Adapt Java bindings to const structs

native_client/java/jni/deepspeech.i

@@ -6,6 +6,8 @@
 %}
 
 %include "typemaps.i"
+%include "enums.swg"
+%javaconst(1);
 
 %include "arrays_java.i"
 // apply to DS_FeedAudioContent and DS_SpeechToText
@@ -15,12 +17,6 @@
 %pointer_functions(ModelState*, modelstatep);
 %pointer_functions(StreamingState*, streamingstatep);
 
-%typemap(newfree) char* "DS_FreeString($1);";
-
-%include "carrays.i"
-%array_functions(struct TokenMetadata, TokenMetadata_array);
-%array_functions(struct CandidateTranscript, CandidateTranscript_array);
-
 %extend struct CandidateTranscript {
   /**
    * Retrieve one TokenMetadata element
@@ -29,8 +25,8 @@
    *
    * @return The TokenMetadata requested or null
    */
-  TokenMetadata getToken(int i) {
+  const TokenMetadata& getToken(int i) {
-    return TokenMetadata_array_getitem(self->tokens, i);
+    return self->tokens[i];
   }
 }
 
@@ -42,8 +38,8 @@
    *
    * @return The CandidateTranscript requested or null
    */
-  CandidateTranscript getTranscript(int i) {
+  const CandidateTranscript& getTranscript(int i) {
-    return CandidateTranscript_array_getitem(self->transcripts, i);
+    return self->transcripts[i];
   }
 
   ~Metadata() {
@@ -58,9 +54,11 @@
 %nodefaultctor TokenMetadata;
 %nodefaultdtor TokenMetadata;
 
+%typemap(newfree) char* "DS_FreeString($1);";
 %newobject DS_SpeechToText;
 %newobject DS_IntermediateDecode;
 %newobject DS_FinishStream;
+%newobject DS_ErrorCodeToErrorMessage;
 
 %rename ("%(strip:[DS_])s") "";
 

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech/DeepSpeechModel.java

@@ -11,8 +11,15 @@ public class DeepSpeechModel {
     }
 
     // FIXME: We should have something better than those SWIGTYPE_*
-    SWIGTYPE_p_p_ModelState _mspp;
+    private SWIGTYPE_p_p_ModelState _mspp;
-    SWIGTYPE_p_ModelState   _msp;
+    private SWIGTYPE_p_ModelState   _msp;
+
+    private void evaluateErrorCode(int errorCode) {
+        DeepSpeech_Error_Codes code = DeepSpeech_Error_Codes.swigToEnum(errorCode);
+        if (code != DeepSpeech_Error_Codes.ERR_OK) {
+            throw new RuntimeException("Error: " + impl.ErrorCodeToErrorMessage(errorCode) + " (0x" + Integer.toHexString(errorCode) + ").");
+        }
+    }
 
    /**
     * @brief An object providing an interface to a trained DeepSpeech model.
@@ -20,10 +27,12 @@ public class DeepSpeechModel {
     * @constructor
     *
     * @param modelPath The path to the frozen model graph.
+    *
+    * @throws RuntimeException on failure.
     */
     public DeepSpeechModel(String modelPath) {
         this._mspp = impl.new_modelstatep();
-        impl.CreateModel(modelPath, this._mspp);
+        evaluateErrorCode(impl.CreateModel(modelPath, this._mspp));
         this._msp  = impl.modelstatep_value(this._mspp);
     }
 
@@ -43,10 +52,10 @@ public class DeepSpeechModel {
      * @param aBeamWidth The beam width used by the model. A larger beam width value
      *                   generates better results at the cost of decoding time.
      *
-     * @return Zero on success, non-zero on failure.
+     * @throws RuntimeException on failure.
      */
-    public int setBeamWidth(long beamWidth) {
+    public void setBeamWidth(long beamWidth) {
-        return impl.SetModelBeamWidth(this._msp, beamWidth);
+        evaluateErrorCode(impl.SetModelBeamWidth(this._msp, beamWidth));
     }
 
    /**
@@ -70,19 +79,19 @@ public class DeepSpeechModel {
     *
     * @param scorer The path to the external scorer file.
     *
-    * @return Zero on success, non-zero on failure (invalid arguments).
+    * @throws RuntimeException on failure.
     */
     public void enableExternalScorer(String scorer) {
-        impl.EnableExternalScorer(this._msp, scorer);
+        evaluateErrorCode(impl.EnableExternalScorer(this._msp, scorer));
     }
 
     /**
     * @brief Disable decoding using an external scorer.
     *
-    * @return Zero on success, non-zero on failure (invalid arguments).
+    * @throws RuntimeException on failure.
     */
     public void disableExternalScorer() {
-        impl.DisableExternalScorer(this._msp);
+        evaluateErrorCode(impl.DisableExternalScorer(this._msp));
     }
 
