Add sentiment analysis sample (#533)

* Add sentiment analysis sample * Move sample review files into resources directory * Remove blank line from end of file * Update set up instructions to point to getting started guide * Update README to remove need to set up gcloud. Itemize what setting up a project entails. * Update NL README to link to Sentiment tutorial code * Coerce number types before comparison

Author: jerjou

language/README.md

@@ -13,5 +13,9 @@ to extract text from images, then uses the NL API to extract entity information
 from those texts, and stores the extracted information in a database in support
 of further analysis and correlation.
 
+- [sentiment](sentiment) contains the [Sentiment Analysis
+ Tutorial](https://cloud.google.com/natural-language/docs/sentiment-tutorial)
+code as used within the documentation.
+
 - [syntax_triples](syntax_triples) uses syntax analysis to find
 subject-verb-object triples in a given piece of text.

language/sentiment/README.md

@@ -0,0 +1,48 @@
+# Introduction
+
+This sample contains the code referenced in the 
+[Sentiment Analysis Tutorial](http://cloud.google.com/natural-language/docs/sentiment-tutorial)
+within the Google Cloud Natural Language API Documentation. A full walkthrough of this sample
+is located within the documentation.
+
+This sample is a simple illustration of how to construct a sentiment analysis
+request and process a response using the API.
+
+## Prerequisites
+
+Set up your 
+[Cloud Natural Language API project](https://cloud.google.com/natural-language/docs/getting-started#set_up_a_project)
+, which includes:
+
+* Enabling the Natural Language API
+* Setting up a service account
+* Ensuring you've properly set up your `GOOGLE_APPLICATION_CREDENTIALS` for proper
+ authentication to the service.
+
+## Download the Code
+
+```
+$ git clone https://github.com/GoogleCloudPlatform/python-dev-samples/language/sentiment/
+$ cd python-docs-samples/language/sentiment
+```
+
+## Run the Code
+
+Open a sample folder, create a virtualenv, install dependencies, and run the sample:
+
+```
+$ virtualenv env
+$ source env/bin/activate
+(env)$ pip install -r requirements.txt
+```
+
+### Usage
+
+This sample provides four sample movie reviews which you can
+provide to the sample on the command line. (You can also
+pass your own text files.)
+
+```
+(env)$ python sentiment_analysis.py textfile.txt
+Sentiment: polarity of -0.1 with magnitude of 6.7
+```

language/sentiment/requirements.txt

@@ -0,0 +1,2 @@
+google-api-python-client==1.5.3
+

language/sentiment/resources/mixed.txt

@@ -0,0 +1,20 @@
+I really wanted to love 'Bladerunner' but ultimately I couldn't get
+myself to appreciate it fully. However, you may like it if you're into
+science fiction, especially if you're interested in the philosophical
+exploration of what it means to be human or machine. Some of the gizmos
+like the flying cars and the Vouight-Kampff machine (which seemed very
+steampunk), were quite cool.
+
+I did find the plot pretty slow and but the dialogue and action sequences
+were good. Unlike most science fiction films, this one was mostly quiet, and
+not all that much happened, except during the last 15 minutes. I didn't
+understand why a unicorn was in the movie. The visual effects were fantastic,
+however, and the musical score and overall mood was quite interesting.
+A futurist Los Angeles that was both highly polished and also falling apart
+reminded me of 'Outland.' Certainly, the style of the film made up for
+many of its pedantic plot holes.
+
+If you want your sci-fi to be lasers and spaceships, 'Bladerunner' may
+disappoint you. But if you want it to make you think, this movie may
+be worth the money.
+

language/sentiment/resources/neg.txt

@@ -0,0 +1,4 @@
+What was Hollywood thinking with this movie! I hated,
+hated, hated it. BORING! I went afterwards and demanded my money back.
+They refused.
+

language/sentiment/resources/neutral.txt

@@ -0,0 +1,3 @@
+I neither liked nor disliked this movie. Parts were interesting, but
+overall I was left wanting more. The acting was pretty good.
+

language/sentiment/resources/pos.txt

@@ -0,0 +1,11 @@
+`Bladerunner` is often touted as one of the best science fiction films ever
+made. Indeed, it satisfies many of the requisites for good sci-fi: a future
+world with flying cars and humanoid robots attempting to rebel against their
+creators. But more than anything, `Bladerunner` is a fantastic exploration
+of the nature of what it means to be human. If we create robots which can
+think, will they become human? And if they do, what makes us unique? Indeed,
+how can we be sure we're not human in any case? `Bladerunner` explored
+these issues before such movies as `The Matrix,' and did so intelligently.
+The visual effects and score by Vangelis set the mood. See this movie
+in a dark theatre to appreciate it fully. Highly recommended!
+

language/sentiment/sentiment_analysis.py

@@ -0,0 +1,54 @@
+# Copyright 2016, Google, Inc.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+'''Demonstrates how to make a simple call to the Natural Language API'''
+
+import argparse
+from googleapiclient import discovery
+from oauth2client.client import GoogleCredentials
+
+
+def main(movie_review_filename):
+ '''Run a sentiment analysis request on text within a passed filename.'''
+
+ credentials = GoogleCredentials.get_application_default()
+ service = discovery.build('language', 'v1beta1', credentials=credentials)
+
+ with open(movie_review_filename, 'r') as review_file:
+ service_request = service.documents().analyzeSentiment(
+ body={
+ 'document': {
+ 'type': 'PLAIN_TEXT',
+ 'content': review_file.read(),
+ }
+ }
+ )
+ response = service_request.execute()
+
+ polarity = response['documentSentiment']['polarity']
+ magnitude = response['documentSentiment']['magnitude']
+
+ print('Sentiment: polarity of {} with magnitude of {}'.format(
+ polarity, magnitude))
+ return 0
+
+
+if __name__ == '__main__':
+ parser = argparse.ArgumentParser(
+ description=__doc__,
+ formatter_class=argparse.RawDescriptionHelpFormatter)
+ parser.add_argument(
+ 'movie_review_filename',
+ help='The filename of the movie review you\'d like to analyze.')
+ args = parser.parse_args()
+ main(args.movie_review_filename)

language/sentiment/sentiment_analysis_test.py

@@ -0,0 +1,46 @@
+# Copyright 2016, Google, Inc.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import re
+from sentiment_analysis import main
+
+
+def test_pos(resource, capsys):
+ main(resource('pos.txt'))
+ out, err = capsys.readouterr()
+ polarity = float(re.search('polarity of (.+?) with', out).group(1))
+ magnitude = float(re.search('magnitude of (.+?)', out).group(1))
+ assert polarity * magnitude > 0
+
+
+def test_neg(resource, capsys):
+ main(resource('neg.txt'))
+ out, err = capsys.readouterr()
+ polarity = float(re.search('polarity of (.+?) with', out).group(1))
+ magnitude = float(re.search('magnitude of (.+?)', out).group(1))
+ assert polarity * magnitude < 0
+
+
+def test_mixed(resource, capsys):
+ main(resource('mixed.txt'))
+ out, err = capsys.readouterr()
+ polarity = float(re.search('polarity of (.+?) with', out).group(1))
+ assert polarity <= 0.3
+ assert polarity >= -0.3
+
+
+def test_neutral(resource, capsys):
+ main(resource('neutral.txt'))
+ out, err = capsys.readouterr()
+ magnitude = float(re.search('magnitude of (.+?)', out).group(1))
+ assert magnitude <= 2.0