include/parquet/level_comparison_inc.h

0001 // Licensed to the Apache Software Foundation (ASF) under one
0002 // or more contributor license agreements.  See the NOTICE file
0003 // distributed with this work for additional information
0004 // regarding copyright ownership.  The ASF licenses this file
0005 // to you under the Apache License, Version 2.0 (the
0006 // "License"); you may not use this file except in compliance
0007 // with the License.  You may obtain a copy of the License at
0008 //
0009 //   http://www.apache.org/licenses/LICENSE-2.0
0010 //
0011 // Unless required by applicable law or agreed to in writing,
0012 // software distributed under the License is distributed on an
0013 // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
0014 // KIND, either express or implied.  See the License for the
0015 // specific language governing permissions and limitations
0016 // under the License.
0017 #pragma once
0018
0019 #include "arrow/util/bit_util.h"
0020 #include "arrow/util/endian.h"
0021 #include "parquet/level_comparison.h"
0022
0023 // Used to make sure ODR rule isn't violated.
0024 #ifndef PARQUET_IMPL_NAMESPACE
0025 #  error "PARQUET_IMPL_NAMESPACE must be defined"
0026 #endif
0027 namespace parquet::internal::PARQUET_IMPL_NAMESPACE {
0028 /// Builds a bitmap by applying predicate to the level vector provided.
0029 ///
0030 /// \param[in] levels Rep or def level array.
0031 /// \param[in] num_levels The number of levels to process (must be [0, 64])
0032 /// \param[in] predicate The predicate to apply (must have the signature `bool
0033 /// predicate(int16_t)`.
0034 /// \returns The bitmap using least significant "bit" ordering.
0035 ///
0036 template <typename Predicate>
0037 inline uint64_t LevelsToBitmap(const int16_t* levels, int64_t num_levels,
0038                                Predicate predicate) {
0039   // Both clang and GCC can vectorize this automatically with SSE4/AVX2.
0040   uint64_t mask = 0;
0041   for (int x = 0; x < num_levels; x++) {
0042     mask |= static_cast<uint64_t>(predicate(levels[x]) ? 1 : 0) << x;
0043   }
0044   return ::arrow::bit_util::ToLittleEndian(mask);
0045 }
0046
0047 inline MinMax FindMinMaxImpl(const int16_t* levels, int64_t num_levels) {
0048   MinMax out{std::numeric_limits<int16_t>::max(), std::numeric_limits<int16_t>::min()};
0049   for (int x = 0; x < num_levels; x++) {
0050     out.min = std::min(levels[x], out.min);
0051     out.max = std::max(levels[x], out.max);
0052   }
0053   return out;
0054 }
0055
0056 inline uint64_t GreaterThanBitmapImpl(const int16_t* levels, int64_t num_levels,
0057                                       int16_t rhs) {
0058   return LevelsToBitmap(levels, num_levels, [rhs](int16_t value) { return value > rhs; });
0059 }
0060
0061 }  // namespace parquet::internal::PARQUET_IMPL_NAMESPACE