001-7. Applying regular expression
<![CDATA[
@
# You can practice regular expression on site of regex crossword
# You should satisfy both left notation and upper notation
# | means "or"
# For example, to satisfy "A|Z" and "A|B", answer is "A"
# A|B
# A|Z "A"
# [ABC]
# [BDF] "B"
# [^AB]
# [ABC] "C"
# A*
# A "A"
# AB* "A"
# A?B?
# A|C "A"
# B "B"
# A+
# A|B "A"
# A|Z "A"
# Parenthesis and \ is a pair
# Using each one separately is useless
# Character behind "\" means index parenthesis
# I can use parenthesis up to 9 ones
# When I reuse parenthesis which is used before, I specify by index number
# (A)\1
# A|B "A"
# A|B "A"
# A(2,) means over 2 times (whose syntax is rarely used)
# A(1) means 1 time (whose syntax is rarely used)
# A(2,)
# A(1) "A"
# B|A "A"
# [^SPEAK]+ EP|IP|EF
# HE|LL|O+ H E
# [PLEASE]+ L P
# Day_01_05_re.py
# Module for regular expression is not involved in basic module of Python
# So we need to import "re" module to use regular expression
import re
# Search "3412 bob 123" on web to grab phone number data
# Click "LinuxFocus" link
db = '''3412 Bob 123
3834 Jonny 333
1248 Kate 634
1423 Tony 567
2567 Peter 435
3567 Alice 535
1548 Kerry 534'''
# db = '3412 Bob 123\n' \
# '3834 Jonny 333\n' \
# '1248 Kate 634'
print(db)
# We will talk about ''' ''' later
# We will talk about raw sequence of characters later
# Some examples of special character, which are used in Python:
# \n, \t, \a
# We can use following notation "re.findall(r'', db)"" as template
# The only thing we should do is to write patterns inside of ' ' by using regular expression
ns = re.findall(r'[0-9]', db)
print(ns)
# output:
# ['3', '4', '1', '2', '1', '2', '3', '3', '8', '3', '4', '3', '3', '3', '1', '2', '4', '8', '6', '3', '4', '1', '4', '2', '3', '5', '6', '7', '2', '5', '6', '7', '4', '3', '5', '3', '5', '6', '7', '5', '3', '5', '1', '5', '4', '8', '5', '3', '4']
ns = re.findall(r'[0-9]+', db)
print(ns)
# output:
# ['3412', '123', '3834', '333', '1248', '634', '1423', '567', '2567', '435', '3567', '535', '1548', '534']
# Quize
# Find only name
# Following code is what beginner generally creates
# But this one has potential bug
# See ascii code
# You can see there are some characters
# between uppercase and lowercase of alphabet
# names = re.findall(r'[A-z]+', db)
# For example, if data is
# db = '''3412 [[[Bob 123
# 3834 Jonny 333
# 1248 Kate 634
# 1423 Tony 567
# 2567 Peter 435
# 3567 Alice 535
# 1548 Kerry 534'''
# [[[ is also searched
# Recommended one is following
names = re.findall(r'[A-Za-z]+', db)
Following one means "one uppercase"+"one lowercase affected by +"
# In other words, I find name which starts with uppercase and ends with lowercase
names = re.findall(r'[A-Z][a-z]+', db)
print(names)
# Quiz
# Find all names which start with T
print(re.findall(r'T[a-z]+', db))
# Find all names which don't start with T
# Following has potential bug because it shows ony from Tony
# So, not staring with T and matched with ony, which we shouldn't find
# print(re.findall(r'[^T][a-z]+', db))
# output:
# ['Bob', 'Jonny', 'Kate', 'ony', 'Peter', 'Alice', 'Kerry']
# print(re.findall(r'[^t][a-z]+', db))
# output: show name which starts with T
# ['Bob', 'Jonny', 'Kate', 'Tony', 'Peter', 'Alice', 'Kerry']
# Following is recommended one
print(re.findall(r'[A-SU-Z][a-z]+', db))
]]>