apacheGH-40872: [C++][Parquet] Encoding: Optimize DecodeArrow/Decode(…

…bitmap) for PlainBooleanDecoder (apache#40876)

### Rationale for this change

This is for enhance boolean decoding. I optimized the `DecodeArrow` for PlainBoolean

### What changes are included in this PR?

Optimize DecodeArrow/Decode(bitmap) for PlainBooleanDecoder, and add benchmarks

### Are these changes tested?

Yes

### Are there any user-facing changes?

Minor optimization. And `Decode` boolean will change the syntax

* GitHub Issue: apache#40872

Lead-authored-by: mwish <maplewish117@gmail.com>
Co-authored-by: Antoine Pitrou <antoine@python.org>
Signed-off-by: mwish <maplewish117@gmail.com>

cpp/src/parquet/encoding.cc

@@ -1192,12 +1192,10 @@ int PlainBooleanDecoder::DecodeArrow(
   int values_decoded = num_values - null_count;
   if (ARROW_PREDICT_FALSE(num_values_ < values_decoded)) {
     // A too large `num_values` was requested.
-    ParquetException::EofException(
-        "A too large `num_values` was requested in PlainBooleanDecoder: remain " +
-        std::to_string(num_values_) + ", requested: " + std::to_string(values_decoded));
+    ParquetException::EofException();
   }
   if (ARROW_PREDICT_FALSE(!bit_reader_->Advance(values_decoded))) {
-    ParquetException::EofException("PlainDecoder doesn't have enough values in page");
+    ParquetException::EofException();
   }
 
   if (null_count == 0) {
@@ -1210,7 +1208,7 @@ int PlainBooleanDecoder::DecodeArrow(
     BitBlockCounter bit_counter(valid_bits, valid_bits_offset, num_values);
     int64_t value_position = 0;
     int64_t valid_bits_offset_position = valid_bits_offset;
-    int64_t previous_value_offset = total_num_values_ - num_values_;
+    int64_t previous_value_offset = 0;
     while (value_position < num_values) {
       auto block = bit_counter.NextWord();
       if (block.AllSet()) {
@@ -1226,7 +1224,8 @@ int PlainBooleanDecoder::DecodeArrow(
       } else {
         for (int64_t i = 0; i < block.length; ++i) {
           if (bit_util::GetBit(valid_bits, valid_bits_offset_position + i)) {
-            bool value = bit_util::GetBit(data_, previous_value_offset);
+            bool value = bit_util::GetBit(
+                data_, total_num_values_ - num_values_ + previous_value_offset);
             builder->UnsafeAppend(value);
             previous_value_offset += 1;
           } else {

cpp/src/parquet/encoding_benchmark.cc

@@ -1518,10 +1518,11 @@ BENCHMARK_DEFINE_F(BM_DecodeArrowBooleanRle, DecodeArrowNonNull)
 (benchmark::State& state) { DecodeArrowNonNullDenseBenchmark(state); }
 BENCHMARK_REGISTER_F(BM_DecodeArrowBooleanRle, DecodeArrowNonNull)
     ->Range(MIN_RANGE, MAX_RANGE);
-BENCHMARK_DEFINE_F(BM_DecodeArrowBooleanRle, DecodeArrowWithNull)
-(benchmark::State& state) { DecodeArrowWithNullDenseBenchmark(state); }
-BENCHMARK_REGISTER_F(BM_DecodeArrowBooleanRle, DecodeArrowWithNull)
-    ->Apply(BooleanWithNullCustomArguments);
+// TODO(mwish): RleBoolean not implemented DecodeArrow with null slots yet.
+// BENCHMARK_DEFINE_F(BM_DecodeArrowBooleanRle, DecodeArrowWithNull)
+//(benchmark::State& state) { DecodeArrowWithNullDenseBenchmark(state); }
+// BENCHMARK_REGISTER_F(BM_DecodeArrowBooleanRle, DecodeArrowWithNull)
+//    ->Apply(BooleanWithNullCustomArguments);
 
 BENCHMARK_DEFINE_F(BM_DecodeArrowBooleanPlain, DecodeArrow)
 (benchmark::State& state) { DecodeArrowDenseBenchmark(state); }