Add DLP replace with info type transformation examples, and update DLP README

dlp/README.md

@@ -44,43 +44,46 @@ Note: image scanning is not currently supported on Google Cloud Storage.
 For more information, refer to the [API documentation](https://cloud.google.com/dlp/docs). 
 Optional flags are explained in [this resource](https://cloud.google.com/dlp/docs/reference/rest/v2beta1/content/inspect#InspectConfig).
 ```
-Commands:
-  -s <string>                   Inspect a string using the Data Loss Prevention API.
-  -f <filepath>                 Inspects a local text, PNG, or JPEG file using the Data Loss Prevention API.
-  -gcs -bucketName <bucketName> -fileName <fileName>  Inspects a text file stored on Google Cloud Storage using the Data Loss
-                                          Prevention API.
-  -ds -projectId [projectId] -namespace [namespace] - kind <kind> Inspect a Datastore instance using the Data Loss Prevention API.
-
-Options:
-  --help               Show help 
-  -minLikelihood       [string] [choices: "LIKELIHOOD_UNSPECIFIED", "VERY_UNLIKELY", "UNLIKELY", "POSSIBLE", "LIKELY", "VERY_LIKELY"]
-                       [default: "LIKELIHOOD_UNSPECIFIED"]
-                       specifies the minimum reporting likelihood threshold.
-  -f, --maxFindings    [number] [default: 0]
-                       maximum number of results to retrieve
-  -q, --includeQuote   [boolean] [default: true] include matching string in results
-  -t, --infoTypes      set of infoTypes to search for [eg. PHONE_NUMBER US_PASSPORT]
-  -customDictionaries  set of comma-separated dictionary words to search for as customInfoTypes
-  -customRegexes       set of regex patterns to search for as customInfoTypes
+usage: com.example.dlp.Inspect
+ -bq,--Google BigQuery         inspect BigQuery table
+ -bucketName <arg>
+ -customDictionaries <arg>
+ -customRegexes <arg>
+ -datasetId <arg>
+ -ds,--Google Datastore        inspect Datastore kind
+ -f,--file path <arg>          inspect input file path
+ -fileName <arg>
+ -gcs,--Google Cloud Storage   inspect GCS file
+ -includeQuote <arg>
+ -infoTypes <arg>
+ -kind <arg>
+ -maxFindings <arg>
+ -minLikelihood <arg>
+ -namespace <arg>
+ -projectId <arg>
+ -s,--string <arg>             inspect string
+ -subscriptionId <arg>
+ -tableId <arg>
+ -topicId <arg>
 ```
 ### Examples
  - Inspect a string:
    ```
-   java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -s "My phone number is (123) 456-7890 and my email address is [email protected]" --infoTypes PHONE_NUMBER EMAIL_ADDRESS
+   java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -s "My phone number is (123) 456-7890 and my email address is [email protected]" -infoTypes PHONE_NUMBER EMAIL_ADDRESS
    java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -s "My phone number is (123) 456-7890 and my email address is [email protected]" -customDictionaries [email protected] -customRegexes "\(\d{3}\) \d{3}-\d{4}"
    ```
  - Inspect a local file (text / image):
    ```
-     java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -f src/test/resources/test.txt --infoTypes PHONE_NUMBER EMAIL_ADDRESS
-     java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -f src/test/resources/test.png --infoTypes PHONE_NUMBER EMAIL_ADDRESS
+     java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -f src/test/resources/test.txt -infoTypes PHONE_NUMBER EMAIL_ADDRESS
+     java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -f src/test/resources/test.png -infoTypes PHONE_NUMBER EMAIL_ADDRESS
    ```
 - Inspect a file on Google Cloud Storage:
   ```
-    java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -gcs -bucketName my-bucket -fileName my-file.txt --infoTypes PHONE_NUMBER EMAIL_ADDRESS
+    java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -gcs -bucketName my-bucket -fileName my-file.txt -infoTypes PHONE_NUMBER EMAIL_ADDRESS
   ```
 - Inspect a Google Cloud Datastore kind:
   ```
-    java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -ds -kind my-kind --infoTypes PHONE_NUMBER EMAIL_ADDRESS
+    java -cp dlp/target/dlp-samples-1.0-jar-with-dependencies.jar com.example.dlp.Inspect -ds -kind my-kind -infoTypes PHONE_NUMBER EMAIL_ADDRESS
   ```
 