     /**
@@ -91,10 +100,10 @@ public class DeepSpeechModel {
     * @param alpha The alpha hyperparameter of the decoder. Language model weight.
     * @param beta The beta hyperparameter of the decoder. Word insertion weight.
     *
-    * @return Zero on success, non-zero on failure (invalid arguments).
+    * @throws RuntimeException on failure.
     */
     public void setScorerAlphaBeta(float alpha, float beta) {
-        impl.SetScorerAlphaBeta(this._msp, alpha, beta);
+        evaluateErrorCode(impl.SetScorerAlphaBeta(this._msp, alpha, beta));
     }
 
    /*
@@ -132,10 +141,12 @@ public class DeepSpeechModel {
     *        and finishStream().
     *
     * @return An opaque object that represents the streaming state.
+    *
+    * @throws RuntimeException on failure.
     */
     public DeepSpeechStreamingState createStream() {
         SWIGTYPE_p_p_StreamingState ssp = impl.new_streamingstatep();
-        impl.CreateStream(this._msp, ssp);
+        evaluateErrorCode(impl.CreateStream(this._msp, ssp));
         return new DeepSpeechStreamingState(impl.streamingstatep_value(ssp));
     }
 

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech_doc/CandidateTranscript.java

@@ -9,8 +9,8 @@
 package org.mozilla.deepspeech.libdeepspeech;
 
 /**
- * A single transcript computed by the model, including a confidence value and
+ * A single transcript computed by the model, including a confidence<br>
- * the metadata for its constituent tokens.
+ *        value and the metadata for its constituent tokens.
  */
 public class CandidateTranscript {
   private transient long swigCPtr;
@@ -36,14 +36,7 @@ public class CandidateTranscript {
   }
 
   /**
-   * Array of TokenMetadata objects
+   *  Array of TokenMetadata objects 
-   */
-  public void setTokens(TokenMetadata value) {
-    implJNI.CandidateTranscript_tokens_set(swigCPtr, this, TokenMetadata.getCPtr(value), value);
-  }
-
-  /**
-   * Array of TokenMetadata objects
    */
   public TokenMetadata getTokens() {
     long cPtr = implJNI.CandidateTranscript_tokens_get(swigCPtr, this);
@@ -51,31 +44,15 @@ public class CandidateTranscript {
   }
 
   /**
-   * Size of the tokens array
+   *  Size of the tokens array 
    */
-  public void setNum_tokens(int value) {
+  public long getNum_tokens() {
-    implJNI.CandidateTranscript_num_tokens_set(swigCPtr, this, value);
-  }
-
-  /**
-   * Size of the tokens array
-   */
-  public int getNum_tokens() {
     return implJNI.CandidateTranscript_num_tokens_get(swigCPtr, this);
   }
 
   /**
-   * Approximated confidence value for this transcript. This is roughly the
+   *  Approximated confidence value for this transcript. This is roughly the<br>
-   * sum of the acoustic model logit values for each timestep/character that
+   * sum of the acoustic model logit values for each timestep/character that<br>
-   * contributed to the creation of this transcript.
-   */
-  public void setConfidence(double value) {
-    implJNI.CandidateTranscript_confidence_set(swigCPtr, this, value);
-  }
-
-  /**
-   * Approximated confidence value for this transcript. This is roughly the
-   * sum of the acoustic model logit values for each timestep/character that
    * contributed to the creation of this transcript.
    */
   public double getConfidence() {
@@ -83,14 +60,14 @@ public class CandidateTranscript {
   }
 
   /**
-   * Retrieve one TokenMetadata element
+   * Retrieve one TokenMetadata element<br>
-   *
+   * <br>
-   * @param i Array index of the TokenMetadata to get
+   * @param i Array index of the TokenMetadata to get<br>
-   *
+   * <br>
    * @return The TokenMetadata requested or null
    */
   public TokenMetadata getToken(int i) {
-    return new TokenMetadata(implJNI.CandidateTranscript_getToken(swigCPtr, this, i), true);
+    return new TokenMetadata(implJNI.CandidateTranscript_getToken(swigCPtr, this, i), false);
   }
 
 }

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech_doc/DeepSpeech_Error_Codes.java

