045-004. replacing string, sub(), raw string in regualr expression
<![CDATA[
@
# Find patter(apple or orange), replace it with fruit in target string 'apple box orange tree'
re.sub('apple|orange', 'fruit', 'apple box orange tree')
# 'fruit box fruit tree'
# Find [0-9] showing over 1 times, replace it with n in target string '1 2 Fizz 4 Buzz Fizz 7 8'
re.sub('[0-9]+', 'n', '1 2 Fizz 4 Buzz Fizz 7 8')
# 'n n Fizz n Buzz Fizz n n'
@
# You can pass method which returns target replacement into sub()
# Method which returns target replacement takes match object m
def multiple10(m):
# Bring matched string by m.group() and convert that string into int
n = int(m.group())
# This method returns final target replacement
return str(n * 10)
# Find [0-9] showing over 1 times, target replacement is what multiple10() returns, target string '1 2 Fizz 4 Buzz Fizz 7 8'
re.sub('[0-9]+', multiple10, '1 2 Fizz 4 Buzz Fizz 7 8')
# '10 20 Fizz 40 Buzz Fizz 70 80'
@
# You can use lambda expression to make method which returns target replacement
re.sub('[0-9]+', lambda m: str(int(m.group()) * 10), '1 2 Fizz 4 Buzz Fizz 7 8')
# '10 20 Fizz 40 Buzz Fizz 70 80'
@
# group1 finds hello
# group2 finds 1234
# '([a-z]+) ([0-9]+)' finds hello 1234
# I replace 'hello 1234' with '\\2 \\1 \\2 \\'
# I replace 'hello 1234' with '1234 hello 1234 hello'
re.sub('([a-z]+) ([0-9]+)', '\\2 \\1 \\2 \\1', 'hello 1234')
# '1234 hello 1234 hello'
@
# ({\s*) group1 finds '{ '
# (\w+) group2 finds 'name'
# (\w+) group3 finds 'james'
# (\s*}) group4 finds ' }'
# '({\s*)"(\w+)":\s*"(\w+)"(\s*})' finds '{ "name": "james" }'
# I replace '{ "name": "james" }' with '<\\2>\\3</\\2>'
# I replace '{ "name": "james" }' with '<name>jamaes</name>'
re.sub('({\s*)"(\w+)":\s*"(\w+)"(\s*})', '<\\2>\\3</\\2>', '{ "name": "james" }')
# '<name>james</name>'
@
# ({\s*) group1 finds '{ '
# (?P<key>\w+) group2(name key) finds 'name'
# (?P<value>\w+) group3(name value) finds 'james'
# (\s*}) group4 finds ' }'
# '({\s*)"(?P<key>\w+)":\s*"(?P<value>\w+)"(\s*})' finds '{ "name": "james" }'
# I replace '{ "name": "james" }' with '<\\g<key>>\\g<value></\\g<key>>'
# I replace '{ "name": "james" }' with '<name>jamaes</name>'
re.sub('({\s*)"(?P<key>\w+)":\s*"(?P<value>\w+)"(\s*})', '<\\g<key>>\\g<value></\\g<key>>', '{ "name": "james" }')
# '<name>james</name>'
@
# Tip
# When you want to find special characters in regular expression pattern, you use \
# But in other way, you can use 'r' to make regular expression pattern as raw string
# In that raw string, you can find special character without \
# Therefore, in raw string, you can use \number, \g<name>, \g<number> instead of \\number, \\g<name>, \\g<number>
# ({\s*) group1 finds '{ '
# (\w+) group2 finds 'name'
# (\w+) group3 finds 'james'
# (\s*}) group4 finds ' }'
# '({\s*)"(\w+)":\s*"(\w+)"(\s*})' finds '{ "name": "james" }'
# I replace '{ "name": "james" }' with r'<\2>\3</\2>'
# I replace '{ "name": "james" }' with '<name>jamaes</name>'
re.sub('({\s*)"(\w+)":\s*"(\w+)"(\s*})', r'<\2>\3</\2>', '{ "name": "james" }')
# '<name>james</name>'
]]>