#!/usr/bin/python

import re

# Notes programs will have to handle dates
# What we want to do is use the following forms:
# 
# "yesterday|today|tomorrow"
# "[+-][1-9][0-9]*"
# "(next|last)-(week|month|year)"
# "[1-9][0-9]*-days(-ago)?"
# "(0?[1-9]|[12][0-9]|30|31)(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|\
#  Sep|Oct|Nov|Dec)([0-9]*)?[0-9]{2}"
# "1st|2nd|3rd|[4-9]th|1[0-9]th|21st|22nd|23rd|2[04-9]th|30th|31st"
# "(?i)(this-)?(mon|tue|wed|thu|fri|sat|sun)"
# "(?i)(this-)?(monday|tuesday|wednesday|thursday|friday|saturday|sunday)"
# "(?i)next-(mon|tue|wed|thu|fri|sat|sun)"
# "(?i)next-(monday|tuesday|wednesday|thursday|friday|saturday|sunday)"
# 
# and normalize them into yyyy-mm-dd format
# "[0-9]{4}-[0-9]{2}-[0-9]{2}"

days = "monday|tuesday|wednesday|thursday|friday|saturday|sunday"
months = "jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec"
r_dates = [re.compile("(?i)^%s$" % p) for p in (
   "(yesterday|today|tomorrow)", 
   "[+-][1-9][0-9]*", 
   "(next|last)-(week|month|year)", 
   "[1-9][0-9]*-days(-ago)?", 
   "(0?[1-9]|[12][0-9]|30|31)(%s)([0-9]*)?[0-9]{2}" % months, 
   "1st|2nd|3rd|[4-9]th|1[0-9]th|21st|22nd|23rd|2[04-9]th|30th|31st", 
   "(this-)?(mon|tue|wed|thu|fri|sat|sun)", 
   "(this-)?(%s)" % days, 
   "next-(mon|tue|wed|thu|fri|sat|sun)", 
   "next-(%s)" % days
)]

def normalizeDate(s, base=None): 
   for i in range(len(r_dates)): 
      if i == 0: 
         # "(yesterday|today|tomorrow)"
         # "[+-][1-9][0-9]*"
         # "(next|last)-(week|month|year)"
         # "[1-9][0-9]*-days(-ago)?"
         # "(0?[1-9]|[12][0-9]|30|31)(%s)([0-9]*)?[0-9]{2}" % months
         # "1st|2nd|3rd|[4-9]th|1[0-9]th|21st|22nd|23rd|2[04-9]th|30th|31st"
         # "(this-)?(mon|tue|wed|thu|fri|sat|sun)"
         # "(this-)?(%s)" % days
         # "next-(mon|tue|wed|thu|fri|sat|sun)"
         # "next-(%s)" % days

# We want to be able to parse things simply from text input
# The following will simply return if it gets a string with a newline
# Otherwise, it'll return a tokenized tuple

r_line = re.compile(r"([A-Za-z]+) ((?:[^@]|@:@:|@(?!:))*)(?:@: +(.+))?")

def tokenize(s):
   s = s.replace('\r\n', '\n')
   s = s.replace('\r', '\n')
   if '\n' in s: return tuple(s)
   m = r_line.match(s)
   if m:
      groups = list(m.groups())
      groups[2] = groups[2] or ''
      groups[2] = tuple([s for s in groups[2].split(' ') if s.strip()])
      if groups[1].endswith(' '): groups[1] = groups[1][:-1]
      return tuple(groups)
   return tuple()

def test(): 
   assert (tokenize("note blargh @: something") == 
           ('note', 'blargh', ('something',)))
   assert (tokenize("note blargh @: something etc. ") == 
           ('note', 'blargh', ('something', 'etc.')))
   print "Tests passed."

if __name__=="__main__": 
   test()