# Write code to split a string at uppercase letters
# using regular expression,
# and print them as string
# Example, a string "IWillDoMorePractice"
# will result in "I Will Do More Practice"

import re

text = "IWillDoMorePractice"


# sample solution #1
regex = re.compile(r'[A-Z][^A-Z]*')
string = ""
matches = regex.findall(text)    # findall() returns a list of matched string
print("matches =", matches)
for match in matches:
    if len(string) > 0:
        string += " "
    string += match      # match is a string and thus does not support group()
print(string)


# sample solution #2
regex = re.compile(r'[A-Z][^A-Z]*')
matches = regex.finditer(text)    # finditer() returns a collection of matched objects
print("matches =", matches)
string = ""
for match in matches:
    if len(string) > 0:
        string += " "

    # match is an object and thus support group(), start(), end()
    string += match.group()    # or match.group(0)
print(string)


# Consider the following code. Differences? What do they return?
# When are they used?

print("1", regex.search(text))
print("2", regex.match(text))
print("3", regex.findall(text))
print("4", regex.finditer(text))

print("5", re.findall('[A-Z][^A-Z]*', text))
print("6", re.finditer('[A-Z][^A-Z]*', text))



# let's modify the regular expression to review the concept of grouping

print("\n---- grouping concept ----")
regex = re.compile(r'([A-Z])([^A-Z]*)')     # divide into 2 groups
matches = regex.finditer(text)

for match in matches:
    print('match =', match)
    print('   match.group() = match.group(0) =', match.group())
    print('   match.groups() = match.groups(0) =', match.groups())

    print('   match.group(1) =', match.group(1))
    print('   match.group(2) =', match.group(2))
    # notice that since the regular expression is divided into 2 groups,
    # we can access group 1 and group 2
    # (if there're n groups, then we can access group 1, 2, ..., n)

    print('   match.groups(1) =', match.groups(1))
    print('   match.groups(2) =', match.groups(2))




# findall() returns the occurrences of the pattern in the target string
#    - create a list of occurrences of a pattern in a string
#    - group(), start(), end() not applicable
# search() gives more information than just the actual patterns
#    - tell where the pattern occurs in the target string
#    - returns a match object for the first occurrence
#    - returns None if pattern is not found
# match() checks if the pattern exists at the very *start* of a string
#    - returns None if pattern not found
#      note: Python processes None as False
# finditer() creates a list (or an iterator) of match objects for where
#      the pattern occurs in a string
#    - thus, can iterate a list of match objects and use group(), start(), end()