oebb_py/workers/conn.py

from bs4 import BeautifulSoup
import requests
import datetime
import pytz
import threading
import queue
import sys

import workers.val
from classes import *

def getStation(name):
 return list(workers.val.validateName(name))[0]

def getService(sid, lines, q, eq = None):
 try:
  dep = lines[0]
  arr = lines[1]
  det = lines[2]

  depst = list(workers.val.validateName(dep.find("td", { "class": "station" }).findAll("a")[0].string))[0]
  depdate = dep.find("td", { "class": "date" }).string.strip() or None
  deptime = dep.find("td", { "class": "timeValue" }).find("span").string.split()[1].strip()

  depprog = (dep.find("span", { "class": "prognosis" }).find("span") or dep.find("span", { "class": "prognosis" })).string.strip() or None
  depplat = (dep.find("td", { "class": "platform" }).find("span") or dep.find("td", { "class": "platform" })).string.strip() or None

  walk = dep.find("img", { "class": "product" }).get("src") == "/img/vs_oebb/fuss_pic.gif"
  name = dep.find("img", { "class": "product" }).get("alt") if not walk else "Walk"

  arrst = list(workers.val.validateName(arr.find("td", { "class": "station" }).findAll("a")[0].string))[0]
  arrdate = (arr.find("td", { "class": "date" }).find("span") or arr.find("td", { "class": "date" })).string.strip() or depdate
  arrtime = arr.find("td", { "class": "timeValue" }).find("span").string.split()[1].strip()
  arrprog = (arr.find("span", { "class": "prognosis" }).find("span") or arr.find("span", { "class": "prognosis" })).string.strip() or None
  arrplat = (arr.find("td", { "class": "platform" }).find("span") or arr.find("td", { "class": "platform" })).string.strip() or None

  if arrdate and not depdate:
   arrdts = datetime.datetime.strptime(arrdate, "%d.%m.%Y")
   depdts = arrdts - datetime.timedelta(days=1)
   depdate = datetime.datetime.strftime(depdts, "%d.%m.%Y")

  dest = None

  if not (walk or depdate):
   purl = dep.find("td", { "class": "product" }).find("a").get("href")
   psource = requests.get(purl).text

   zuppa = BeautifulSoup(psource, "html5lib")
   depdate = zuppa.findAll("div", { "class": "block" })[1].text.strip()
   arrdate = depdate
   dest = list(workers.val.validateName(zuppa.findAll("div", { "class": "block" })[2].text.split(":")[1].strip()))[0]

  elif not depdate:
   depdate = "01.01.2000"
   arrdate = depdate

  depts = datetime.datetime.strptime("%s %s" % (depdate, deptime), "%d.%m.%Y %H:%M")
  arrts = datetime.datetime.strptime("%s %s" % (arrdate, arrtime), "%d.%m.%Y %H:%M")

  depprog = deptime if depprog == "pünktlich" else depprog
  arrprog = arrtime if arrprog == "pünktlich" else arrprog

  svc = Service(name, depst, depts, arrst, arrts, dest, depplat, depprog, arrplat, arrprog)
  q.put((sid, svc))

 except Exception as e:
  if eq:
   eq.put(sys.exc_info())

def getDetails(cid, url, q, via = [], eq = None):
 try:
  ssource = requests.get(url).text
  suppe = BeautifulSoup(ssource, "html5lib")

  cont = suppe.find("tr", id="trC0-%i" % cid)
  if not cont:
   return

#   buyurl = None
#
#   for url in cont.findAll("a"):
#    if url.get("href") and "https://tickets.oebb.at/de/ticket/ticket?" in url.get("href"):
#     buyurl = url.get("href")

  conn = Connection(True)

  for vst in via:
   conn.addVia(vst)

  lines = cont.findAll("tr", { "class": "tpDetails" })[1:]
  threads = []
  iq = queue.PriorityQueue()

  for line in range(0, len(lines), 3):
   t = threading.Thread(target=getService, args=(line, lines[line:line + 3], iq, eq))
   t.start()
   threads += [t]

  for t in threads:
   t.join()  

  wdate = None

  while not iq.empty():
   svc = iq.get()[1]

