-
Notifications
You must be signed in to change notification settings - Fork 3.3k
/
byte_array_decoder.go
130 lines (110 loc) · 4.34 KB
/
byte_array_decoder.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package encoding
import (
"encoding/binary"
"github.com/apache/arrow/go/v17/arrow"
"github.com/apache/arrow/go/v17/arrow/array"
"github.com/apache/arrow/go/v17/arrow/memory"
"github.com/apache/arrow/go/v17/internal/utils"
"github.com/apache/arrow/go/v17/parquet"
pqutils "github.com/apache/arrow/go/v17/parquet/internal/utils"
"golang.org/x/xerrors"
)
// PlainByteArrayDecoder decodes a data chunk for bytearrays according to
// the plain encoding. The byte arrays will use slices to reference the
// data rather than copying it.
//
// The parquet spec defines Plain encoding for ByteArrays as a 4 byte little
// endian integer containing the length of the bytearray followed by that many
// bytes being the raw data of the byte array.
type PlainByteArrayDecoder struct {
decoder
}
// Type returns parquet.Types.ByteArray for this decoder
func (PlainByteArrayDecoder) Type() parquet.Type {
return parquet.Types.ByteArray
}
// Decode will populate the slice of bytearrays in full or until the number
// of values is consumed.
//
// Returns the number of values that were decoded.
func (pbad *PlainByteArrayDecoder) Decode(out []parquet.ByteArray) (int, error) {
max := utils.Min(len(out), pbad.nvals)
for i := 0; i < max; i++ {
// there should always be at least four bytes which is the length of the
// next value in the data.
if len(pbad.data) < 4 {
return i, xerrors.New("parquet: eof reading bytearray")
}
// the first 4 bytes are a little endian int32 length
byteLen := int32(binary.LittleEndian.Uint32(pbad.data[:4]))
if byteLen < 0 {
return i, xerrors.New("parquet: invalid BYTE_ARRAY value")
}
if int64(len(pbad.data)) < int64(byteLen)+4 {
return i, xerrors.New("parquet: eof reading bytearray")
}
out[i] = pbad.data[4 : byteLen+4 : byteLen+4]
pbad.data = pbad.data[byteLen+4:]
}
pbad.nvals -= max
return max, nil
}
// DecodeSpaced is like Decode, but expands the slice out to leave empty values
// where the validBits bitmap has 0s
func (pbad *PlainByteArrayDecoder) DecodeSpaced(out []parquet.ByteArray, nullCount int, validBits []byte, validBitsOffset int64) (int, error) {
toRead := len(out) - nullCount
valuesRead, err := pbad.Decode(out[:toRead])
if err != nil {
return valuesRead, err
}
if valuesRead != toRead {
return valuesRead, xerrors.New("parquet: number of values / definition levels read did not match")
}
return spacedExpand(out, nullCount, validBits, validBitsOffset), nil
}
func (d *DictByteArrayDecoder) InsertDictionary(bldr array.Builder) error {
conv := d.dictValueDecoder.(*ByteArrayDictConverter)
dictLength := cap(conv.dict)
conv.ensure(pqutils.IndexType(dictLength))
byteArrayData := memory.NewResizableBuffer(d.mem)
defer byteArrayData.Release()
byteArrayOffsets := memory.NewResizableBuffer(d.mem)
defer byteArrayOffsets.Release()
var totalLen int
for _, v := range conv.dict {
totalLen += len(v)
}
byteArrayData.ResizeNoShrink(totalLen)
byteArrayOffsets.ResizeNoShrink((dictLength + 1) * arrow.Int32SizeBytes)
byteData := byteArrayData.Bytes()
byteOffsets := arrow.Int32Traits.CastFromBytes(byteArrayOffsets.Bytes())
var offset int32
for i, v := range conv.dict {
n := copy(byteData, v)
byteData, byteOffsets[i] = byteData[n:], offset
offset += int32(n)
}
byteOffsets[dictLength] = offset
data := array.NewData(bldr.Type().(*arrow.DictionaryType).ValueType, dictLength,
[]*memory.Buffer{nil, byteArrayOffsets, byteArrayData}, nil, 0, 0)
defer data.Release()
arr := array.NewBinaryData(data)
defer arr.Release()
binaryBldr := bldr.(*array.BinaryDictionaryBuilder)
return binaryBldr.InsertDictValues(arr)
}