Merge branch 'parquet' into 'dev'

Add Apache Parquet export support

See merge request mass-spectrometry/corems!229

Author: corilo

README.md

@@ -99,6 +99,7 @@ As an open source project, CoreMS welcomes contributions of all forms. See the [
 - Pandas data frame (can be saved using pickle, h5, etc)
 - Text Files (.csv, tab separated .txt, etc)
 - Microsoft Excel (xlsx)
+- Apache Parquet (.parquet)
 - Automatic JSON for metadata storage and reuse
 - Self-containing Hierarchical Data Format (.hdf5) including raw data and time-series data-point for processed data-sets with all associated metadata stored as json attributes
 
@@ -368,6 +369,8 @@ mass_spectrum_obj.to_csv("filename")
 mass_spectrum_obj.to_hdf("filename")
 # to pandas Datarame pickle
 mass_spectrum_obj.to_pandas("filename")
+# to parquet
+mass_spectrum_obj.to_parquet("filename")
 
 # Extract data as a pandas Dataframe
 df = mass_spectrum_obj.to_dataframe()

corems/mass_spectra/factory/GC_Class.py

@@ -556,6 +556,31 @@ def to_pandas(self, out_file_path, write_metadata=True, id_label="corems:"):
 
         return out_file_path.with_suffix(".pkl")
 
+    def to_parquet(self, out_file_path, write_metadata=True, id_label="corems:"):
+        """Export the GC-MS data to a Parquet file.
+
+        Parameters
+        ----------
+        out_file_path : str, pathlib.Path, or s3path.S3Path
+            Path object containing the file location.
+        write_metadata : bool, optional
+            If True, write the metadata. Defaults to True.
+        id_label : str, optional
+            Label of the ID. Defaults to 'corems:'.
+
+        Notes
+        -----
+        Requires ``pyarrow`` (or ``fastparquet``) for pandas Parquet I/O.
+        """
+
+        if isinstance(out_file_path, str):
+            out_file_path = Path(out_file_path)
+
+        exportMS = LowResGCMSExport(out_file_path, self)
+        exportMS.to_parquet(id_label=id_label, write_metadata=write_metadata)
+
+        return out_file_path.with_suffix(".parquet")
+
     def to_dataframe(self, id_label="corems:"):
         """Export the GC-MS data to a Pandas dataframe.