Making engine for querying more robust in case numexpr isn't installed

FAST-HEP · Nov 6, 2020 · 51cccc6 · 51cccc6
1 parent 59e1857
commit 51cccc6
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/fast_plotter/postproc/functions.py b/fast_plotter/postproc/functions.py
@@ -29,7 +29,7 @@ def __call__(self, row):
 
 
 def handle_one_df(df, query=None, replacements=[],
-                  combine_dims=[], combine_dims_ignore=None, combine_delim="__", engine_query="numexpr"):
+                  combine_dims=[], combine_dims_ignore=None, combine_delim="__", engine_query=None):
     if query:
         df.query(query, inplace=True, engine=engine_query)
     if df.empty:
@@ -51,7 +51,7 @@ def handle_one_df(df, query=None, replacements=[],
     return df
 
 
-def query(df, query, engine_query="numexpr"):
+def query(df, query, engine_query=None):
     """
     Keep only rows that satisfy requirements of the query string,
     See: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.query.html