@@ -0,0 +1,65 @@
+/* ----------------------------------------------------------------------------
+ * This file was automatically generated by SWIG (http://www.swig.org).
+ * Version 4.0.1
+ *
+ * Do not make changes to this file unless you know what you are doing--modify
+ * the SWIG interface file instead.
+ * ----------------------------------------------------------------------------- */
+
+package org.mozilla.deepspeech.libdeepspeech;
+
+public enum DeepSpeech_Error_Codes {
+  ERR_OK(0x0000),
+  ERR_NO_MODEL(0x1000),
+  ERR_INVALID_ALPHABET(0x2000),
+  ERR_INVALID_SHAPE(0x2001),
+  ERR_INVALID_SCORER(0x2002),
+  ERR_MODEL_INCOMPATIBLE(0x2003),
+  ERR_SCORER_NOT_ENABLED(0x2004),
+  ERR_FAIL_INIT_MMAP(0x3000),
+  ERR_FAIL_INIT_SESS(0x3001),
+  ERR_FAIL_INTERPRETER(0x3002),
+  ERR_FAIL_RUN_SESS(0x3003),
+  ERR_FAIL_CREATE_STREAM(0x3004),
+  ERR_FAIL_READ_PROTOBUF(0x3005),
+  ERR_FAIL_CREATE_SESS(0x3006),
+  ERR_FAIL_CREATE_MODEL(0x3007);
+
+  public final int swigValue() {
+    return swigValue;
+  }
+
+  public static DeepSpeech_Error_Codes swigToEnum(int swigValue) {
+    DeepSpeech_Error_Codes[] swigValues = DeepSpeech_Error_Codes.class.getEnumConstants();
+    if (swigValue < swigValues.length && swigValue >= 0 && swigValues[swigValue].swigValue == swigValue)
+      return swigValues[swigValue];
+    for (DeepSpeech_Error_Codes swigEnum : swigValues)
+      if (swigEnum.swigValue == swigValue)
+        return swigEnum;
+    throw new IllegalArgumentException("No enum " + DeepSpeech_Error_Codes.class + " with value " + swigValue);
+  }
+
+  @SuppressWarnings("unused")
+  private DeepSpeech_Error_Codes() {
+    this.swigValue = SwigNext.next++;
+  }
+
+  @SuppressWarnings("unused")
+  private DeepSpeech_Error_Codes(int swigValue) {
+    this.swigValue = swigValue;
+    SwigNext.next = swigValue+1;
+  }
+
+  @SuppressWarnings("unused")
+  private DeepSpeech_Error_Codes(DeepSpeech_Error_Codes swigEnum) {
+    this.swigValue = swigEnum.swigValue;
+    SwigNext.next = this.swigValue+1;
+  }
+
+  private final int swigValue;
+
+  private static class SwigNext {
+    private static int next = 0;
+  }
+}
+

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech_doc/Metadata.java

@@ -40,14 +40,7 @@ public class Metadata {
   }
 
   /**
-   * Array of CandidateTranscript objects
+   *  Array of CandidateTranscript objects 
-   */
-  public void setTranscripts(CandidateTranscript value) {
-    implJNI.Metadata_transcripts_set(swigCPtr, this, CandidateTranscript.getCPtr(value), value);
-  }
-
-  /**
-   * Array of CandidateTranscript objects
    */
   public CandidateTranscript getTranscripts() {
     long cPtr = implJNI.Metadata_transcripts_get(swigCPtr, this);
@@ -55,28 +48,21 @@ public class Metadata {
   }
 
   /**
-   * Size of the transcripts array
+   *  Size of the transcripts array 
    */
-  public void setNum_transcripts(int value) {
+  public long getNum_transcripts() {
-    implJNI.Metadata_num_transcripts_set(swigCPtr, this, value);
-  }
-
-  /**
-   * Size of the transcripts array
-   */
-  public int getNum_transcripts() {
     return implJNI.Metadata_num_transcripts_get(swigCPtr, this);
   }
 
   /**
-   * Retrieve one CandidateTranscript element
+   * Retrieve one CandidateTranscript element<br>
-   *
+   * <br>
-   * @param i Array index of the CandidateTranscript to get
+   * @param i Array index of the CandidateTranscript to get<br>
-   *
+   * <br>
    * @return The CandidateTranscript requested or null
    */
   public CandidateTranscript getTranscript(int i) {
-    return new CandidateTranscript(implJNI.Metadata_getTranscript(swigCPtr, this, i), true);
+    return new CandidateTranscript(implJNI.Metadata_getTranscript(swigCPtr, this, i), false);
   }
 
 }

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech_doc/README.rst

@@ -4,7 +4,7 @@ Javadoc for Sphinx
 
 This code is only here for reference for documentation generation.
 
-To update, please build SWIG (4.0 at least) and then run from native_client/java:
+To update, please install SWIG (4.0 at least) and then run from native_client/java:
 
 .. code-block::
 

native_client/java/libdeepspeech/src/main/java/org/mozilla/deepspeech/libdeepspeech_doc/TokenMetadata.java

@@ -35,42 +35,21 @@ public class TokenMetadata {
   }
 
   /**
-   * The text corresponding to this token
+   *  The text corresponding to this token 
-   */
-  public void setText(String value) {
-    implJNI.TokenMetadata_text_set(swigCPtr, this, value);
-  }
-
-  /**
-   * The text corresponding to this token
    */
   public String getText() {
     return implJNI.TokenMetadata_text_get(swigCPtr, this);
   }
 
   /**
-   * Position of the token in units of 20ms
+   *  Position of the token in units of 20ms 
    */
-  public void setTimestep(int value) {
+  public long getTimestep() {
-    implJNI.TokenMetadata_timestep_set(swigCPtr, this, value);
-  }
-
-  /**
-   * Position of the token in units of 20ms
-   */
-  public int getTimestep() {
     return implJNI.TokenMetadata_timestep_get(swigCPtr, this);
   }
 
   /**
-   * Position of the token in seconds
+   *  Position of the token in seconds 
-   */
-  public void setStart_time(float value) {
-    implJNI.TokenMetadata_start_time_set(swigCPtr, this, value);
-  }
-
-  /**
-   * Position of the token in seconds
    */
   public float getStart_time() {
     return implJNI.TokenMetadata_start_time_get(swigCPtr, this);