Release 0.2 Fixes Train Parsing Problems

RailIN.egg-info/PKG-INFO

@@ -0,0 +1,12 @@
+Metadata-Version: 1.1
+Name: RailIN
+Version: 0.2
+Summary: Unofficial API for Indian Railways.
+Home-page: https://github.com/ash2shukla/RailIN
+Author: Ashish Shukla
+Author-email: ash2shukla@gmail.com
+License: MIT
+Download-URL: https://github.com/ash2shukla/RailIN/archive/0.1.tar.gz
+Description: UNKNOWN
+Keywords: Railways,API,IR,unofficial
+Platform: UNKNOWN

RailIN.egg-info/SOURCES.txt

@@ -0,0 +1,11 @@
+setup.cfg
+setup.py
+RailIN/CaptchaBreak.py
+RailIN/RailIN.py
+RailIN/__init__.py
+RailIN/prettify.py
+RailIN.egg-info/PKG-INFO
+RailIN.egg-info/SOURCES.txt
+RailIN.egg-info/dependency_links.txt
+RailIN.egg-info/requires.txt
+RailIN.egg-info/top_level.txt

RailIN.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

RailIN.egg-info/requires.txt

@@ -0,0 +1,5 @@
+bs4
+requests==2.18.1
+user_agent==0.1.8
+pillow==4.2.1
+pytesseract==0.1.7

RailIN.egg-info/top_level.txt

@@ -0,0 +1 @@
+RailIN

RailIN/CaptchaBreak.py

@@ -17,15 +17,15 @@ def _add_background(self,image):
         return image
 
     def decode(self,raw_image):
-        file = open('x.png','wb')
+        file = open('/tmp/x.png','wb')
         file.write(raw_image)
         file.close()
-        data = Image.open('x.png')
+        data = Image.open('/tmp/x.png')
         # adding background after scaling will create blurry image
         data = self._add_background(data)
         # resize the image to 300 x 128 for better recognition by tesseract
         data = data.resize((300,128),Image.ANTIALIAS)
-        remove('x.png')
+        remove('/tmp/x.png')
         # convert all pixel's alpha to 255
         string = its(data)
         try:

RailIN/RailIN.py

@@ -72,7 +72,7 @@ def getFare(self,TN,F,T):
         URL_Fare = "https://erail.in/data.aspx?Action=GetTrainFare&train="+str(TN)+"&from="+F+"&to="+T
         return Prettify().FareToJson(get(URL_Fare).text)
 
-    def getStatus(self,TN,DD,MMM,YYYY,CD):
+    def getStatus(self,TN,DD,MMM,YYYY,STN):
         D = '-'.join([str(DD), MMM, str(YYYY)])
-        URL_Live = "https://data.erail.in/getIR.aspx?&jsonp=true&Data=RUNSTATUS~0_"+str(TN)+"_"+D+"_"+CD
+        URL_Live = "https://data.erail.in/getIR.aspx?&jsonp=true&Data=RUNSTATUS~0_"+str(TN)+"_"+D+"_"+STN
         return loads(get(URL_Live).text.strip('()'))

RailIN/prettify.py

@@ -11,7 +11,6 @@ def TrainsToJson(self,string):
 
     def MakeSenseTrain(self,string):
         data = string.split('~~~~~~~~')
-
         data1 = data[0].split('~')
         retval = {}
 
@@ -32,10 +31,16 @@ def MakeSenseTrain(self,string):
         dic1['travel_time'] = data1[12]
         dic1['running_days'] = data1[13]
 
-        retval['train_base']=dic1
+        retval['train_base'] = dic1
+
+        # dealing with the second part , third part isn't interpreted yet
+        data2 = data[1].split('~~')
+        # produces atleast 6 parts at max 8
+
 
-        data2 = data[1].split('~~~')
         data2_1 = data2[0].split('~')
+        # 0th - Extract coach types etc
+
         retval['coach_types']={}
         retval['coach_types']['1A'] = data2_1[0][0]
         retval['coach_types']['2A'] = data2_1[0][1]
@@ -47,67 +52,60 @@ def MakeSenseTrain(self,string):
 
         retval['train_base']['source_depart'] = data2_1[5]
         retval['train_base']['dstn_reach'] = data2_1[6]
+        retval['train_base']['type'] = data2_1[11]
+        retval['train_base']['train_id']= data2_1[12]
+        retval['train_base']['distance_from_to'] = data2_1[18]
+        retval['train_base']['average_speed'] = data2_1[19]
 
