From 6ce01a828deba5283fd7493b9ab35154d5bf133a Mon Sep 17 00:00:00 2001
From: Nico Matentzoglu <nicolas.matentzoglu@gmail.com>
Date: Sat, 9 Nov 2024 17:04:25 +0200
Subject: [PATCH] Handle two warnings that pollute the output of sssom-py CLI
 (#561)

See commits for a more detailed description of the changes;

This PR should not introduce any functional changes.
---
 src/sssom/cli.py     |  4 ++++
 src/sssom/parsers.py |  4 +---
 src/sssom/util.py    | 16 ++++++++++------
 3 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/src/sssom/cli.py b/src/sssom/cli.py
index 9762d700..3bd8b72a 100644
--- a/src/sssom/cli.py
+++ b/src/sssom/cli.py
@@ -54,6 +54,7 @@
     filter_redundant_rows,
     invert_mappings,
     merge_msdf,
+    pandas_set_no_silent_downcasting,
     reconcile_prefix_and_data,
     remove_unmatched,
     sort_df_rows_columns,
@@ -126,6 +127,9 @@
 def main(verbose: int, quiet: bool):
     """Run the SSSOM CLI."""
     logger = _logging.getLogger()
+
+    pandas_set_no_silent_downcasting()
+
     if verbose >= 2:
         logger.setLevel(level=_logging.DEBUG)
     elif verbose == 1:
diff --git a/src/sssom/parsers.py b/src/sssom/parsers.py
index c2348fe3..28f96b43 100644
--- a/src/sssom/parsers.py
+++ b/src/sssom/parsers.py
@@ -424,9 +424,7 @@ def from_sssom_dataframe(
     # Need to revisit this solution.
     # This is to address: A value is trying to be set on a copy of a slice from a DataFrame
     if CONFIDENCE in df.columns:
-        df2 = df.copy()
-        df2[CONFIDENCE].replace(r"^\s*$", np.nan, regex=True, inplace=True)
-        df = df2
+        df.replace({CONFIDENCE: r"^\s*$"}, np.nan, regex=True, inplace=True)
 
     mapping_set = _get_mapping_set_from_df(df=df, meta=meta)
     doc = MappingSetDocument(mapping_set=mapping_set, converter=converter)
diff --git a/src/sssom/util.py b/src/sssom/util.py
index f506c44a..6684f9a1 100644
--- a/src/sssom/util.py
+++ b/src/sssom/util.py
@@ -158,14 +158,9 @@ def from_mapping_set_document(cls, doc: MappingSetDocument) -> "MappingSetDataFr
         df = pd.DataFrame(get_dict_from_mapping(mapping) for mapping in doc.mapping_set.mappings)
         meta = _extract_global_metadata(doc)
 
-        if pandas_version >= (2, 0, 0):
-            # For pandas >= 2.0.0, use the 'copy' parameter
-            df = df.infer_objects(copy=False)
-        else:
-            # For pandas < 2.0.0, call 'infer_objects()' without any parameters
-            df = df.infer_objects()
         # remove columns where all values are blank.
         df.replace("", np.nan, inplace=True)
+        df = df.infer_objects()
         df.dropna(axis=1, how="all", inplace=True)  # remove columns with all row = 'None'-s.
 
         slots = _get_sssom_schema_object().dict["slots"]
@@ -1493,3 +1488,12 @@ def safe_compress(uri: str, converter: Converter) -> str:
     :return: A CURIE
     """
     return converter.compress_or_standardize(uri, strict=True)
+
+
+def pandas_set_no_silent_downcasting(no_silent_downcasting=True):
+    """Set pandas future.no_silent_downcasting option. Context https://github.com/pandas-dev/pandas/issues/57734."""
+    try:
+        pd.set_option("future.no_silent_downcasting", no_silent_downcasting)
+    except KeyError:
+        # Option does not exist in this version of pandas
+        pass