Merge pull request #432 from GoogleCloudPlatform/language-gcs

gguuss · web-flow · commit 7770807e017b · 2016-12-05T13:32:18.000-08:00
Add GCS examples for Natural Language
diff --git a/language/analysis/README.md b/language/analysis/README.md
@@ -30,12 +30,12 @@ mvn clean compile assembly:single
 ```
 
 We can then run the assembled JAR file with the `java` command. The variable $COMMAND takes
-three values `entities`, `sentiment` or `syntax`.
+three values `entities`, `sentiment`, or `syntax`.
 
 ```
 MAIN_CLASS=com.google.cloud.language.samples.Analyze
 JAR_FILE=target/language-entities-1.0-jar-with-dependencies.jar
-java -cp $JAR_FILE $MAIN_CLASS <sentiment|entities|syntax> <text>
+java -cp $JAR_FILE $MAIN_CLASS <sentiment|entities|syntax> <text|path>
 ```
 
 Example usage:
@@ -47,7 +47,9 @@ QUOTE="Larry Page, Google's co-founder, once described the 'perfect search
     offer products beyond search, but the spirit of what he said remains."
 
 java -cp $JAR_FILE $MAIN_CLASS entities "$QUOTE"
+java -cp $JAR_FILE $MAIN_CLASS entities "gs://bucket/file.txt"
 java -cp $JAR_FILE $MAIN_CLASS sentiment "$QUOTE"
+java -cp $JAR_FILE $MAIN_CLASS sentiment "gs://bucket/file.txt"
 java -cp $JAR_FILE $MAIN_CLASS syntax "$QUOTE"
+java -cp $JAR_FILE $MAIN_CLASS syntax "gs://bucket/file.txt"
 ```
-
diff --git a/language/analysis/src/main/java/com/google/cloud/language/samples/Analyze.java b/language/analysis/src/main/java/com/google/cloud/language/samples/Analyze.java
@@ -60,11 +60,23 @@ public static void main(String[] args) throws IOException, GeneralSecurityExcept
     Analyze app = new Analyze(LanguageServiceClient.create());
 
     if (command.equals("entities")) {
-      printEntities(System.out, app.analyzeEntities(text));
+      if (text.startsWith("gs://")) {
+        printEntities(System.out, app.analyzeEntitiesFile(text));
+      } else {
+        printEntities(System.out, app.analyzeEntitiesText(text));
+      }
     } else if (command.equals("sentiment")) {
-      printSentiment(System.out, app.analyzeSentiment(text));
+      if (text.startsWith("gs://")) {
+        printSentiment(System.out, app.analyzeSentimentFile(text));
+      } else {
+        printSentiment(System.out, app.analyzeSentimentText(text));
+      }
     } else if (command.equals("syntax")) {
-      printSyntax(System.out, app.analyzeSyntax(text));
+      if (text.startsWith("gs://")) {
+        printSyntax(System.out, app.analyzeSyntaxFile(text));
+      } else {
+        printSyntax(System.out, app.analyzeSyntaxText(text));
+      }
     }
   }
 
@@ -111,6 +123,9 @@ public static void printSentiment(PrintStream out, Sentiment sentiment) {
     out.printf("\tScore: %.3f\n", sentiment.getScore());
   }
 
+  /**
+   * Prints the Syntax for the {@code tokens}.
+   */
   public static void printSyntax(PrintStream out, List<Token> tokens) {
     if (tokens == null || tokens.size() == 0) {
       out.println("No syntax found");
@@ -153,7 +168,7 @@ public Analyze(LanguageServiceClient languageApi) {
   /**
    * Gets {@link Entity}s from the string {@code text}.
    */
-  public List<Entity> analyzeEntities(String text) throws IOException {
+  public List<Entity> analyzeEntitiesText(String text) throws IOException {
     Document doc = Document.newBuilder()
         .setContent(text).setType(Type.PLAIN_TEXT).build();
     AnalyzeEntitiesRequest request = AnalyzeEntitiesRequest.newBuilder()
@@ -163,20 +178,43 @@ public List<Entity> analyzeEntities(String text) throws IOException {
     return response.getEntitiesList();
   }
 
