = "{0}/data/{1}/?startperiod={2}&endperiod={3}"
@@ -24,7 +23,7 @@ def url(self):
     @property
     def dsd_url(self):
         """API DSD URL"""
-        if not isinstance(self.symbols, string_types):
+        if not isinstance(self.symbols, str):
             raise ValueError("data name must be string")
 
         return "{0}/datastructure/ESTAT/DSD_{1}".format(self._URL, self.symbols)

pandas_datareader/famafrench.py

@@ -6,7 +6,7 @@
 from pandas import read_csv, to_datetime
 
 from pandas_datareader.base import _BaseReader
-from pandas_datareader.compat import StringIO, lmap
+from pandas_datareader.compat import StringIO
 
 _URL = "http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/"
 _URL_PREFIX = "ftp/"
@@ -162,4 +162,4 @@ def get_available_datasets(self):
             if ds.startswith(_URL_PREFIX) and ds.endswith(_URL_SUFFIX)
         ]
 
-        return lmap(lambda x: x[len(_URL_PREFIX) : -len(_URL_SUFFIX)], datasets)
+        return list(map(lambda x: x[len(_URL_PREFIX) : -len(_URL_SUFFIX)], datasets))

pandas_datareader/io/sdmx.py

@@ -7,7 +7,7 @@
 
 import pandas as pd
 
-from pandas_datareader.compat import HTTPError, str_to_bytes
+from pandas_datareader.compat import HTTPError
 from pandas_datareader.io.util import _read_content
 
 _STRUCTURE = "{http://www.sdmx.org/resources/sdmxml/schemas/v2_1/structure}"
@@ -234,8 +234,10 @@ def _read_zipped_sdmx(path_or_buf):
     """Unzipp data contains SDMX-XML"""
     data = _read_content(path_or_buf)
 
+    if not isinstance(data, bytes):
+        data = data.encode("ascii")
     zp = BytesIO()
-    zp.write(str_to_bytes(data))
+    zp.write(data)
     f = zipfile.ZipFile(zp)
     files = f.namelist()
     assert len(files) == 1

pandas_datareader/io/util.py

@@ -2,7 +2,7 @@
 
 import os
 
-from pandas_datareader.compat import get_filepath_or_buffer, string_types
+from pandas_datareader.compat import get_filepath_or_buffer
 
 
 def _read_content(path_or_buf):
@@ -12,7 +12,7 @@ def _read_content(path_or_buf):
 
     filepath_or_buffer = get_filepath_or_buffer(path_or_buf)[0]
 
-    if isinstance(filepath_or_buffer, string_types):
+    if isinstance(filepath_or_buffer, str):
         try:
             exists = os.path.exists(filepath_or_buffer)
         except (TypeError, ValueError):

pandas_datareader/moex.py

@@ -1,9 +1,10 @@
 import datetime as dt
+from io import StringIO
 
 import pandas as pd
 
 from pandas_datareader.base import _DailyBaseReader
-from pandas_datareader.compat import StringIO, binary_type, concat, is_list_like
+from pandas_datareader.compat import is_list_like
 
 
 class MoexReader(_DailyBaseReader):
@@ -108,7 +109,7 @@ def _get_metadata(self):
                     "{} request returned no data; check URL for invalid "
                     "inputs: {}".format(service, self.__url_metadata)
                 )
-            if isinstance(text, binary_type):
+            if isinstance(text, bytes):
                 text = text.decode("windows-1251")
 
             header_str = "secid;boardid;"
@@ -200,7 +201,7 @@ def read_all_boards(self):
                 "check URL or correct a date".format(self.__class__.__name__)
             )
         elif len(dfs) > 1:
-            b = concat(dfs, axis=0, join="outer", sort=True)
+            b = pd.concat(dfs, axis=0, join="outer", sort=True)
         else:
             b = dfs[0]
         return b
@@ -228,7 +229,7 @@ def _read_url_as_String(self, url, params=None):
                 "{} request returned no data; check URL for invalid "
                 "inputs: {}".format(service, self.url)
             )
-        if isinstance(text, binary_type):
+        if isinstance(text, bytes):
             text = text.decode("windows-1251")
         return text
 

pandas_datareader/naver.py

@@ -3,7 +3,6 @@
 
 import numpy as np
 from pandas import DataFrame, to_datetime
-from six import string_types
 
 from pandas_datareader.base import _DailyBaseReader
 
@@ -32,7 +31,7 @@ def __init__(
         get_actions=False,
         adjust_dividends=True,
     ):
-        if not isinstance(symbols, string_types):
+        if not isinstance(symbols, str):
             raise NotImplementedError("Bulk-fetching is not implemented")
 
         super().__init__(