Pandas pd.eval not correctly evaluating equality of local string variables

Hi,

The following evaluates False when I'd expect it to evaluate True.

import pandas as pd
import numpy as np
df=pd.DataFrame({'val':[1,2,3]})
print(df.eval('@fruit=="apple"',local_dict={'fruit':'apple'}))

This works as expected

df=pd.DataFrame({'val':[1,2,3]})
print(df.eval('val>@x',local_dict={'x':2}))

As does this:

df=pd.DataFrame({'val':[1,2,3]})
print(df.eval('@fruit',local_dict={'fruit':'apple'}))

I was hoping to utilize the '.attrs' property of dataframes and the 'local_dict' feature of df.eval to minimize the memory footprint of my analysis.

Thanks,