|
| 1 | +// Licensed to the Apache Software Foundation (ASF) under one |
| 2 | +// or more contributor license agreements. See the NOTICE file |
| 3 | +// distributed with this work for additional information |
| 4 | +// regarding copyright ownership. The ASF licenses this file |
| 5 | +// to you under the Apache License, Version 2.0 (the |
| 6 | +// "License"); you may not use this file except in compliance |
| 7 | +// with the License. You may obtain a copy of the License at |
| 8 | +// |
| 9 | +// http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +// |
| 11 | +// Unless required by applicable law or agreed to in writing, |
| 12 | +// software distributed under the License is distributed on an |
| 13 | +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| 14 | +// KIND, either express or implied. See the License for the |
| 15 | +// specific language governing permissions and limitations |
| 16 | +// under the License. |
| 17 | + |
| 18 | +#include "vec/exec/format/parquet/delta_bit_pack_decoder.h" |
| 19 | + |
| 20 | +namespace doris::vectorized { |
| 21 | +Status DeltaLengthByteArrayDecoder::_decode_lengths() { |
| 22 | + RETURN_IF_ERROR(_len_decoder.set_bit_reader(_bit_reader)); |
| 23 | + // get the number of encoded lengths |
| 24 | + int num_length = _len_decoder.valid_values_count(); |
| 25 | + _buffered_length.resize(num_length); |
| 26 | + |
| 27 | + // decode all the lengths. all the lengths are buffered in buffered_length_. |
| 28 | + int ret; |
| 29 | + RETURN_IF_ERROR(_len_decoder.decode(_buffered_length.data(), num_length, &ret)); |
| 30 | + DCHECK_EQ(ret, num_length); |
| 31 | + _length_idx = 0; |
| 32 | + _num_valid_values = num_length; |
| 33 | + return Status::OK(); |
| 34 | +} |
| 35 | + |
| 36 | +Status DeltaLengthByteArrayDecoder::_get_internal(Slice* buffer, int max_values, |
| 37 | + int* out_num_values) { |
| 38 | + // Decode up to `max_values` strings into an internal buffer |
| 39 | + // and reference them into `buffer`. |
| 40 | + max_values = std::min(max_values, _num_valid_values); |
| 41 | + if (max_values == 0) { |
| 42 | + *out_num_values = 0; |
| 43 | + return Status::OK(); |
| 44 | + } |
| 45 | + |
| 46 | + int32_t data_size = 0; |
| 47 | + const int32_t* length_ptr = _buffered_length.data() + _length_idx; |
| 48 | + for (int i = 0; i < max_values; ++i) { |
| 49 | + int32_t len = length_ptr[i]; |
| 50 | + if (PREDICT_FALSE(len < 0)) { |
| 51 | + return Status::InvalidArgument("Negative string delta length"); |
| 52 | + } |
| 53 | + buffer[i].size = len; |
| 54 | + if (common::add_overflow(data_size, len, data_size)) { |
| 55 | + return Status::InvalidArgument("Excess expansion in DELTA_(LENGTH_)BYTE_ARRAY"); |
| 56 | + } |
| 57 | + } |
| 58 | + _length_idx += max_values; |
| 59 | + |
| 60 | + _buffered_data.resize(data_size); |
| 61 | + char* data_ptr = _buffered_data.data(); |
| 62 | + for (int j = 0; j < data_size; j++) { |
| 63 | + if (!_bit_reader->GetValue(8, data_ptr + j)) { |
| 64 | + return Status::IOError("Get length bytes EOF"); |
| 65 | + } |
| 66 | + } |
| 67 | + |
| 68 | + for (int i = 0; i < max_values; ++i) { |
| 69 | + buffer[i].data = data_ptr; |
| 70 | + data_ptr += buffer[i].size; |
| 71 | + } |
| 72 | + // this->num_values_ -= max_values; |
| 73 | + _num_valid_values -= max_values; |
| 74 | + *out_num_values = max_values; |
| 75 | + return Status::OK(); |
| 76 | +} |
| 77 | + |
| 78 | +Status DeltaByteArrayDecoder::_get_internal(Slice* buffer, int max_values, int* out_num_values) { |
| 79 | + // Decode up to `max_values` strings into an internal buffer |
| 80 | + // and reference them into `buffer`. |
| 81 | + max_values = std::min(max_values, _num_valid_values); |
| 82 | + if (max_values == 0) { |
| 83 | + *out_num_values = max_values; |
| 84 | + return Status::OK(); |
| 85 | + } |
| 86 | + |
| 87 | + int suffix_read; |
| 88 | + RETURN_IF_ERROR(_suffix_decoder.decode(buffer, max_values, &suffix_read)); |
| 89 | + if (PREDICT_FALSE(suffix_read != max_values)) { |
| 90 | + return Status::IOError("Read {}, expecting {} from suffix decoder", |
| 91 | + std::to_string(suffix_read), std::to_string(max_values)); |
| 92 | + } |
| 93 | + |
| 94 | + int64_t data_size = 0; |
| 95 | + const int32_t* prefix_len_ptr = _buffered_prefix_length.data() + _prefix_len_offset; |
| 96 | + for (int i = 0; i < max_values; ++i) { |
| 97 | + if (PREDICT_FALSE(prefix_len_ptr[i] < 0)) { |
| 98 | + return Status::InvalidArgument("negative prefix length in DELTA_BYTE_ARRAY"); |
| 99 | + } |
| 100 | + if (PREDICT_FALSE(common::add_overflow(data_size, static_cast<int64_t>(prefix_len_ptr[i]), |
| 101 | + data_size) || |
| 102 | + common::add_overflow(data_size, static_cast<int64_t>(buffer[i].size), |
| 103 | + data_size))) { |
| 104 | + return Status::InvalidArgument("excess expansion in DELTA_BYTE_ARRAY"); |
| 105 | + } |
| 106 | + } |
| 107 | + _buffered_data.resize(data_size); |
| 108 | + |
| 109 | + std::string_view prefix {_last_value}; |
| 110 | + |
| 111 | + char* data_ptr = _buffered_data.data(); |
| 112 | + for (int i = 0; i < max_values; ++i) { |
| 113 | + if (PREDICT_FALSE(static_cast<size_t>(prefix_len_ptr[i]) > prefix.length())) { |
| 114 | + return Status::InvalidArgument("prefix length too large in DELTA_BYTE_ARRAY"); |
| 115 | + } |
| 116 | + memcpy(data_ptr, prefix.data(), prefix_len_ptr[i]); |
| 117 | + // buffer[i] currently points to the string suffix |
| 118 | + memcpy(data_ptr + prefix_len_ptr[i], buffer[i].data, buffer[i].size); |
| 119 | + buffer[i].data = data_ptr; |
| 120 | + buffer[i].size += prefix_len_ptr[i]; |
| 121 | + data_ptr += buffer[i].size; |
| 122 | + prefix = std::string_view {buffer[i].data, buffer[i].size}; |
| 123 | + } |
| 124 | + _prefix_len_offset += max_values; |
| 125 | + _num_valid_values -= max_values; |
| 126 | + _last_value = std::string {prefix}; |
| 127 | + |
| 128 | + if (_num_valid_values == 0) { |
| 129 | + _last_value_in_previous_page = _last_value; |
| 130 | + } |
| 131 | + *out_num_values = max_values; |
| 132 | + return Status::OK(); |
| 133 | +} |
| 134 | +} // namespace doris::vectorized |
0 commit comments