Parsing a string pattern - Python

Question

I have a string pattern (for a xml test reporter) in the following pattern:

'testsets.testcases.[testset].[testcase]-[date-stamp]'

For example:

a='testsets.testcases.test_different_blob_sizes.TestDifferentBlobSizes-20150430130436'

I know I always can parse the testset and testcase names by doing:

temp = a.split("-")[0]
current = temp.split(".")
testset = '.'.join(current[:-1]) + ".py"
testcase = current[-1]

However, I want to accomplish that using a more pythonic way, like regex or any other expression that I would do it in a single line. How can I accomplish that?

• Joel Hinz: possible duplicate of Python Regular Expression example
• Malik Brahimi: What are s and its name that you suddenly begin to use?
• cybertextron: @MalikBrahimi sorry will update the question
• cybertextron: @JoelHinz I dont think they are possible duplicates ... I'm looking for a more general pattern than the one asked in that question

Answer 1

You can try:

testset, testcase = re.search('(.*)\.(.*)-.*', a).group(1, 2)
testset += '.py'

re.search returns a MatchObject on matches, and it has a group method we can use to extract match groups for the regex ("()"s in the regex).

• Malik Brahimi: This an incorrect regex. Look at the OP where brackets indicated the desired groups in a certain string.

Answer 2

Just use the groups that are obtained from the regular expression searched groups:

data = re.search(r'.+\..+\.(.+)\.(.+)-(\d+)', string).groups()

Answer 3

If you strictly want to pull out the testset and testcase, i.e. "test_different_blob_sizes" and "TestDifferentBlobSizes", as in the first part of your question, you can just do:

testset, testcase = re.split('[.-]',s)[2:4]

For compact regexp-based code based on what you have, see Ziyao Wei's response.