如何从 python 中的大量列表中拆分所有其他元素

Question

I have a huge list in python in a single column and i need to split all the fruits, colours etc from the list and make a dafaframe.

example
details=['banana', 
'type:', 
'fruit', 
'color:', 
'yellow', 
'orange', 
'type:', 
'fruit', 
'color:', 
'orange',
'blueberry', 
'type:', 
'fruit', 
'color:', 
'blue']

what I'm expecting to achieve is if I extract all color from above then the result should be a single column of list like below.

Out[1]:

['yellow',
 'orange',
 'blue']

Answer 1

details=['banana',
         'type:',
         'fruit',
         'color:',
         'yellow',
         'orange',
         'type:',
         'fruit',
         'color:',
         'orange',
         'blueberry',
         'type:',
         'fruit',
         'color:',
        'blue']
# split each fruit as a list and index colors 
details = [details[i:i+5] for i in range(0,len(details),5)]
fruits = []
color = []
for i in details:
  fruits.append(i[0])
  color.append(i[4])

Answer 2

One possible approach if the data-structure doesn't change, is using list comprehension:

eg: [details[i+1] for i, x in enumerate(details) if x == 'color:']

Full code:

details=['banana',
'type:',
'fruit',
'color:',
'yellow',
'orange',
'type:',
'fruit',
'color:',
'orange',
'blueberry',
'type:',
'fruit',
'color:',
'blue']

colors = [details[i+1] for i, x in enumerate(details) if x == 'color:']
fruits = [details[i-1] for i, x in enumerate(details) if x == 'type:']
types = [details[i+1] for i, x in enumerate(details) if x == 'type:']

print('fruits: ', fruits)
print('types: ', types)
print('colors: ', colors)

Output:

fruits:  ['banana', 'orange', 'blueberry']
types:  ['fruit', 'fruit', 'fruit']
colors:  ['yellow', 'orange', 'blue']

Or as Dataframe

# to make datafame
import pandas as pd

df = pd.DataFrame()
df['fruits'] = [details[i-1] for i, x in enumerate(details) if x == 'type:']
df['types'] = [details[i+1] for i, x in enumerate(details) if x == 'type:']
df['colors'] = [details[i+1] for i, x in enumerate(details) if x == 'color:']

print(df)

Output:

      fruits  types  colors
0     banana  fruit  yellow
1     orange  fruit  orange
2  blueberry  fruit    blue

Explanation:

1. value of colour comes after the string color , so find the indices of that string in the list
2. add 1 to this index to get the value of color
3. use this with list comprehension to get desired output array
4. repeat for other values

Answer 3

If you want to extract only colors, you can use the package colour - https://github.com/vaab/colour

In [7]: from colour import Color

In [8]: details
Out[8]:
['banana',
 'type:',
 'fruit',
 'color:',
 'yellow',
 'orange',
 'type:',
 'fruit',
 'color:',
 'orange',
 'blueberry',
 'type:',
 'fruit',
 'color:',
 'blue']

In [9]: colors_only = set()

In [10]: for i in details:
    ...:     try:
    ...:         Color(i)
    ...:         colors_only.add(i)
    ...:     except: pass
    ...:

In [11]: colors_only
Out[11]: {'yellow', 'orange', 'blue'}