ZmVtXaJs

1import httplib, urllib, base64
2import sys
3
4import websocket
5import thread
6import json
7import requests
8import urllib
9import wave
10import audioop
11from time import sleep
12import StringIO
13import struct
14import sys
15import codecs
16from xml.etree import ElementTree
17
18def imagetotext(url):
19    headers = {
20        # Request headers
21        'Content-Type': 'application/json',
22        'Ocp-Apim-Subscription-Key': 'a81d6f58ee334c79b4b6956faeca20d3',
23    }
24
25    params = urllib.urlencode({
26        # Request parameters
27        'language': 'unk',
28        'detectOrientation ': 'true',
29    })
30
31    try:
32        conn = httplib.HTTPSConnection('api.projectoxford.ai')
33        conn.request("POST", "/vision/v1.0/ocr?%s" % params, "{\"url\":\""+url+"\"}", headers)
34        response = conn.getresponse()
35        data = response.read()
36        #print(data)
37        word=""
38        x=data.find("bounding")
39        data=data[x+1:]
40        data=data[data.find('"')+1:]
41        data=data[data.find('"')+1:]
42        #print data
43        coordinates=data[:data.find('"')]
44        #print coordinates
45        while (data.find("text")!=-1):
46            data=data[data.find("text")+len("text")+3:]
47            #print data
48            x=data.find('"')
49            word=word+" "+str(data[:x]).lower()
50        return word,coordinates
51        conn.close()
52    except Exception as e:
53        print("[Errno {0}] {1}".format(e.errno, e.strerror))
54def detectLanguage(text):
55    headers = {
56        # Request headers
57        'Content-Type': 'application/json',
58        'Ocp-Apim-Subscription-Key': '299921e134524fae91de7cfe72d5d370',
59    }
60
61    params = urllib.urlencode({
62        # Request parameters
63        'numberOfLanguagesToDetect': '1',
64    })
65
66    try:
67        #text="Madame"
68        body="{\"documents\": [{\"id\": \"string\",\"text\": \""+text+"\"} ]}"
69        conn = httplib.HTTPSConnection('westus.api.cognitive.microsoft.com')
70        conn.request("POST", "/text/analytics/v2.0/languages?%s" % params, body, headers)
71        response = conn.getresponse()
72        data = response.read()
73        #print(data)
74        x=data.find("iso6391Name\"")
75        data= data[x+len("iso6391Name\""):]
76        data=data[data.find('"')+1:]
77        #print data
78        x=data[:data.find('"')]
79        conn.close()
80        return x
81    except Exception as e:
82        print("[Errno {0}] {1}".format(e.errno, e.strerror))
83
84def entityextraction(inputvalue,lang):
85    headers = {
86        # Request headers
87        'Content-Type': 'application/json',
88        'Ocp-Apim-Subscription-Key': 'afad113729f74919b1f2846a34d099da',
89    }
90
91    params = urllib.urlencode({
92    })
93
94    try:
95        conn = httplib.HTTPSConnection('api.projectoxford.ai')
96        #inputvalue="Please do not step on grass"
97        body="{\"language\" : \""+lang+"\",\"analyzerIds\" : [\"4fa79af1-f22c-408d-98bb-b7d7aeef7f04\"],\"text\" : \""+inputvalue+"\" }"
98        conn.request("POST", "/linguistics/v1.0/analyze?%s" % params, body, headers)
99        response = conn.getresponse()
100        data = response.read()
101        value=data.find("result");
102        value=data.find("result")+len("result")+4;
103        data = data[value:]
104        x=0
105        ind=[]
106        while (x!=-1):
107            x=data.find('"')
108            data=data[x+1:]
109            sub= data[:data.find('"')]
110            ind.append(sub)
111            x=data.find('"')
112            data=data[x+1:]
113        del ind[-1]
114        print ind
115        actualstring=inputvalue.split()
116        print actualstring
