From 3894427ecd6b6682eeb2d6c542667dea00fa5e6e Mon Sep 17 00:00:00 2001
From: Liangcai Li <firestarmanllc@gmail.com>
Date: Fri, 11 Nov 2022 13:10:51 +0800
Subject: [PATCH] Add JNI for `substring` without 'end' parameter. (#12113)

Authors:
  - Liangcai Li (https://github.com/firestarman)

Approvers:
  - Robert (Bobby) Evans (https://github.com/revans2)
  - Nghia Truong (https://github.com/ttnghia)

URL: https://github.com/rapidsai/cudf/pull/12113
---
 java/src/main/java/ai/rapids/cudf/ColumnView.java | 14 +++++++++++---
 java/src/main/native/src/ColumnViewJni.cpp        | 15 +++++++++++++--
 2 files changed, 24 insertions(+), 5 deletions(-)

diff --git a/java/src/main/java/ai/rapids/cudf/ColumnView.java b/java/src/main/java/ai/rapids/cudf/ColumnView.java
index e639320b028..57849b9ba0a 100644
--- a/java/src/main/java/ai/rapids/cudf/ColumnView.java
+++ b/java/src/main/java/ai/rapids/cudf/ColumnView.java
@@ -2632,12 +2632,13 @@ public final ColumnVector stringSplitRecord(String delimiter) {
 
   /**
    * Returns a new strings column that contains substrings of the strings in the provided column.
-   * Overloading subString to support if end index is not provided. Appending -1 to indicate to
-   * read until end of string.
+   * The character positions to retrieve in each string are `[start, <the string end>)`..
+   *
    * @param start first character index to begin the substring(inclusive).
    */
   public final ColumnVector substring(int start) {
-    return substring(start, -1);
+    assert type.equals(DType.STRING) : "column type must be a String";
+    return new ColumnVector(substringS(getNativeView(), start));
   }
 
   /**
@@ -3983,6 +3984,13 @@ private static native long stringSplitRecord(long nativeHandle, String pattern,
    */
   private static native long substring(long columnView, int start, int end) throws CudfException;
 
+  /**
+   * Native method to extract substrings from a given strings column.
+   * @param columnView native handle of the cudf::column_view being operated on.
+   * @param start      first character index to begin the substrings (inclusive).
+   */
+  private static native long substringS(long columnView, int start) throws CudfException;
+
   /**
    * Native method to calculate substring from a given string column.
    * @param columnView native handle of the cudf::column_view being operated on.
diff --git a/java/src/main/native/src/ColumnViewJni.cpp b/java/src/main/native/src/ColumnViewJni.cpp
index 9d442772261..4acc14c760c 100644
--- a/java/src/main/native/src/ColumnViewJni.cpp
+++ b/java/src/main/native/src/ColumnViewJni.cpp
@@ -1397,6 +1397,18 @@ JNIEXPORT jlong JNICALL Java_ai_rapids_cudf_ColumnView_binaryOpVS(JNIEnv *env, j
   CATCH_STD(env, 0);
 }
 
+JNIEXPORT jlong JNICALL Java_ai_rapids_cudf_ColumnView_substringS(JNIEnv *env, jclass,
+                                                                  jlong cv_handle, jint start) {
+  JNI_NULL_CHECK(env, cv_handle, "column is null", 0);
+  try {
+    cudf::jni::auto_set_device(env);
+    auto const cv = reinterpret_cast<cudf::column_view const *>(cv_handle);
+    auto const scv = cudf::strings_column_view{*cv};
+    return release_as_jlong(cudf::strings::slice_strings(scv, start));
+  }
+  CATCH_STD(env, 0);
+}
+
 JNIEXPORT jlong JNICALL Java_ai_rapids_cudf_ColumnView_substring(JNIEnv *env, jclass,
                                                                  jlong column_view, jint start,
                                                                  jint end) {
@@ -1405,8 +1417,7 @@ JNIEXPORT jlong JNICALL Java_ai_rapids_cudf_ColumnView_substring(JNIEnv *env, jc
     cudf::jni::auto_set_device(env);
     cudf::column_view *cv = reinterpret_cast<cudf::column_view *>(column_view);
     cudf::strings_column_view scv(*cv);
-    return release_as_jlong((end == -1 ? cudf::strings::slice_strings(scv, start) :
-                                         cudf::strings::slice_strings(scv, start, end)));
+    return release_as_jlong(cudf::strings::slice_strings(scv, start, end));
   }
   CATCH_STD(env, 0);
 }