-        retval['train_base'] = dic1
-
-        data2_arr = data2[1].split('~~')
-
-        info_count = len(data2_arr)
-
-        base_arr = data2_arr[0].split('~')
+        # 1st - Extract notification
 
-        retval['train_base']['type'] = base_arr[0]
-        retval['train_base']['train_id'] = base_arr[1]
-        retval['train_base']['distance_from_to'] = base_arr[7]
-        retval['train_base']['average_speed'] = base_arr[8]
-        if len(base_arr)==16:
-            retval['train_base']['notif_coach'] = base_arr[12]
-        else:
-            pass
-        try:
-            retval['train_base']['notif_special'] = data2_arr[1].split('~')[3]
-        except:
-            pass
-        try:
-            # if this field has more words than a train type i.e. 3-4 then it is a notif
-            if len(data_arr[2].split())>5:
-                retval['train_base']['notif_special2'] = data2_arr[2]
-            else:
-                pass
-        except:
-            pass
-        try:
-            retval['train_base']['owned_by'] = data2_arr[3].split('~')[1]
-        except:
-            pass
         try:
-            retval['train_base']['owned_by'] = data2_arr[3].split('~')[1]
-        except:
-            pass
-        try:
-            # if this part is having a alpha only string then it is owned by or region where train comes from
-            val = data2_arr[4].split('~')[1]
-            if all(i.isalpha() for i in val):
-                retval['train_base']['owned_by'] = val
-            else:
-                retval['train_base']['rake_share'] = val
-        except:
-            pass
-        try:
-            val = data2_arr[5].split('~')[1]
-            if ':' in val:
-                retval['coach_arrangement'] = []
-                for i in val.split(':'):
-                    index = str(val.index(i))
-                    i = i.split(',')
-                    retval['coach_arrangement'].append({index : {'tag':i[0],'coach_id':i[1],'type':i[2]}})
-            else:
-                pass
+            notif = data2[1].split('~')[3]
+            retval['train_base']['notif_coach'] = notif
         except:
             pass
 
+        # 2nd - Extract train type or notification
+        for i in range(2,7):
+            try:
+                if data2[i] is not None:
+                    key,value = self._extract(data2[i])
+                    retval[key]=value
+            except:
+                pass
         return retval
 
+    def _extract(self,string):
+        val = string.strip('~')
+        val = val.split('~')
+        length = len(val)
+        if length == 1:
+            if val[0] == "" :
+                return 'No Data'
+            else:
+                words = len(val[0].split())
+                if words >3:
+                    return "notif",val[0]
+                if all((i.isalpha()) or (i=='&') or (i==' ') for i in val[0]) and (words <4) and (val[0]!='BG'):
+                    return "train_type",val[0]
+        if length >= 2:
+            if all((i.isalpha()) or (i=='&') or (i==' ') for i in val[1]) and (len(val[1].split()) <4) and (val[1]!='BG') and len(val[1])>4:
+                # Check camel case
+                if all(i[0].isupper() and all(j.islower() for j in i[1:]) for i in val[1].split() if len(i)>1):
+                    return "train_type",val[1]
+            elif all(i.isalpha() and i.isupper() for i in val[1]) :
+                # it is a region
+                return "region",val[1]
+            elif all((i.isdigit())or (i==",") for i in val[1]):
+                return "rake_share",val[1].split(',')
+            elif ':' in val[1]:
+                coach = []
+                data = val[1].strip(':').split(':')
+                for i in range(len(data)):
+                    f = data[i].split(',')
+                    coach.append({str(i+1) : {'tag':f[0],'coach_id':f[1],'type':f[2]}})
+                return "coach_arrangement",coach
+
     def StationToJson(self,string):
         retval = []
         data = string.split('^')

build/lib/RailIN/CaptchaBreak.py

@@ -0,0 +1,45 @@
+from PIL import Image
+
+from os import remove
+
+from pytesseract import image_to_string as its
+
+class Captcha:
+    def _add_background(self,image):
+        data = image.convert('RGBA').getdata()
+        to_White = []
+        for i in data:
+            if i[3]==0:
+                to_White.append((255,255,255,255))
+            else:
+                to_White.append(i)
+        image.putdata(to_White)
+        return image
+
+    def decode(self,raw_image):
+        file = open('/tmp/x.png','wb')
+        file.write(raw_image)
+        file.close()
+        data = Image.open('/tmp/x.png')
+        # adding background after scaling will create blurry image
+        data = self._add_background(data)
+        # resize the image to 300 x 128 for better recognition by tesseract
+        data = data.resize((300,128),Image.ANTIALIAS)
+        remove('/tmp/x.png')
+        # convert all pixel's alpha to 255
+        string = its(data)
+        try:
+            end = string.index('=')
+        except ValueError:
+            # if = doesn't exist in string then
+            # sometimes = is recognized as semicolon
+            end = string.index(':')
+        finally:
+            # return evaluated string before = or :
+            # if neither = nor : is found then something wrong went
+            # rather than raising an error return 0
+            try:
+                return str(eval(string[:end]))
+            except NameError:
+                # if end was undefined i.e. = or : isn't defined.
+                return '0'