117        grammar=["VB","RB","NN","NNP","VBG","NNS","DT","JJ"]
118        count=0
119        result=[]
120        for i in ind:
121            for x in grammar:
122                if i==x:
123                    result.append(actualstring[count])
124            count=count+1
125        #print result
126        conn.close()
127        return result
128    except Exception as e:
129        print("[Errno {0}] {1}".format(e.errno, e.strerror))
130def genImageLink(inputValue):
131    headers = {
132        # Request headers
133        'Ocp-Apim-Subscription-Key': '14f2ca99aac74c768b3166655becdc1f',
134    }
135
136    params = urllib.urlencode({
137        # Request parameters
138        'q': inputValue,
139        'count': '10',
140        'offset': '0',
141        'mkt': 'en-us',
142        'safeSearch': 'Moderate',
143    })
144
145    try:
146        conn = httplib.HTTPSConnection('api.cognitive.microsoft.com')
147        conn.request("GET", "/bing/v5.0/images/search?%s" % params, "{body}", headers)
148        response = conn.getresponse()
149        data = response.read()
150        #print(data)
151        x=data.find("contentUrl")
152        data=data[x+1:]
153        x=data.find('"')
154        data=data[x+1:]
155        #print data[0]
156        x=data.find('"')
157        #print data[x:]
158        data=data[data.find('"')+1:]
159        #print data
160        sub=data[x-2:data.find('"')]
161        #print sub
162        conn.close()
163        return sub
164    except Exception as e:
165        print("[Errno {0}] {1}".format(e.errno, e.strerror))
166 
167#inputvalue="Please do not step on grass"   
168#inputvalue="No food or drinks"   
169#inputvalue="No running in library"   
170#inputvalue="No trespassing"   
171#inputvalue="No smoking"   
172#inputvalue="Wet floor ahead"   
173#inputvalue="Slow down kids playing"   
174#inputvalue="Do not enter workers only"  
175#url=sys.argv[1]
176def GetToken(): #Get the access token from ADM, token is good for 10 minutes
177    urlArgs = {
178        'client_id': 'imagetranslationapi',
179        'client_secret': 'WHbVzXbaQdWipvmfQ30rytDoP4TaK3PhC+ZiNWLr4WA=',
180        'scope': 'http://api.microsofttranslator.com',
181        'grant_type': 'client_credentials'
182    }
183
184    oauthUrl = 'https://datamarket.accesscontrol.windows.net/v2/OAuth2-13'
185
186    try:
187        oauthToken = json.loads(requests.post(oauthUrl, data = urllib.urlencode(urlArgs)).content) #make call to get ADM token and parse json
188        finalToken = "Bearer " + oauthToken['access_token'] #prepare the token
189    except OSError:
190        pass
191
192    return finalToken
193#End GetToken
194
195def GetTextAndTranslate(finalToken,fromLangCode,textToTranslate):
196
197    toLangCode = "en"
198    
199    #Call to Microsoft Translator Service
200    headers = {"Authorization ": finalToken}
201    translateUrl = "http://api.microsofttranslator.com/v2/Http.svc/Translate?text={}&to={}".format(textToTranslate, toLangCode)
202   
203    try:
204        translationData = requests.get(translateUrl, headers = headers) #make request
205        translation = ElementTree.fromstring(translationData.text.encode('utf-8')) # parse xml return values
206        return translation.text #display translation
207
208    except OSError:
209        pass
210
211    print " "
212 
213#End GetTextAndTranslate()
214
215url="http://il3.picdn.net/shutterstock/videos/12627623/thumb/11.jpg"
216string,coordinates=imagetotext(url)
217lang=detectLanguage(string)
218#print string
219print coordinates
220#print string
221token=GetToken()
222string=GetTextAndTranslate(token,lang,string)
223listvalues=entityextraction(string,"en")
224for i in listvalues:
225    print genImageLink(i)