 ## Automatic redaction of sensitive data from images

dlp/src/main/java/com/example/dlp/DeIdentification.java

@@ -42,6 +42,7 @@
 import com.google.privacy.dlp.v2.RecordTransformations;
 import com.google.privacy.dlp.v2.ReidentifyContentRequest;
 import com.google.privacy.dlp.v2.ReidentifyContentResponse;
+import com.google.privacy.dlp.v2.ReplaceWithInfoTypeConfig;
 import com.google.privacy.dlp.v2.Table;
 import com.google.privacy.dlp.v2.Value;
 import com.google.protobuf.ByteString;
@@ -71,6 +72,71 @@
 
 public class DeIdentification {
 
+  // [START dlp_deidentify_replace_with_info_type]
+  /**
+   * Deidentify a string by replacing sensitive information with its info type using the DLP API.
+   *
+   * @param string The string to deidentify.
+   * @param projectId ID of Google Cloud project to run the API under.
+   */
+  private static void deIdentifyReplaceWithInfoType(
+      String string,
+      List<InfoType> infoTypes,
+      String projectId) {
+
+    // instantiate a client
+    try (DlpServiceClient dlpServiceClient = DlpServiceClient.create()) {
+
+      ContentItem contentItem = ContentItem.newBuilder().setValue(string).build();
+
+      // Create the deidentification transformation configuration
+      PrimitiveTransformation primitiveTransformation =
+          PrimitiveTransformation.newBuilder()
+              .setReplaceWithInfoTypeConfig(ReplaceWithInfoTypeConfig.getDefaultInstance())
+              .build();
+
+      InfoTypeTransformation infoTypeTransformationObject =
+          InfoTypeTransformation.newBuilder()
+              .setPrimitiveTransformation(primitiveTransformation)
+              .build();
+
+      InfoTypeTransformations infoTypeTransformationArray =
+          InfoTypeTransformations.newBuilder()
+              .addTransformations(infoTypeTransformationObject)
+              .build();
+
+      InspectConfig inspectConfig =
+          InspectConfig.newBuilder()
+              .addAllInfoTypes(infoTypes)
+              .build();
+
+      DeidentifyConfig deidentifyConfig =
+          DeidentifyConfig.newBuilder()
+              .setInfoTypeTransformations(infoTypeTransformationArray)
+              .build();
+
+      // Create the deidentification request object
+      DeidentifyContentRequest request =
+          DeidentifyContentRequest.newBuilder()
+              .setParent(ProjectName.of(projectId).toString())
+              .setInspectConfig(inspectConfig)
+              .setDeidentifyConfig(deidentifyConfig)
+              .setItem(contentItem)
+              .build();
+
+      // Execute the deidentification request
+      DeidentifyContentResponse response = dlpServiceClient.deidentifyContent(request);
+
+      // Print the redacted input value
+      // e.g. "My SSN is 123456789" --> "My SSN is [US_SOCIAL_SECURITY_NUMBER]"
+      String result = response.getItem().getValue();
+      System.out.println(result);
+    } catch (Exception e) {
+      System.out.println("Error in deIdentifyReplaceWithInfoType: " + e.getMessage());
+    }
+  }
+  // [END dlp_deidentify_replace_with_info_type]
+
   // [START dlp_deidentify_masking]
   /**
    * Deidentify a string by masking sensitive information with a character using the DLP API.
@@ -512,6 +578,10 @@ public static void main(String[] args) throws Exception {
     OptionGroup optionsGroup = new OptionGroup();
     optionsGroup.setRequired(true);
 
+    Option deidentifyReplaceWithInfoTypeOption =
+        new Option("it", "info_type_replace", true, "Deidentify by replacing with info type.");
+    optionsGroup.addOption(deidentifyReplaceWithInfoTypeOption);
+
     Option deidentifyMaskingOption =
         new Option("m", "mask", true, "Deidentify with character masking.");
     optionsGroup.addOption(deidentifyMaskingOption);
@@ -606,7 +676,11 @@ public static void main(String[] args) throws Exception {
       }
     }
 
-    if (cmd.hasOption("m")) {
+    if (cmd.hasOption("it")) {
+      // replace with info type
+      String val = cmd.getOptionValue(deidentifyReplaceWithInfoTypeOption.getOpt());
+      deIdentifyReplaceWithInfoType(val, infoTypesList, projectId);
+    } else if (cmd.hasOption("m")) {
       // deidentification with character masking
       int numberToMask = Integer.parseInt(cmd.getOptionValue(numberToMaskOption.getOpt(), "0"));
       char maskingCharacter = cmd.getOptionValue(maskingCharacterOption.getOpt(), "*").charAt(0);

dlp/src/test/java/com/example/dlp/DeIdentificationIT.java

@@ -57,6 +57,18 @@ public void setUp() {
     assertNotNull(System.getenv("DLP_DEID_KEY_NAME"));
   }
 
+  @Test
+  public void testDeidReplaceWithInfoType() throws Exception {
+    String text = "\"My SSN is 372819127\"";
+    DeIdentification.main(
+        new String[] {
+          "-it", text,
+          "-infoTypes", "US_SOCIAL_SECURITY_NUMBER"
+        });
+    String output = bout.toString();
+    assertThat(output, containsString("My SSN is [US_SOCIAL_SECURITY_NUMBER]"));
+  }
+
   @Test
   public void testDeidStringMasksCharacters() throws Exception {
     String text = "\"My SSN is 372819127\"";