|
@@ -0,0 +1,28 @@
|
|
|
+#! /usr/bin/env python3
|
|
|
+# filter dataframe by matching regex
|
|
|
+
|
|
|
+# https://stackoverflow.com/questions/37080612/pandas-dataframe-filter-regex/37080814#37080814
|
|
|
+
|
|
|
+import pandas as pd
|
|
|
+data = {
|
|
|
+ 'Company' : ['Ford','Ford','Ford','Ford','Chevy','Chevy'],
|
|
|
+ 'Type' : ['Mercury','Lincoln','Lincoln','Econoline','Malabu','Pickups'],
|
|
|
+ 'Profit' : [1,100,40,99,2,3]
|
|
|
+}
|
|
|
+df = pd.DataFrame(data)
|
|
|
+
|
|
|
+# print(df)
|
|
|
+
|
|
|
+# Company Type Profit
|
|
|
+# 0 Ford Mercury 1
|
|
|
+# 1 Ford Lincoln 100
|
|
|
+# 2 Ford Lincoln 40
|
|
|
+# 3 Ford Econoline 99
|
|
|
+# 4 Chevy Malabu 2
|
|
|
+# 5 Chevy Pickups 3
|
|
|
+
|
|
|
+# Now print only rows that have an "e" in the Type
|
|
|
+# preceeded by a capitol letter (that's how I choose
|
|
|
+# my cars :-))
|
|
|
+
|
|
|
+print(df[df["Type"].str.contains('[A-Z]e',regex=True)])
|