使用正则表达式对字符串进行分区

Question

Trying to break the string into 2 parts.

#Need to get 'I1234' and 'I56/I78'
name1 = 'I1234/I56/I78'

#Need to get '\I1234 ' and 'I56/I78'
name2 = '\I1234 /I56/I78'      

#Need to get '\I1234 ' and '\I56 /I78'
name3 = '\I1234 /\I56 /I78'

#Need to get '\1234 ' and '\I56 /\I78 '
name4 = '\I1234 /\I56 /\I78 '

I tried this, and it worked:

pat_a = re.compile(r'(.+)(/)(.+)')

Is there a better way ?

result = re.findall(pat_a, name2[::-1])

EDIT

There are more complicated strings possible, for example:

\I78_[0]/abcd_/efg_ /I1234/I56

Answer 1

Not sure if its better, but you can use partition or split with maxsplit=1 given to avoid the re module import:

print('I1234/I56/I78'.partition("/"))   # ('I1234', '/', 'I56/I78')

print('I1234/I56/I78'.split("/",1))     # ['I1234', 'I56/I78']

For partition you would need to look at the 0th and 2nd index of the tuple:

first, _ , last = 'I1234/I56/I78'.partition("/")

Doku:

• https://docs.python.org/3/library/stdtypes.html#str.partition
• https://docs.python.org/3/library/stdtypes.html#str.split

---

Full example:

name1 = 'I1234/I56/I78' 
name2 = '\I1234 /I56/I78'       
name3 = '\I1234 /\I56 /I78' 
name4 = '\I1234 /\I56 /\I78 '

for n in [name1,name2,name3,name4]:
    print(n.partition("/"))   # ('I1234', '/', 'I56/I78')
    print(n.split("/",1))     # ['I1234', 'I56/I78']

Output (backslashes are escaped - thats why they are doubled up):

('I1234', '/', 'I56/I78')           # using partition
['I1234', 'I56/I78']                # using split

('\\I1234 ', '/', 'I56/I78')        # partition
['\\I1234 ', 'I56/I78']             # split .. etc.

('\\I1234 ', '/', '\\I56 /I78')
['\\I1234 ', '\\I56 /I78']

('\\I1234 ', '/', '\\I56 /\\I78 ')
['\\I1234 ', '\\I56 /\\I78 ']

Answer 2

This answer uses string.split , which seems to be the cleanest method over regex. I looked at using string.partition , but it produces a tuple , which requires index splitting. Plus the output related to string.partition doesn't give the output that you requested.

This first example takes a single string and outputs a pair of strings based on your split request.

# Need to get '\I1234 ' and '\I56 /I78'
name3 = '\I1234 /\I56 /I78'

# The input name (name3) can be change in a for loop linked to your input. 
split_input = name3.split('/', 1) # maxsplit=1
print (split_input)
# outputs 
#####################################################################
# NOTE: the escaped backslashes, which doesn't match your requirement. 
#####################################################################
['\\I1234 ', '\\I56 /I78'] 

The original output above created escaped backslashes, so this code removes them.

# Need to get '\I1234 ' and '\I56 /I78'
name3 = '\I1234 /\I56 /I78'

# The input name (name3) can be change in a for loop linked to your input. 
split_input = str(name3.split('/', 1)).encode('utf-8').decode('unicode_escape')
print (split_input)
# outputs 
['\I1234 ', '\I56 /I78'] # Do you need that trailing space?

I'm not sure where your input values are originally coming from (eg, file, website, etc.), so I added the ones from your question to a list for faster testing. The next example use list comprehension and string.split.

my_strings = ['I1234/I56/I78', '\I1234 /I56/I78', '\I1234 /\I56 /I78', '\I1234 /\I56 /\I78', '\I78_[0]/abcd_/efg_ /I1234/I56']

# Uses list comprehension and string.split to split the elements in your strings
split_input = [x.split('/', 1) for x in my_strings]

# The original output created escaped backslashes, so this code removes them.
decode_output = (str(split_input).encode('utf-8').decode('unicode_escape'))

print (decode_output)
# outputs 
[['I1234', 'I56/I78'], ['\I1234 ', 'I56/I78'], ['\I1234 ', '\I56 /I78'], ['\I1234 ', '\I56 /\I78'], ['\I78_[0]', 'abcd_/efg_ /I1234/I56']]