+  /**
+   * Gets {@link Entity}s from the contents of the object at the given GCS {@code path}.
+   */
+  public List<Entity> analyzeEntitiesFile(String path) throws IOException {
+    Document doc = Document.newBuilder()
+        .setGcsContentUri(path).setType(Type.PLAIN_TEXT).build();
+    AnalyzeEntitiesRequest request = AnalyzeEntitiesRequest.newBuilder()
+        .setDocument(doc)
+        .setEncodingType(EncodingType.UTF16).build();
+    AnalyzeEntitiesResponse response = languageApi.analyzeEntities(request);
+    return response.getEntitiesList();
+  }
+
   /**
    * Gets {@link Sentiment} from the string {@code text}.
    */
-  public Sentiment analyzeSentiment(String text) throws IOException {
+  public Sentiment analyzeSentimentText(String text) throws IOException {
     Document doc = Document.newBuilder()
         .setContent(text).setType(Type.PLAIN_TEXT).build();
     AnalyzeSentimentResponse response = languageApi.analyzeSentiment(doc);
     return response.getDocumentSentiment();
   }
 
+  /**
+   * Gets {@link Sentiment} from the contents of the object at the given GCS {@code path}.
+   */
+  public Sentiment analyzeSentimentFile(String path) throws IOException {
+    Document doc = Document.newBuilder()
+        .setGcsContentUri(path).setType(Type.PLAIN_TEXT).build();
+    AnalyzeSentimentResponse response = languageApi.analyzeSentiment(doc);
+    return response.getDocumentSentiment();
+  }
+
   /**
    * Gets {@link Token}s from the string {@code text}.
    */
-  public List<Token> analyzeSyntax(String text) throws IOException {
+  public List<Token> analyzeSyntaxText(String text) throws IOException {
     Document doc = Document.newBuilder()
         .setContent(text).setType(Type.PLAIN_TEXT).build();
     AnalyzeSyntaxRequest request = AnalyzeSyntaxRequest.newBuilder()
@@ -185,4 +223,17 @@ public List<Token> analyzeSyntax(String text) throws IOException {
     AnalyzeSyntaxResponse response = languageApi.analyzeSyntax(request);
     return response.getTokensList();
   }
+
+  /**
+   * Gets {@link Token}s from the contents of the object at the given GCS {@code path}.
+   */
+  public List<Token> analyzeSyntaxFile(String path) throws IOException {
+    Document doc = Document.newBuilder()
+        .setGcsContentUri(path).setType(Type.PLAIN_TEXT).build();
+    AnalyzeSyntaxRequest request = AnalyzeSyntaxRequest.newBuilder()
+        .setDocument(doc)
+        .setEncodingType(EncodingType.UTF16).build();
+    AnalyzeSyntaxResponse response = languageApi.analyzeSyntax(request);
+    return response.getTokensList();
+  }
 }
diff --git a/language/analysis/src/test/java/com/google/cloud/language/samples/AnalyzeIT.java b/language/analysis/src/test/java/com/google/cloud/language/samples/AnalyzeIT.java
@@ -48,7 +48,7 @@ public class AnalyzeIT {
   @Test public void analyzeEntities_withEntities_returnsLarryPage() throws Exception {
     // Act
     List<Entity> entities =
-        analyzeApp.analyzeEntities(
+        analyzeApp.analyzeEntitiesText(
             "Larry Page, Google's co-founder, once described the 'perfect search engine' as"
             + " something that 'understands exactly what you mean and gives you back exactly what"
             + " you want.' Since he spoke those words Google has grown to offer products beyond"
@@ -59,32 +59,75 @@ public class AnalyzeIT {
     assertThat(got).named("entity names").contains("Larry Page");
   }
 
-  @Test public void analyzeSentiment_returnPositive() throws Exception {
+  @Test public void analyzeEntities_withEntitiesFile_containsGod() throws Exception {
+      // Act
+      List<Entity> entities =
+          analyzeApp.analyzeEntitiesFile("gs://cloud-samples-tests/natural-language/gettysburg.txt");
+      List<String> got = entities.stream().map(e -> e.getName()).collect(Collectors.toList());
+
+      // Assert
+      assertThat(got).named("entity names").contains("God");
+    }
+
+  @Test public void analyzeSentimentText_returnPositive() throws Exception {
     // Act
     Sentiment sentiment =
-        analyzeApp.analyzeSentiment(
+        analyzeApp.analyzeSentimentText(
             "Tom Cruise is one of the finest actors in hollywood and a great star!");
 
