# vim: ts=2 sw=2 expandtab
import sys
from os import mkdir
from os.path import exists
import time
from urllib2 import urlopen
from xml.dom import Node
from xml.dom.minidom import parseString
from time import strptime, strftime, localtime, time, sleep
from calendar import timegm
 
debug=True
 
def getRemain():
  url='http://twitter.com/account/rate_limit_status.xml'
  raw=urlopen(url).read()
  dom=parseString(raw)
  tag=dom.getElementsByTagName('remaining-hits').item(0)
  remain=int(tag.firstChild.data)
  return remain
 
def getResetTime():
  url='http://twitter.com/account/rate_limit_status.xml'
  raw=urlopen(url).read()
  dom=parseString(raw)
  tag=dom.getElementsByTagName('reset-time').item(0)
  resetTime=tag.firstChild.data
  return resetTime
 
def getIds(dom):
  ids=[]
  for idNode in dom.getElementsByTagName('id'):
    ids.append(int(idNode.childNodes[0].data))
  return ids
 
def getPage(user, page):
  url='http://twitter.com/statuses/user_timeline/'
  rawData=urlopen(url+user+'.xml?page='+str(page)).read()
  return rawData
 
def getEpoch(str):
	dd = strptime(str, "%a %b %d %H:%M:%S +0000 %Y")
	return timegm(dd)
 
def addEpochAttr(dom):
	ds=dom.getElementsByTagName('created_at')
	for d in ds:
		dtext=d.firstChild.data
		depoch=getEpoch(dtext)
		d.setAttribute('epoch', str(depoch))
 
def debug(str):
  if debug: print 'DEBUG: '+str
 
### MAIN ####################################################
if(len(sys.argv)<2):
  print 'Usage: '+sys.argv[0]+' <twitter account name> [page offset]'
else:
  user=sys.argv[1]
  print 'twitter account name: '+user
  dir='twitter-log-'+user
  sleepTime=3
 
  # CONFIRM
  s=raw_input('Now logging start, take several times. r u OK? [y/N]: ')
 
  if s.lower()=='y' or s.lower()=='yes':
    # START LOGGING
    remain=getRemain()
    if(remain>0):
      if(not exists(dir)): mkdir(dir) # make dir
      debug('make dir: '+dir)
      idSet=set()                           # init
      if(len(sys.argv)>2): page=int(sys.argv[2])
      else:                page=1
      pageStr='%04d'%page
      rawData=getPage(user, page)           # get log
      dom=parseString(rawData)
      addEpochAttr(dom)
      debug('get log: page='+str(page))
      newIdSet=set(getIds(dom))-idSet
      idSet=idSet|newIdSet
      debug('new tweet: '+str(len(newIdSet))+', accum. tweet: '+str(len(idSet)))
      remain=getRemain()
      debug('remain hits: '+str(remain))
 
      while(len(newIdSet)>0 and remain>0):
        rawData=dom.toxml("utf-8")
        path=strftime("%Y%m%d-%H%M%S", localtime(time()))+'-page'+pageStr+'.xml'
        f=open(dir+'/'+path, 'w')
        f.write(rawData)
        debug('write file: '+path)
 
        debug('wait '+str(sleepTime)+'sec.')
        sleep(sleepTime);                   # wait
 
        page+=1
        pageStr='%04d'%page
        rawData=getPage(user, page)           # get log
        dom=parseString(rawData)
        addEpochAttr(dom)
        debug('get log: page='+str(page))
        newIdSet=set(getIds(dom))-idSet
        idSet=idSet|newIdSet
        debug('new tweet: '+str(len(newIdSet))+', accum. tweet: '+str(len(idSet)))
        remain=getRemain()
        debug('remain hits: '+str(remain))
 
      if(remain==0):
        print 'Error: no more request.'
        print 'Next reset time(UTC): '+getResetTime()
      else: print 'OK!'
 
    else:
      print 'Error: no more request.'
      print 'Next reset time(UTC): '+getResetTime()
  else: print 'Aborted.'