BUG: query with invalid dtypes should fallback to python engine · Issue #10486 · pandas-dev/pandas (original) (raw)

...
File "/Applications/anaconda/lib/python3.4/site-packages/numexpr/necompiler.py", line 629, in getType
raise ValueError("unkown type %s" % a.dtype.name)

ValueError: unkown type str128
# -*- coding: utf-8 -*-
import re
import pandas as pd

def get_gene_site(tp):
    site = re.sub('T.$', '',tp)
    gene = site.split('%%')[0]
    t = int(re.search(r'T(\d+)$',tp).group(1))
    return gene, site, t

sub_sample_list = ['Actb%%qeqwT0',
 'Actb%%qeqwT1',
 'Actb%%qeqwT2',
 'Actb%%qeqwT3',
 'Actb%%tralalaT0',
 'Actb%%tralalaT1',
 'Actb%%tralalaT2',
 'Actb%%tralalaT3',
 'Dummy%%rrrrT0',
 'Dummy%%rrrrT1',
 'Dummy%%rrrrT2',
 'Dummy%%rrrrT3',
 ]

dict_samples = {'sample':[], 'gene':[], 'site':[], 'timepoint':[]}
for tp in sub_sample_list:
    gene, site, t = get_gene_site(tp)
    dict_samples['sample'].append(tp)
    dict_samples['gene'].append(str(gene))
    dict_samples[ 'site'].append(site)
    dict_samples['timepoint'].append(t) 

df = pd.DataFrame(dict_samples , 
                columns = ['sample', 'gene', 'site', 'timepoint'])

df.query('gene = "Actb"')