Parse a line containing a sequence of entries, inferring data

I tried every possible pattern, but still finding difficulty. Any help please?

So i have this string : 'C+11.,18.,25.6.,2.,23.7.,27.8.23'

I want my regex to extract the days numbers ( they are situated before every month), and also that are only related to that specific month. Or days in between two months that belong only to the month after days.

Example:

Case1: days of month 6 that are before it |

Case2: days of month 7 that are after month 6. The month is defined as: *r'\.\d{1}\\.'* For example 6, 7, and 8 are months here. and any number before them is its days.

Code:

import re
captured_pattern = "C+11.,18.,25.6.,2.,23.7.,27.8.23"
pattern = r'(\d{1,2})\.(?=(\d{1})\.(?!.*,\d{1,}\.6))'
matches = re.findall(pattern, captured_pattern)
print(matches)
OUTPUT : [('25', '6'), ('23', '7'), ('27', '8')]

The days are defined as:

Example: 11.,18.,25 are days of the month 6 (juin). 2.,23 are days of month 7 (july).

The year is defined in the last part as two-digit format. 23

Thank you in advance

You can match the text portions with days up to their respective month and use

import re
s="C+11.,18.,25.6.,2.,23.7.,27.8.23"
rx = r'\b(\d{1,2}(?:\.,\d+)*)\.(\d{1,2})\b'
results = [(days.split('.,'),month,) for days, month in re.findall(rx, s)]
print(results)
# => [(['11', '18', '25'], '6'), (['2', '23'], '7'), (['27'], '8')]

See the Python demo. See the regex demo. Note that the days regex part can be defined as (0?[1-9]|[12]\d|3[01]) and the month regex part can be further precised as (0?[1-9]|1[0-2]).

Regex details:

• \b - a word boundary
• (\d{1,2}(?:\.,\d+)*) - Group 1: one or two digits, and then zero or more repetitions of ., + one or more digits
• \. - a . char
• (\d{1,2}) - Group 2 (month): one or two digits
• \b - a word boundary

import re
string = "C+11.,18.,25.6.,2.,23.7.,27.8.23"
# Remove anything that is not a digit, comma or period
cleaned = re.sub(r'[^\d,.]+', '', string)
matches = {}
days = []
for part in cleaned.split(','):
    day, month, *_ = part.split('.')
    days.append(int(day))
    if month:
        matches[int(month)] = days
        days = []

Result:

{6: [11, 18, 25], 7: [2, 23], 8: [27]}

If the parsing task is too complex for a single regex, use a custom parser and relegate regexes to simpler parts.

Regexes are good for parsing simple data structures but are not good for parsing data sequences (and plain cannot parse recursive structures like arbitrarily-deep nesting). They are also unable to do any custom processing logic like inferring parts of data in an application-specific way.

import re
from dataclasses import dataclass

@dataclass
class MutableDate:
    day = None
    month = None
    year = None

def parse(data: str):

  # split off the prefix
  data = re.match(r'[^+]+\+(.*)',data).group(1)

  # a regex for each entry
  entry_re = re.compile(r'(?P<day>\d+)\.(?:(?P<month>\d+)\.(?P<year>\d+)?)?')

  # parse the sequence into separate entries
  entries=data.split(',')

  # parse individual entries
  result=[]
  for entry in entries:
    m = entry_re.match(entry)
    result.append(MutableDate(**m.groupdict()))

  # process the parsed data with your custom logic
  # (infer missing parts of dates)
  last_month_entry = last_year_entry = None
  for i,entry in enumerate(result):
    current_month = entry.month
    if current_month != None:
      for earlier_entry in result[i-1:last_month_entry:-1]:
        earlier_entry.month = current_month
      last_month_entry = i

    current_year = entry.year
    if current_year != None:
       for earlier_entry in result[i-1:last_year_entry:-1]:
         earlier_entry.year = current_year
       last_year_entry = i

  # optionally convert all entries in the result to `datetime.date`.
  # this would have a useful side effect of validating the dates

  return result

For your data, this returns:

[MutableDate(day='11', month='6', year='23'),
 MutableDate(day='18', month='6', year='23'),
 MutableDate(day='25', month='6', year='23'),
 MutableDate(day='2', month='7', year='23'),
 MutableDate(day='23', month='7', year='23'),
 MutableDate(day='27', month='8', year='23')]

Here's my suggestion:

import re
captured_pattern = "C+11.,18.,25.6.,2.,23.7.,27.8.23"
matches = [tuple(f"{v}.{value}".split(".")) for key,value in re.findall(r"\b(\d{1,2}(?:\.,\d+)*|(?:\.,\d+)\.\d)\.(\d{1,2})\b", captured_pattern) for v in key.split(".,") if v]
print(matches)

Result:

[('11', '6'), ('18', '6'), ('25', '6'), ('2', '7'), ('23', '7'), ('27', '8', '23')]

Link to RegEx