issue #62 implemented

diff --git a/service-analysis/Dockerfile b/service-analysis/Dockerfile
index ad239ec90879ddb796656346ce9b6524ae2908eb..be70c7c4eb2136b3b72f309eb7c74176110eb7e3 100644 (file)
--- a/service-analysis/Dockerfile
+++ b/service-analysis/Dockerfile
@@ -6,6 +6,8 @@ COPY    . /app/
 
 RUN            pip install -r requirements.txt
 
+RUN            python -m textblob.download_corpora
+
 RUN            chmod a+x *.py
 
 CMD            ["python3.7", "./sentiment_analysis.py"]

diff --git a/service-analysis/README.md b/service-analysis/README.md
index 1572bebc71d14eabadb6cc36173d7a8a1f5f005c..57c6b0c795acc0564b84331d2b9beed93cfd6b5b 100644 (file)
--- a/service-analysis/README.md
+++ b/service-analysis/README.md
@@ -9,7 +9,7 @@ The service uses the the sentiment analysis service indico.
 
 `GET`: `/health` Returns HTTP Status 200 if service is running
 
-`POST`: `/` Calculates the average sentiment for a given list of texts
+`POST`: `/` Calculates the average sentiment for a given list of texts using the online sentiment analysis tool indico (https://www.indico.io).
 - param: Term[ ] as Content-Tye: `application/json`
 example body of request:
 ```json
@@ -25,6 +25,22 @@ example body of request:
        }
 ]
 ```
+`POST`: `/offline_analysis` Calculates the average sentiment for a given list of texts using TextBlob (https://textblob.readthedocs.io)
+- param:  Term[ ] as Content-Tye: `application/json`
+example body of request:
+```json
+[
+       {
+               'text': 'sample text 1',
+       },
+       {       
+               'text': 'sample text 2',
+       },
+       {       
+               'text': 'sample text 3',
+       }
+]
+```
 JSON string may contain more value-key pairs than 'text', but 'text' is needed for the sentiment analysis. Everything else will be ignored.
 For more examples see curl commands in local -> commands.
 
@@ -33,8 +49,9 @@ For more examples see curl commands in local -> commands.
 - `503`: indico sentiment analysis not available
 
 ## external services
-This sentiment analysis service uses and relies on the online sentiment analysis indico (https://www.indico.io).
-The api key can be set in the file indico_api_key.txt. The current key does work, however excessive use of this service will require you to provide your own key, since the amount of requests for this key is limited.
+This sentiment analysis service uses the online sentiment analysis indico (https://www.indico.io).
+The indico api key can be set in the file indico_api_key.txt. The current key does work, however excessive use of this service will require you to provide your own key, since the amount of requests for this key is limited.
+In case of an 503 error, you can use the `/offline_analysis`.
 
 ## run with docker
 - `docker build -t sentiment_analysis .`
@@ -46,8 +63,11 @@ The api key can be set in the file indico_api_key.txt. The current key does work
 - pip: indicoio
 - pip: flask
 - pip: flask_restful
+- pip: textblob
 ### commands
-- `python3.7 sentiment_analysis.py`
-- positive sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081 -d '[{"text":"happy birthday, i love you"}]'`
-- negative sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081 -d '[{"text":"i hate you, please die"}]'`
+- `python3.7 sentiment_analysis.py` to start the service
+- positive online sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081 -d '[{"text":"happy birthday, i love you"}]'`
+- negative online sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081 -d '[{"text":"i hate you, please die"}]'`
+- positive offline sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081/offline_analysis -d '[{"text":"happy birthday, i love you"}]'`
+- negative offline sentiment test example: `curl -v -H 'content-type: application/json' -X POST http://localhost:8081/offline_analysis -d '[{"text":"i hate you, please die"}]'`
 - health check test example: `curl -v -X GET http://localhost:8081/health`

diff --git a/service-analysis/analysis.py b/service-analysis/analysis.py
index 291b8e0591c4e8c05e04d9ff144536604b57ade2..b0c43b2bc401ed3cf60d6d40ca5e00f5de4cfac1 100644 (file)
--- a/service-analysis/analysis.py
+++ b/service-analysis/analysis.py
@@ -10,7 +10,7 @@ service_error_503 = 'The sentiment analysis service is currently unavailable.'
 
 class Analysis(Resource):
        def get(self):
-               return "POST a list of JSON objects as content-type: application/json. JSON objects must contain key 'text'.", 405, {'Allow': 'GET'}
+               return "POST a list of JSON objects as content-type: application/json. JSON objects must contain key 'text'.", 405
                
        def post(self):
                if not request.json or not isinstance(request.json, (list,)):

diff --git a/service-analysis/offline_analysis.py b/service-analysis/offline_analysis.py
new file mode 100644 (file)
index 0000000..9c9da3a
--- /dev/null
+++ b/service-analysis/offline_analysis.py
@@ -0,0 +1,26 @@
+# -*- coding: utf-8 -*-
+import json
+
+from flask import request
+from flask_restful import Resource, abort
+from textblob import TextBlob
+
+input_error_409 = 'Input must be a list of JSON objects. JSON objects must contain contain key text.'
+
+class Offline_Analysis(Resource):
+       def get(self):
+               return "POST a list of JSON objects as content-type: application/json. JSON objects must contain key 'text'.", 405
+               
+       def post(self):
+               if not request.json or not isinstance(request.json, (list,)):
+                       return abort(409, message=input_error_409)
+               texts = request.json
+               value = 0
+               for text in texts:
+                       if 'text' not in text:
+                               return abort(409, message=input_error_409)
+                       analysis = TextBlob(text['text'])
+                       value += (analysis.sentiment.polarity/2) + 0.5
+               sentiment = value/len(texts)
+               data = {'sentiment': sentiment}
+               return data

diff --git a/service-analysis/requirements.txt b/service-analysis/requirements.txt
index 206be2fc4720465c00c05d46d8ffeac8e4904f78..65e542034e17e24f35cac2a203bb52c9dd053f0f 100644 (file)
--- a/service-analysis/requirements.txt
+++ b/service-analysis/requirements.txt
@@ -1,3 +1,4 @@
 flask
 flask_restful
 indicoio
+textblob

diff --git a/service-analysis/sentiment_analysis.py b/service-analysis/sentiment_analysis.py
index e27b0af28be0f0fc487bc83b7a5c21e8d0315220..055a48de5e32b48754779d794a197e614d5be74c 100644 (file)
--- a/service-analysis/sentiment_analysis.py
+++ b/service-analysis/sentiment_analysis.py
@@ -3,6 +3,7 @@ import json
 import indicoio
 
 from analysis import Analysis
+from offline_analysis import Offline_Analysis
 from health_check import Health_Check
 
 from flask import Flask
@@ -15,6 +16,7 @@ with open('indico_api_key.txt', 'r') as key:
        indicoio.config.api_key = key.read().replace('\n','')
 
 api.add_resource(Analysis, '/')
+api.add_resource(Offline_Analysis, '/offline_analysis')
 api.add_resource(Health_Check, '/health')
 
 if __name__ == '__main__':