closes #149. Removed default loggers (#152)

HISTORY.rst

@@ -1,3 +1,12 @@
+Mass_Composition 0.4.11 (2024-05-14)
+====================================
+
+Other Tasks
+-----------
+
+- Removed default loggers. (#149)
+
+
 Mass_Composition 0.4.10 (2024-05-13)
 ====================================
 

elphick/mass_composition/mass_composition.py

@@ -1012,12 +1012,17 @@ def __truediv__(self, other: 'MassComposition') -> 'MassComposition':
 
         return res
 
+    def __eq__(self, other):
+        if isinstance(other, MassComposition):
+            return self.__dict__ == other.__dict__
+        return False
+
     @staticmethod
     def _check_cols_in_data_cols(cols: List[str], cols_data: List[str]):
         for col in cols:
             if (col is not None) and (col not in cols_data):
                 msg: str = f"{col} not in the data columns: {cols_data}"
-                logging.error(msg)
+                logging.getLogger(__name__).error(msg)
                 raise IndexError(msg)
 
     @staticmethod

elphick/mass_composition/mc_status.py

@@ -6,14 +6,15 @@
 
 class Status:
     def __init__(self, df_oor: pd.DataFrame):
+        self._logger = logging.getLogger(__name__)
         self.oor: pd.DataFrame = df_oor
         self.num_oor: int = len(df_oor)
         self.failing_components: List[str] = list(df_oor.dropna(axis=1).columns) if self.num_oor > 0 else None
 
     @property
     def ok(self) -> bool:
         if self.num_oor > 0:
-            logging.warning(f'{self.num_oor} OOR records exist.')
+            self._logger.warning(f'{self.num_oor} OOR records exist.')
         return True if self.num_oor == 0 else False
 
     def __str__(self) -> str:

elphick/mass_composition/network.py

@@ -978,13 +978,13 @@ def _check_indexes(cls, streams):
                     s: str = stream.name
                     tmp_nans: pd.Series = stream_nans.query('stream==@s').sum(axis=1)
                     if tmp_nans.iloc[0] > 0:
-                        logging.debug(f'The {s} stream has missing coarse sizes')
+                        logger.debug(f'The {s} stream has missing coarse sizes')
                         first_zero_index = tmp_nans.loc[tmp_nans == 0].index[0]
                         if tmp_nans[tmp_nans.index <= first_zero_index].sum() > 0:
-                            logging.debug(f'The {s} stream has missing sizes requiring interpolation')
+                            logger.debug(f'The {s} stream has missing sizes requiring interpolation')
                             raise NotImplementedError('Coming soon - we need interpolation!')
                         else:
-                            logging.debug(f'The {s} stream has missing coarse sizes only')
+                            logger.debug(f'The {s} stream has missing coarse sizes only')
                             stream_df = df_streams_full.loc[:, (slice(None), s)].droplevel(-1, axis=1).fillna(0)
                             # recreate the stream from the dataframe
                             stream.set_data(stream_df)

elphick/mass_composition/utils/loader.py

@@ -11,15 +11,23 @@
 from elphick.mass_composition.utils.parallel import TqdmParallel
 from elphick.mass_composition.utils.pd_utils import column_prefix_counts, column_prefixes
 
+logger = logging.getLogger(__name__)
+
 
 def create_mass_composition(stream_data: Tuple[Union[int, str], pd.DataFrame],
                             interval_edges: Optional[Union[Iterable, int]] = None) -> Tuple[
-                            Union[int, str], MassComposition]:
+                                Union[int, str], MassComposition]:
     stream, data = stream_data
-    if interval_edges is not None:
-        return stream, MassComposition(data=data, name=stream).resample_1d(interval_edges=interval_edges)
-    else:
-        return stream, MassComposition(data=data, name=stream)
+    res = None
+    try:
+        if interval_edges is not None:
+            res = stream, MassComposition(data=data, name=stream).resample_1d(interval_edges=interval_edges)
+        else:
+            res = stream, MassComposition(data=data, name=stream)
+    except Exception as e:
+        logger.error(f"Error creating MassComposition object for {stream}: {e}")
+
+    return res
 
 
 def streams_from_dataframe(df: pd.DataFrame,
@@ -40,11 +48,10 @@ def streams_from_dataframe(df: pd.DataFrame,
     Returns:
 
     """
-    logger: logging.Logger = logging.getLogger(__name__)
-
     stream_data: Dict[str, pd.DataFrame] = {}
     index_names: List[str] = []
     if mc_name_col:
+        logger.debug("Creating MassComposition objects by name column.")
         if mc_name_col in df.index.names:
             index_names = df.index.names
             df.reset_index(mc_name_col, inplace=True)
@@ -58,6 +65,7 @@ def streams_from_dataframe(df: pd.DataFrame,
             df.reset_index(inplace=True)
             df.set_index(index_names, inplace=True)
     else:
+        logger.debug("Creating MassComposition objects by column prefixes.")
         # wide case - find prefixes where there are at least 3 columns
         prefix_counts = column_prefix_counts(df.columns)
         prefix_cols = column_prefixes(df.columns)
@@ -69,6 +77,7 @@ def streams_from_dataframe(df: pd.DataFrame,
                     columns={col: col.replace(f'{prefix}_', '') for col in df.columns})
 
     if interval_edges is not None:
+        logger.debug("Resampling MassComposition objects to new interval edges.")
         # unify the edges - this will also interp missing grades
         if not isinstance(df.index, pd.IntervalIndex):
             raise NotImplementedError(f"The index `{df.index}` of the dataframe is not a pd.Interval. "
@@ -82,8 +91,8 @@ def streams_from_dataframe(df: pd.DataFrame,
             indx = pd.IntervalIndex.from_arrays(left=all_edges[0:-1], right=all_edges[1:])
             interval_edges = _upsample_grid_by_factor(indx=indx, factor=interval_edges)
 
-    with TqdmParallel(desc="Creating MassComposition objects", n_jobs=n_jobs, prefer="processes",
-                      total=len(stream_data)) as p:
+    with TqdmParallel(desc="Creating MassComposition objects", n_jobs=n_jobs,
+                      prefer=None, total=len(stream_data)) as p:
         res = p(delayed(create_mass_composition)(stream_data, interval_edges) for stream_data in stream_data.items())
     res = dict(res)
 

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "mass-composition"
-version = "0.4.10"
+version = "0.4.11"
 description = "For managing multi-dimensional mass-composition datasets, supporting weighted mathematical operations and visualisation."
 authors = ["Greg <greg@elphick.com.au>"]
 packages = [{ include = "elphick/mass_composition" }]