added code search (#160)

* added code search

* added single code table fetch

* update table fetching for codes

* added codes test

* run pre-commit

* update docs

docs/api.rst

@@ -24,6 +24,14 @@ Top-level functions
    magic_numbers
    file_pattern
 
+Codes
+=====
+
+.. autosummary::
+   :toctree: generated/
+
+   codes.get_dict
+   codes.search
 
 Physics
 =======

docs/whats_new.rst

@@ -11,6 +11,11 @@ What's New
 (Unreleased)
 ------------
 
+New Features
+~~~~~~~~~~~~
+
+- Added :py:meth:`codes.search` for searching in the code table using keyword arguments (:pull:`160`).
+
 Internal Changes
 ~~~~~~~~~~~~~~~~
 

pyremo/codes.py

@@ -40,16 +40,30 @@
 
 """
 
-# from .tables import codes as code_table  # code_table, read_table
-
 import numpy as np
 import pandas as pd
 
 from .tables import codes
 
+tables = codes.tables
+
 # table = pd.concat([table for name, table in code_table.items()])
 
 
+def _code_from_varname(varname):
+    """Returns a code from a varname.
+
+    Used typically for varnames create by a `cdo -f nc copy` command
+    on IEG files.
+    """
+    if "var" in varname:
+        import re
+
+        return int(re.findall("[0-9]+", varname)[0])
+    else:
+        return None
+
+
 def get_dict(id):
     """Returns a dictionary with variable info.
 
@@ -124,15 +138,52 @@ def get_dict_by_code(code):
     return dict
 
 
-def _code_from_varname(varname):
-    """Returns a code from a varname.
+def _search_df(df, **kwargs):
+    """Search dataframe by arbitray conditions
+
+    Converts kwargs to pandas search conditions. If kwargs is a list,
+    pandas isin is used as condition.
 
-    Used typically for varnames create by a `cdo -f nc copy` command
-    on IEG files.
     """
-    if "var" in varname:
-        import re
+    df = df.reset_index()
+    condition_list = []
+    for key, item in kwargs.items():
+        if isinstance(item, (list, tuple)):
+            cond = "(df['{0}'].isin({1}))".format(key, repr(item))
+        else:
+            cond = "(df['{0}'] == {1})".format(key, repr(item))
+        condition_list.append(cond)
+    conditions = " & ".join(condition_list)
+    return df[eval(conditions)]
+
+
+def search(**kwargs):
+    """Returns a tables with variabl meta data.
+
+    Searches the code table by arbitrary attributes.
+    All search parameters can also be iteratables.
+
+    Parameters
+    ----------
+    code: int
+        Variable code.
+    variable: str
+        Variable name (REMO standard).
+    cf_name: str
+        CF Variable name (Climate and Forecast convention).
+    description: str
+        Variable description (REMO standard).
+    units: str
+        Unit (REMO standard).
+    time_cell_method: str
+        Time cell method for standard Remo output (point or mean).
+
+
+    Returns
+    -------
+    df : pd.DataFrame
+        Search result.
 
-        return int(re.findall("[0-9]+", varname)[0])
-    else:
-        return None
+    """
+    table = pd.concat(codes.tables.values())
+    return _search_df(table, **kwargs)

pyremo/tables/_resources.py

@@ -1,3 +1,5 @@
+from os import path as op
+
 import pandas as pd
 import pooch
 
@@ -69,12 +71,19 @@ def read_remo_domain_tables():
     }
 
 
+# def read_remo_code_tables():
+#     resource = CODE_RESOURCE
+#     return {
+#         table.split(".")[0]: read_remote_table(table, resource, index_col="code")
+#         for table in resource.registry.keys()
+#     }
+
+
 def read_remo_code_tables():
-    resource = CODE_RESOURCE
-    return {
-        table.split(".")[0]: read_remote_table(table, resource, index_col="code")
-        for table in resource.registry.keys()
-    }
+    filename = pooch.retrieve(
+        op.join(base_url, "code-list", "table.csv"), known_hash=None, path=cache_url
+    )
+    return {"codes": pd.read_csv(filename, index_col="code")}
 
 
 def read_remo_vc_tables():

tests/test_codes.py

@@ -0,0 +1,33 @@
+import pandas as pd
+
+from pyremo import codes
+
+
+def test_codes():
+    expected = {
+        "variable": "T",
+        "description": "temperature",
+        "units": "K",
+        "layer": 110.0,
+        "time_cell_method": None,
+        "cf_name": "ta",
+        "code": 130,
+    }
+    assert codes.get_dict(130) == expected
+    assert codes.get_dict("T") == expected
+
+
+def test_code_search():
+    df = pd.DataFrame(
+        {
+            "code": 167,
+            "variable": "TEMP2",
+            "description": "2m temperature",
+            "units": "K",
+            "layer": 1.0,
+            "time_cell_method": "mean",
+            "cf_name": "tas",
+        },
+        index=[0],
+    )
+    assert df.equals(codes.search(code=167).reset_index(drop=True))