build/lib/RailIN/RailIN.py

@@ -0,0 +1,78 @@
+from requests import get,post,Session
+from CaptchaBreak import Captcha
+from user_agent import generate_user_agent as gua
+from json import dumps,loads
+from prettify import Prettify
+from datetime import date
+'''
+This API scrapes data of Indian Railways from erail.in and parses into JSON for use in personal applications.
+Not intended for commercial use.
+
+1. Get Route ( X )
+2. Get Availability ( X )
+3. Get Fare ( X )
+4. Get Train Status ( X )
+5. Get StationCode ( X )
+6. Get PNR ( X )
+
+'''
+
+class RailIN:
+    # Website asks for captcha which can be pre-generated due to the flaw
+    def getPNR(self,PNR):
+        if (len(str(PNR))<10) or (len(str(PNR))>10):
+            return dumps({'error':'PNR must be 10 digit.'})
+        URL_captcha = 'http://www.indianrail.gov.in/enquiry/captchaDraw.png'
+        # create a Session
+        s = Session()
+        s.headers['User-Agent'] = gua()
+        probable_captcha = Captcha().decode(s.get(URL_captcha).content)
+        try:
+            URL = 'http://www.indianrail.gov.in/enquiry/CommonCaptcha?inputCaptcha='+probable_captcha+'&inputPnrNo='+PNR+'&inputPage=PNR'
+            return loads(s.get(URL).text)
+        except:
+            return dumps({'error':'some captcha error occured'})
+
+    def getRoute(self,TN):
+        ID = self.getTrain(TN)
+        try:
+            ID['error']
+        except KeyError:
+            ID = ID['train_base']['train_id']
+        URL_Route = "https://erail.in/data.aspx?Action=TRAINROUTE&Password=2012&Data1="+ID+"&Data2=0&Cache=true"
+        return Prettify().StationToJson(get(URL_Route).text)
+
+    def getAllTrains(self,F,T):
+        URL_Trains = "https://erail.in/rail/getTrains.aspx?Station_From="+F+"&Station_To="+T+"&DataSource=0&Language=0&Cache=true"
+        return Prettify().TrainsToJson(get(URL_Trains,headers = {'User-Agent':gua()}).text)
+
+    # Pass in date month and year
+    def getTrainsOn(self,F,T,DD,MM,YYYY):
+        retval = []
+        D = date(YYYY,MM,DD).weekday()
+        for i in self.getAllTrains(F,T):
+            if i['train_base']['running_days'][D]=='1':
+                retval.append(i)
+        return dumps(retval)
+
+    def getTrain(self,TN):
+        URL_Train = "https://erail.in/rail/getTrains.aspx?TrainNo="+str(TN)+"&DataSource=0&Language=0&Cache=true"
+        try:
+            return Prettify().TrainsToJson(get(URL_Train).text)[0]
+        except:
+            return {'error':'Unexpected Server Response'}
+
+    def getAvailability(self,TN,SSTN,DSTN,CLS,QT,DD,MM):
+        URL_Avail = "https://d.erail.in/AVL_Request?Key="
+        val = '_'.join([str(TN),SSTN,DSTN,CLS,QT,str(DD)+'-'+str(MM)])
+        return Prettify().AvailToJson(get(URL_Avail+val).text)
+
+
+    def getFare(self,TN,F,T):
+        URL_Fare = "https://erail.in/data.aspx?Action=GetTrainFare&train="+str(TN)+"&from="+F+"&to="+T
+        return Prettify().FareToJson(get(URL_Fare).text)
+
+    def getStatus(self,TN,DD,MMM,YYYY,STN):
+        D = '-'.join([str(DD), MMM, str(YYYY)])
+        URL_Live = "https://data.erail.in/getIR.aspx?&jsonp=true&Data=RUNSTATUS~0_"+str(TN)+"_"+D+"_"+STN
+        return loads(get(URL_Live).text.strip('()'))

build/lib/RailIN/__init__.py

@@ -0,0 +1,3 @@
+from RailIN import RailIN
+
+__all__ = ['RailIN']