这一讲重要的内容从13-Specifying Tokens开始。但是一开始就出现了这么个东西：
def t_RANGLES(token)
r‘>‘
return token
前面完全没提怎么来的，看着有点迷糊，特别是r‘>‘这个，这是什么语法，于是第一次就放弃了。后来知道是在用PLY这个库，也看到文档中是这么说的：
When a function is used, the regular expression rule is specified in the function documentation string.
哦，这才想起来r‘>‘这玩意就是doc string啊，真是穿了个马甲就不认识了。。
token的定义顺序很重要。
对html中comment的处理。
其实这讲没什么好总结的，主要还是PLY。
作业：
# Hexadecimal Numbers
#
# In this exercise you will write a lexical analyzer that breaks strings up
# into whitespace-separated identifiers and numbers. An identifier is a
# sequence of one or more upper- or lower-case letters. In this exercise,
# however, there are two types of numbers: decimal numbers, and
# _hexadecimal_ numbers.
#
# Humans usually write numbers using "decimal" or "base 10" notation. The
# number# 234 means 2*10^2 + 3*10 + 4*1.
#
# It is also possible to write numbers using other "bases", like "base 16"
# or "hexadecimal". Computers often use base 16 because 16 is a convenient
# power of two (i.e., it is a closer fit to the "binary" system that
# computers use internally). A hexadecimal number always starts with the
# two-character prefix "0x" so that you know not to mistake it for a binary
# number. The number 0x234 means
# 2 * 16^2
# + 3 * 16^1
# + 4 * 16^0
# = 564 decimal.
#
# Because base 16 is larger than base 10, the letters ‘a‘ through ‘f‘ are
# used to represent the numbers ‘10‘ through ‘15‘. So the hexadecimal
# number 0xb is the same as the decimal number 11. When read out loud, the
# "0x" is often pronounced like "hex". "0x" must always be followed by at
# least one hexadecimal digit to count as a hexadecimal number.
#
# Modern programming languages like Python can understand hexadecimal
# numbers natively! Try it:
#
# print 0x234 # uncomment me to see 564 printed
# print 0xb # uncomment me to see 11 printed
#
# This provides an easy way to test your knowledge of hexadecimal.
#
# For this assignment you must write token definition rules (e.g., t_ID,
# t_NUM_hex) that will break up a string of whitespace-separated
# identifiers and numbers (either decimal or hexadecimal) into ID and NUM
# tokens. If the token is an ID, you should store its text in the
# token.value field. If the token is a NUM, you must store its numerical
# value (NOT a string) in the token.value field. This means that if a
# hexadecimal string is found, you must convert it to a decimal value.
#
# Hint 1: When presented with a hexadecimal string like "0x2b4", you can
# convert it to a decimal number in stages, reading it from left to right:
# number = 0 # ‘0x‘
# number = number * 16
# number = number + 2 # ‘2‘
# number = number * 16
# number = number + 11 # ‘b‘
# number = number * 16
# number = number + 4 # ‘4‘
# Of course, since you don‘t know the number of digits in advance, you‘ll
# probably want some sort of loop. There are other ways to convert a
# hexadecimal string to a number. You may use any way that works.
#
# Hint 2: The Python function ord() will convert a single letter into
# an ordered internal numerical representation. This allows you to perform
# simple arithmetic on numbers:
#
# print ord(‘c‘) - ord(‘a‘) == 2
import ply.lex as lex
tokens = (‘NUM‘, ‘ID‘)
####
# Fill in your code here.
####
def t_NUM_hex(token): #this should be placed before t_NUM_decimal
r‘0x[0-9a-f]+‘
token.value = int(token.value, 16)
token.type = ‘NUM‘
return token
def t_NUM_decimal(token):
r‘[0-9]+‘
token.value = int(token.value) # won‘t work on hex numbers!
token.type = ‘NUM‘
return token
def t_ID(token):
r‘[a-zA-z_]+‘
return token
t_ignore = ‘ \t\v\r‘
def t_error(t):
print "Lexer: unexpected character " + t.value[0]
t.lexer.skip(1)
# We have included some testing code to help you check your work. You will
# probably want to add your own additional tests.
lexer = lex.lex()
def test_lexer(input_string):
lexer.input(input_string)
result = [ ]
while True:
tok = lexer.token()
if not tok: break
result = result + [(tok.type, tok.value)]
return result
question1 = "0x19 equals 25" # 0x19 = (1*16) + 9
answer1 = [(‘NUM‘, 25), (‘ID‘, ‘equals‘), (‘NUM‘, 25) ]
print test_lexer(question1) == answer1
question2 = "0xfeed MY 0xface"
answer2 = [(‘NUM‘, 65261), (‘ID‘, ‘MY‘), (‘NUM‘, 64206) ]
print test_lexer(question2) == answer2
question3 = "tricky 0x0x0x"
answer3 = [(‘ID‘, ‘tricky‘), (‘NUM‘, 0), (‘ID‘, ‘x‘), (‘NUM‘, 0), (‘ID‘, ‘x‘)]
print test_lexer(question3) == answer3
question4 = "in 0xdeed"
print test_lexer(question4)
question5 = "where is the 0xbeef"
print test_lexer(question5)
Hexadecimal Numbers
# Email Addresses & Spam
#
# In this assignment you will write Python code to to extract email
# addresses from a string of text. To avoid unsolicited commercial email
# (commonly known as "spam"), users sometimes add the text NOSPAM to an
# other-wise legal email address, trusting that humans will be smart enough
# to remove it but that machines will not. As we shall see, this provides
# only relatively weak protection.
#
# For the purposes of this exercise, an email address consists of a
# word, an ‘@‘, and a domain name. A word is a non-empty sequence
# of upper- or lower-case letters. A domain name is a sequence of two or
# more words, separated by periods.
#
# Example: [email protected]
# Example: [email protected]
# Example: [email protected]
#
# If an email address has the text NOSPAM (uppercase only) anywhere in it,
# you should remove all such text. Example:
# ‘[email protected]‘ -> ‘[email protected]‘
# ‘[email protected]‘ -> ‘[email protected]‘
#
# You should write a procedure addresses() that accepts as input a string.
# Your procedure should return a list of valid email addresses found within
# that string -- each of which should have NOSPAM removed, if applicable.
#
# Hint 1: Just as we can FIND a regular expression in a string using
# re.findall(), we can also REPLACE or SUBSTITUTE a regular expression in a
# string using re.sub(regexp, new_text, haystack). Example:
#
# print re.sub(r"[0-9]+", "NUMBER", "22 + 33 = 55")
# "NUMBER + NUMBER = NUMBER"
#
# Hint 2: Don‘t forget to escape special characters.
#
# Hint 3: You don‘t have to write very much code to complete this exercise:
# you just have to put together a few concepts. It is possible to complete
# this exercise without using a lexer at all. You may use any approach that
# works.
import ply.lex as lex
import re
# Fill in your answer here.
def addresses(haystack):
emails = re.findall(r‘[a-zA-Z][email protected][a-zA-Z]+(?:\.[a-zA-Z]+)+‘, haystack)
return [re.sub(‘NOSPAM‘, ‘‘, email) for email in emails]
# We have provided a single test case for you. You will probably want to
# write your own.
input1 = """[email protected] (1814-1871) was an advocate for
democracy. [email protected] (1905-1982) wrote about
the early nazi era. [email protected] was honored with a 1994
deutsche bundespost stamp. [email protected] is not actually an email address."""
output1 = [‘[email protected]‘, ‘[email protected]‘, ‘[email protected]‘]
print addresses(input1) == output1
Email Addresses And Spam
时间： 2024-10-03 17:25:47