Extract string and number from a string which is in multiple format using regex in python?

Use groups to specify the different sections of the pattern. You can name them for easier extraction later, too:

pattern = re.compile(r"(?P<FolderName>.+)-(?P<Version>\d+)(?:-(?P<Env>dev|stage|prod))?\.tgz")

m = pattern.match(ex)
print(m.groups())
# ('hello-21234', '12345', 'prod')
print(m.group('FolderName'), m.group('Version'), m.group('Env'))
# ('hello-21234', '12345', 'prod')

ex2 = "hello-21234-1234.tgz" # No environment
m = pattern.match(ex)
print(m.groups())
# ('hello-21234', '12345', None)
print(m.group('FolderName'), m.group('Version'), m.group('Env'))
# ('hello-21234', '12345', None)

See if this pattern works

import re
exampleString = 'hello-21234-12345-prod.tgz'
build_found = re.search(r'([\w-]+)-(\d+)-(dev|stage|prod)?', exampleString)

folder_name = build_found[1]
version = build_found[2]
environment = build_found[3]

print(folder_name)
print(version)
print(environment)

Output

hello-21234
12345
prod

Surely not the best approach, but here's one idea.

Start by determining whether you have the first or second case.

-(dev|stage|prod)\.tgz$

This regex will determine whether or not you have case 1 or 2.

If it's case 1, you can extract the foldername with:

.*-

And you can extract the version with:

-\d+.tgz$

If it's case 2, you can extract the combined foldername/versionnumber with:

.*-

From there, you can extract the foldername with (again):

.*-

And the version number with:

-\d+

You need to use a regular expression that captures the components you're looking for within the string, then use .groups() to extract the captures. This worked in my testing:

re.search(r'^(.+)-(\d+)\D*$', exampleString)

example in ipython:

In [1]: import re

In [2]: s1 = 'hello-21234-12345-prod.tgz'

In [3]: s2 = 'hello-1234.tgz'

In [4]: re.search(r'^(.+)-(\d+)\D*$', s1).groups()
Out[4]: ('hello-21234', '12345')

In [5]: re.search(r'^(.+)-(\d+)\D*$', s2).groups()
Out[5]: ('hello', '1234')

The trick is the capture groups ((...)) within the regular expression r'^(.+)-(\d+)\D*$'. There are two groups - it's actually easier to decode it by looking at the second capture group first, then the first.

The second part of the regex - r'(\d+)\D*$' matches the final series of \d digits. You know it is the final series of digits, because the \D*$ part will match and swallow up all non-digit characters up to the end of the string.

The first part of the regex - r'^(.+)-' matches everything before the second part. It captures everything except the "-" character, and gives you the FolderName

Note that you'll need something a bit more complex if you have any digit characters in your environment or in the file ending (such as if you're using bzip2 compression)