     // Assert
     assertThat((double)sentiment.getMagnitude()).isGreaterThan(0.0);
     assertThat((double)sentiment.getScore()).isGreaterThan(0.0);
   }
 
+  @Test public void analyzeSentimentFile_returnPositiveFile() throws Exception {
+      // Act
+      Sentiment sentiment =
+          analyzeApp.analyzeSentimentFile("gs://cloud-samples-tests/natural-language/"
+                  + "sentiment/bladerunner-pos.txt");
+
+      // Assert
+      assertThat((double)sentiment.getMagnitude()).isGreaterThan(0.0);
+      assertThat((double)sentiment.getScore()).isGreaterThan(0.0);
+    }
+
   @Test public void analyzeSentiment_returnNegative() throws Exception {
     // Act
     Sentiment sentiment =
-        analyzeApp.analyzeSentiment(
+        analyzeApp.analyzeSentimentText(
             "That was the worst performance I've seen in awhile.");
 
     // Assert
     assertThat((double)sentiment.getMagnitude()).isGreaterThan(0.0);
     assertThat((double)sentiment.getScore()).isLessThan(0.0);
   }
 
+  @Test public void analyzeSentiment_returnNegativeFile() throws Exception {
+      // Act
+      Sentiment sentiment =
+          analyzeApp.analyzeSentimentFile("gs://cloud-samples-tests/natural-language/"
+                  + "sentiment/bladerunner-neg.txt");
+
+      // Assert
+      assertThat((double)sentiment.getMagnitude()).isGreaterThan(0.0);
+      assertThat((double)sentiment.getScore()).isLessThan(0.0);
+    }
+
+  @Test public void analyzeSentiment_returnNeutralFile() throws Exception {
+      // Act
+      Sentiment sentiment =
+          analyzeApp.analyzeSentimentFile("gs://cloud-samples-tests/natural-language/"
+                  + "sentiment/bladerunner-neutral.txt");
+
+      // Assert
+      assertThat((double)sentiment.getMagnitude()).isGreaterThan(1.0);
+      assertThat((double)sentiment.getScore()).isWithin(0.1);
+    }
+
   @Test public void analyzeSyntax_partOfSpeech() throws Exception {
     // Act
     List<Token> token =
-        analyzeApp.analyzeSyntax(
+        analyzeApp.analyzeSyntaxText(
             "President Obama was elected for the second term");
 
     List<Tag> got = token.stream().map(e -> e.getPartOfSpeech().getTag())
@@ -94,4 +137,20 @@ public class AnalyzeIT {
     assertThat(got).containsExactly(Tag.NOUN, Tag.NOUN, Tag.VERB,
         Tag.VERB, Tag.ADP, Tag.DET, Tag.ADJ, Tag.NOUN).inOrder();
   }
+
+  @Test public void analyzeSyntax_partOfSpeechFile() throws Exception {
+      // Act
+      List<Token> token =
+          analyzeApp.analyzeSyntaxFile("gs://cloud-samples-tests/natural-language/"
+                  + "sentiment/bladerunner-neutral.txt");
+
+      List<Tag> got = token.stream().map(e -> e.getPartOfSpeech().getTag())
+          .collect(Collectors.toList());
+
+      // Assert
+      assertThat(got).containsExactly(Tag.PRON, Tag.CONJ, Tag.VERB, Tag.CONJ, Tag.VERB,
+          Tag.DET, Tag.NOUN, Tag.PUNCT, Tag.NOUN, Tag.VERB, Tag.ADJ, Tag.PUNCT, Tag.CONJ,
+          Tag.ADV, Tag.PRON, Tag.VERB, Tag.VERB, Tag.VERB, Tag.ADJ, Tag.PUNCT, Tag.DET,
+          Tag.NOUN, Tag.VERB, Tag.ADV, Tag.ADJ,Tag.PUNCT).inOrder();
+    }
 }