   if not wdate or svc.arrtime > wdate:
    wdate = svc.arrtime
   elif svc.deptime < wdate:
    ttime0 = datetime.datetime(wdate.year, wdate.month, wdate.day)
    ttime1 = ttime0 + datetime.timedelta(hours=svc.deptime.hour, minutes=svc.deptime.minute)
    ttime2 = ttime0 + datetime.timedelta(hours=svc.arrtime.hour, minutes=svc.arrtime.minute)

    if ttime1 < wdate:
     ttime1 += datetime.timedelta(days=1)
     ttime2 += datetime.timedelta(days=1)

    if ttime1 > ttime2:
     ttime2 += datetime.timedelta(days=1)

    svc.deptime = ttime1
    svc.arrtime = ttime2

   conn.addService(svc)

  q.put((cid, conn))

 except:
  if eq:
   eq.put(sys.exc_info())

def connRequest(frm, to, count = 3, time = datetime.datetime.now(), mode = False, details = False, via = []):
 outdate = datetime.datetime.strftime(time, "%d.%m.%Y")
 outtime = datetime.datetime.strftime(time, "%H:%M")

 url = "http://fahrplan.oebb.at/bin/query.exe/dn?start=1&S=%s&Z=%s&REQ0JourneyDate=%s&time=%s&REQ0HafasNumCons0=%s%s" % (frm.extid if frm.extid else frm.name, to.extid if to.extid else to.name, outdate, outtime, count, "&timesel=arrive" if mode else "")

 for i in range(len(via)):
  url += "&REQ0JourneyStops%i.0G=%s&REQ0JourneyStops%i.0A=1" % (i + 1, via[i].extid if via[i].extid else via[i].name, i + 1)

 source = requests.get(url).text

 if "GO_conViewMode=outward" not in source:
  raise ValueError("No connection found.")

 juha = BeautifulSoup(source, "html5lib")
 
 if details:
  conns = []

  for a in juha.findAll("a"):
   if a.get("href") and "GO_conViewMode" in a.get("href"):
    conns += [a.get("href")]

  threads = []
  eq = queue.Queue()
  q = queue.PriorityQueue()

  for i in range(len(conns)):
   t = threading.Thread(target=getDetails, args=(i, conns[i], q, via, eq))
   t.start()
   threads += [t]

  for t in threads:
   t.join()

  if not eq.empty():
   exc = eq.get()
   raise exc[1].with_traceback(exc[2])

  while not q.empty():
   yield q.get()[1]

 else:
  for i in range(0, count):
   det = juha.find("tr", id="trOverviewC0-%i" % i)
   if not det:
    break

   stations = det.find("td", { "class": "station" }).findAll("div")
   depst = getStation(stations[0].text.strip())
   arrst = getStation(stations[-1].text.strip())

   dates = list(det.find("td", { "class": "date" }).strings)
   depdate = dates[0]
   try:
    arrdate = dates[1]
   except:
    arrdate = depdate

   times = det.find("div", { "class": "planed" }).text
   deptime = times.split()[0]
   arrtime = times.split()[2]

   projections = det.find("div", { "class": "prognosis" })
   curdep = None
   curarr = None

   depts = datetime.datetime.strptime("%s %s" % (depdate, deptime), "%d.%m.%Y %H:%M")
   arrts = datetime.datetime.strptime("%s %s" % (arrdate, arrtime), "%d.%m.%Y %H:%M")

   name = "/".join([img.get("title") for img in det.findAll("img", { "class": "product" })])

#   ticketurl = det.find("td", { "class": "fares" }).find("a").get("href")

   svc = Service(name, depst, depts, arrst, arrts, currdep = curdep, curarr = curarr)
   conn = Connection(details)

   for vst in via:
    conn.addVia(vst)

   conn.addService(svc)

   yield conn

def worker(frm, to, count = 3, time = datetime.datetime.now(pytz.timezone("Europe/Vienna")), mode = False, details = False, json = False, via = None):
 conns = list(connRequest(getStation(frm), getStation(to), count, time, mode, details, [getStation(vst) for vst in via] if via else []))

 output = """<?xml version="1.0" encoding="UTF-8"?>
<connections>
""" if not json else """{
 \"connections\": [
 """

 for i in range(len(conns)):
  output += (conns[i].xml(1, i) + "\n") if not json else (conns[i].json(2, i) + ("\n" if i == len(conns) - 1 else ",\n"))

 output += "</connections>" if not json else " ]\n}"

 return output