Specifying rules for splitting strings Python

Question

I'm trying to split a sentence into three different variables for later use, and I need to specify some rules which will split it in a way I need.

EXAMPLE SENTENCE:

sentence = 'SUPER Jumper Colour BROWN-8'

From this I need three Variables

textBeforeColour = 'SUPER Jumper Colour'
Colour = 'BROWN'
Size = '8'

PS. the Colour (BROWN-8) will always be in CAPS Anything before the colour (BROWN-8) might have some words in CAPS but not all.

I've created a script that will do just that but I know that if the text changes slightly the script will break. For example

import re
text = 'SUPER Jumper Colour BROWN-8'
list = text.split()
myList = []
lastWord = list[-1]

for iterating_var in list:
   if iterating_var is not list[-1]: #THIS GIVES ME THE 'BEFORE COLOUR' TEXT
        myList.append(iterating_var)

if lastWord == 'SIZE':
     print('ONE SIZE') #This is used when the Size is not a number but comes as ONE SIZE
else:
    splitText = re.split('-',lastWord)
    print(splitText[0])
    print(splitText[1])
    Colour = splitText[0]
    size = splitText[1]

Now all of this works. But if the string will use a colour: LIGHT BLUE - this script will keep the 'LIGHT' with the sentence variable not with the colour Variable.

A small edit to the original question. I've realized that some sentences could end with a letter size e.g (SUPER Jumper Colour BROWN-S) or in some cases (SUPER Jumper Colour BROWN-S/M). - Does anyone know know maybe how to get this rule involved? @Patrick — babis95
– babis95, Commented Oct 3, 2019 at 13:52

Simon Crane · Accepted Answer · 2019-09-30 13:45:02Z

3

import re
text = "blah Blah LIGHT BLUE-8"

if text.split()[-1] == "SIZE":
    print("ONE SIZE")
else:
    colour = re.findall("([A-Z ]+)-[0-9]$", text)[0][1:]
    print(colour)
    size = int(re.findall("[0-9]+$", text)[0])
    print(size)
    sentence = re.findall("(.*[^A-Z ])[A-Z ]+-[0-9]$", text)[0]
    print(sentence)

For the colour: a sequence of zero or more capital letters and spaces, which are followed by a hyphen, zero or more digits, and the end of the string

For the size: zero or more digits at the end of the string

For the sentence: zero or more characters, then a character that is not a capital letter or space, then the pattern for the colour

answered Sep 30, 2019 at 13:45

Simon Crane

2,1822 gold badges13 silver badges22 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

babis95 Over a year ago

Its perfect!! Thank you. Simple followup question - is there a manual somewhere for the re.findall/split etc. - class where i can look up parts of these commands?

Simon Crane Over a year ago

Glad I could help :) There is a good list here: w3schools.com/python/python_regex.asp

babis95 Over a year ago

Got it, I will look more into them - THANKS AGAIN!

Patrick Haugh · Accepted Answer · 2019-09-30 13:49:48Z

3

You should be able to do this in a single regex with capturing groups:

import re

pat = re.compile(r'^([\w\s]+?)\s+([A-Z\s]+)-(\d+)$')

sentence = 'SUPER Jumper Colour LIGHT BLUE-88'

match = pat.match(sentence)
if match:
    text, color, number = match.groups()
    print(text)    # SUPER Jumper Colour
    print(color)   # LIGHT BLUE
    print(number)  # 88

Regular expressions are powerful but can get complicated. If you're unfamiliar with them, here is the documentation for the re module

answered Sep 30, 2019 at 13:49

Patrick Haugh

61.3k13 gold badges94 silver badges101 bronze badges

5 Comments

babis95 Over a year ago

Also Works like a charm - Thank you for help. I will look into documentation as it sure looks complex!

babis95 Over a year ago

Hi. A small edit to the original question. I've realized that some sentences could end with a letter size e.g (SUPER Jumper Colour BROWN-S) or in some cases (SUPER Jumper Colour BROWN-S/M). - Would you know how to incorporate these changes?

Patrick Haugh Over a year ago

You could try just changing that final capturing group (\d+) to capture all characters: (.+).

babis95 Over a year ago

Thank you it did it. Last question. I just seen that some sentences have more than one hyphen (-) e.g. 'Super-Cool Jumper BROWN-S'. The script produces no results when this occurs.

Patrick Haugh Over a year ago

You can add hyphens to the first character class to make them acceptable there too: ^([\w\s-]+?)\s+([A-Z\s]+)-(\d+)$

Collectives™ on Stack Overflow

Specifying rules for splitting strings Python

2 Answers 2

3 Comments

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related