Replace assert with IAE in StringsToAutomaton#build if data is not so…

…rted (#12427)

Also update the API to accept more general Iterable<BytesRef> instead of Collection<BytesRef>

lucene/CHANGES.txt

@@ -23,6 +23,10 @@ API Changes
 
 * GITHUB#12718: Make IndexSearcher#getSlices final as it is not expected to be overridden (Luca Cavanna)
 
+* GITHUB#12427: Automata#makeStringUnion #makeBinaryStringUnion now accept Iterable<BytesRef> instead of
+  Collection<BytesRef>. They also now explicitly throw IllegalArgumentException if input data is not properly sorted
+  instead of relying on assert. (Shubham Chaudhary)
+
 New Features
 ---------------------
 

lucene/core/src/java/org/apache/lucene/util/automaton/Automata.java

@@ -43,8 +43,8 @@
  */
 public final class Automata {
   /**
-   * {@link #makeStringUnion(Collection)} limits terms of this max length to ensure the stack
-   * doesn't overflow while building, since our algorithm currently relies on recursion.
+   * {@link #makeStringUnion(Iterable)} limits terms of this max length to ensure the stack doesn't
+   * overflow while building, since our algorithm currently relies on recursion.
    */
   public static final int MAX_STRING_UNION_TERM_LENGTH = 1000;
 
@@ -576,8 +576,8 @@ public static Automaton makeString(int[] word, int offset, int length) {
    * @return An {@link Automaton} accepting all input strings. The resulting automaton is codepoint
    *     based (full unicode codepoints on transitions).
    */
-  public static Automaton makeStringUnion(Collection<BytesRef> utf8Strings) {
-    if (utf8Strings.isEmpty()) {
+  public static Automaton makeStringUnion(Iterable<BytesRef> utf8Strings) {
+    if (utf8Strings.iterator().hasNext() == false) {
       return makeEmpty();
     } else {
       return DaciukMihovAutomatonBuilder.build(utf8Strings, false);
@@ -593,8 +593,8 @@ public static Automaton makeStringUnion(Collection<BytesRef> utf8Strings) {
    * @return An {@link Automaton} accepting all input strings. The resulting automaton is binary
    *     based (UTF-8 encoded byte transition labels).
    */
-  public static Automaton makeBinaryStringUnion(Collection<BytesRef> utf8Strings) {
-    if (utf8Strings.isEmpty()) {
+  public static Automaton makeBinaryStringUnion(Iterable<BytesRef> utf8Strings) {
+    if (utf8Strings.iterator().hasNext() == false) {
       return makeEmpty();
     } else {
       return DaciukMihovAutomatonBuilder.build(utf8Strings, true);

lucene/core/src/java/org/apache/lucene/util/automaton/DaciukMihovAutomatonBuilder.java

@@ -18,7 +18,6 @@
 
 import java.io.IOException;
 import java.util.Arrays;
-import java.util.Collection;
 import java.util.HashMap;
 import java.util.IdentityHashMap;
 import org.apache.lucene.util.ArrayUtil;
@@ -31,13 +30,12 @@
  * Builds a minimal, deterministic {@link Automaton} that accepts a set of strings. The algorithm
  * requires sorted input data, but is very fast (nearly linear with the input size).
  *
- * @see #build(Collection)
- * @see Automata#makeStringUnion(Collection)
- * @see Automata#makeBinaryStringUnion(Collection)
+ * @see Automata#makeStringUnion(Iterable)
+ * @see Automata#makeBinaryStringUnion(Iterable)
  * @see Automata#makeStringUnion(BytesRefIterator)
  * @see Automata#makeBinaryStringUnion(BytesRefIterator)
  * @deprecated Visibility of this class will be reduced in a future release. Users can access this
- *     functionality directly through {@link Automata#makeStringUnion(Collection)}
+ *     functionality directly through {@link Automata#makeStringUnion(Iterable)}
  */
 @Deprecated
 public final class DaciukMihovAutomatonBuilder {
@@ -244,18 +242,18 @@ private Automaton completeAndConvert() {
    * Build a minimal, deterministic automaton from a sorted list of {@link BytesRef} representing
    * strings in UTF-8. These strings must be binary-sorted.
    *
-   * @deprecated Please see {@link Automata#makeStringUnion(Collection)} instead
+   * @deprecated Please see {@link Automata#makeStringUnion(Iterable)} instead
    */
   @Deprecated
-  public static Automaton build(Collection<BytesRef> input) {
+  public static Automaton build(Iterable<BytesRef> input) {
     return build(input, false);
   }
 
   /**
    * Build a minimal, deterministic automaton from a sorted list of {@link BytesRef} representing
    * strings in UTF-8. These strings must be binary-sorted.
    */
-  static Automaton build(Collection<BytesRef> input, boolean asBinary) {
+  static Automaton build(Iterable<BytesRef> input, boolean asBinary) {
     final DaciukMihovAutomatonBuilder builder = new DaciukMihovAutomatonBuilder();
 
     for (BytesRef b : input) {
@@ -290,9 +288,11 @@ private void add(BytesRef current, boolean asBinary) {
               + current);
     }
     assert stateRegistry != null : "Automaton already built.";
-    assert previous == null || previous.get().compareTo(current) <= 0
-        : "Input must be in sorted UTF-8 order: " + previous.get() + " >= " + current;
-    assert setPrevious(current);
+    if (previous != null && previous.get().compareTo(current) > 0) {
+      throw new IllegalArgumentException(
+          "Input must be in sorted UTF-8 order: " + previous.get() + " >= " + current);
+    }
+    setPrevious(current);
 
     // Reusable codepoint information if we're building a non-binary based automaton
     UnicodeUtil.UTF8CodePoint codePoint = null;