I am running a query for a dataframe with a column of type String that returns rows that contain a certain the world ‘select’:

df.query('code.str.contains("select")')

However, when I try to run the same query in the pandasgui as a filter that takes a query expression, it is listed as a failed filter and does not evaluate it.

Picture of the query in pandasgui and columns for the dataframe

I tried a combination of " or ' around contains, and no ` around the column name but can’t seem to find a method that enables the filter. Does anyone know if there’s a way to query a column in a dataframe through pandasgui to search a string value for a specific substring? I’d appreciate any help!

I also ended up searching through the pandasgui source code, and a good starting point would be in the pandasgui/store.py file, where the authors used the apply_filters() method in numerous places. It looks to me like the query is considered failed after catching an exception in the apply_filters() method, which is where self.filters[index].failed is set to True from False.

Edit: It should be noted that boolean expressions like code == 'select' works, but only returns rows that are exactly ‘select’. I am looking to find rows that contain the word ‘select’ with a query in the filters tab of pandasgui.

@status_message_decorator("Applying filters...")
def apply_filters(self):
    df = self.df_unfiltered.copy()
    df['_temp_range_index'] = df.reset_index().index

    for ix, filt in enumerate(self.filters):
        if filt.enabled and not filt.failed:
            try:
                df = df.query(filt.expr)
                # Handle case where filter returns only one row
                if isinstance(df, pd.Series):
                    df = df.to_frame().T
            except Exception as e:
                self.filters[ix].failed = True
                logger.exception(e)

    # self.filtered_index_map is used elsewhere to map unfiltered index to filtered index
    self.filtered_index_map = df['_temp_range_index'].reset_index(drop=True)
    df = df.drop('_temp_range_index', axis=1)

    self.df = df
    self.data_changed()