be/src/storage/segment/parsed_page.h
Line | Count | Source |
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 | | #pragma once |
19 | | |
20 | | #include <gen_cpp/segment_v2.pb.h> |
21 | | |
22 | | #include <memory> |
23 | | |
24 | | #include "common/status.h" |
25 | | #include "storage/segment/binary_dict_page.h" |
26 | | #include "storage/segment/common.h" |
27 | | #include "storage/segment/encoding_info.h" |
28 | | #include "storage/segment/options.h" |
29 | | #include "storage/segment/page_decoder.h" |
30 | | #include "storage/segment/page_handle.h" |
31 | | #include "util/rle_encoding.h" |
32 | | |
33 | | namespace doris { |
34 | | namespace segment_v2 { |
35 | | |
36 | | // This contains information when one page is loaded, and ready for read |
37 | | // This struct can be reused, client should call reset first before reusing |
38 | | // this object |
39 | | struct ParsedPage { |
40 | | static Status create(PageHandle handle, const Slice& body, const DataPageFooterPB& footer, |
41 | | const EncodingInfo* encoding, const PagePointer& page_pointer, |
42 | | uint32_t page_index, ParsedPage* result, |
43 | 5.93M | PageDecoderOptions opts = PageDecoderOptions()) { |
44 | 5.93M | result->~ParsedPage(); |
45 | 5.93M | ParsedPage* page = new (result)(ParsedPage); |
46 | 5.93M | page->page_handle = std::move(handle); |
47 | | |
48 | 5.93M | auto null_size = footer.nullmap_size(); |
49 | 5.93M | page->has_null = null_size > 0; |
50 | 5.93M | page->null_bitmap = Slice(body.data + body.size - null_size, null_size); |
51 | | |
52 | 5.93M | if (page->has_null) { |
53 | 350k | page->null_decoder = |
54 | 350k | RleDecoder<bool>((const uint8_t*)page->null_bitmap.data, null_size, 1); |
55 | 350k | } |
56 | | |
57 | 5.93M | Slice data_slice(body.data, body.size - null_size); |
58 | 5.93M | PageDecoder* decoder; |
59 | 5.93M | RETURN_IF_ERROR(encoding->create_page_decoder(data_slice, opts, &decoder)); |
60 | 5.93M | page->data_decoder.reset(decoder); |
61 | 5.93M | RETURN_IF_ERROR(page->data_decoder->init()); |
62 | | |
63 | 5.93M | if (encoding->encoding() == DICT_ENCODING) { |
64 | 515k | auto dict_decoder = static_cast<BinaryDictPageDecoder*>(page->data_decoder.get()); |
65 | 515k | page->is_dict_encoding = dict_decoder->is_dict_encoding(); |
66 | 515k | } |
67 | | |
68 | 5.93M | page->first_ordinal = footer.first_ordinal(); |
69 | 5.93M | page->num_rows = footer.num_values(); |
70 | | |
71 | 5.93M | page->page_pointer = page_pointer; |
72 | 5.93M | page->page_index = page_index; |
73 | | |
74 | 5.93M | page->next_array_item_ordinal = footer.next_array_item_ordinal(); |
75 | | |
76 | 5.93M | return Status::OK(); |
77 | 5.93M | } |
78 | | |
79 | 29.6M | ~ParsedPage() { data_decoder = nullptr; } |
80 | | |
81 | | PageHandle page_handle; |
82 | | |
83 | | bool has_null; |
84 | | Slice null_bitmap; |
85 | | RleDecoder<bool> null_decoder; |
86 | | std::unique_ptr<PageDecoder> data_decoder; |
87 | | |
88 | | // ordinal of the first value in this page |
89 | | ordinal_t first_ordinal = 0; |
90 | | // number of rows including nulls and not-nulls |
91 | | ordinal_t num_rows = 0; |
92 | | // record it to get the last array element's size |
93 | | // should be none zero if set in page |
94 | | ordinal_t next_array_item_ordinal = 0; |
95 | | |
96 | | PagePointer page_pointer; |
97 | | uint32_t page_index = 0; |
98 | | |
99 | | // current offset when read this page |
100 | | // this means next row we will read |
101 | | ordinal_t offset_in_page = 0; |
102 | | |
103 | | bool is_dict_encoding = false; |
104 | | |
105 | 6.51M | bool contains(ordinal_t ord) { |
106 | 6.51M | return ord >= first_ordinal && ord < (first_ordinal + num_rows); |
107 | 6.51M | } |
108 | | |
109 | 8.12M | operator bool() const { return data_decoder != nullptr; } |
110 | | |
111 | 7.59M | bool has_remaining() const { return offset_in_page < num_rows; } |
112 | | |
113 | 10.7M | size_t remaining() const { return num_rows - offset_in_page; } |
114 | | }; |
115 | | |
116 | | } // namespace segment_v2 |
117 | | } // namespace doris |