|
1
|
|
|
<?php |
|
2
|
|
|
declare(strict_types=1); |
|
3
|
|
|
|
|
4
|
|
|
namespace Level23\Druid\Concerns; |
|
5
|
|
|
|
|
6
|
|
|
use Level23\Druid\InputFormats\FlattenSpec; |
|
7
|
|
|
use Level23\Druid\InputFormats\CsvInputFormat; |
|
8
|
|
|
use Level23\Druid\InputFormats\TsvInputFormat; |
|
9
|
|
|
use Level23\Druid\InputFormats\OrcInputFormat; |
|
10
|
|
|
use Level23\Druid\InputFormats\JsonInputFormat; |
|
11
|
|
|
use Level23\Druid\InputFormats\ParquetInputFormat; |
|
12
|
|
|
use Level23\Druid\InputFormats\ProtobufInputFormat; |
|
13
|
|
|
use Level23\Druid\InputFormats\InputFormatInterface; |
|
14
|
|
|
|
|
15
|
|
|
trait HasInputFormat |
|
16
|
|
|
{ |
|
17
|
|
|
/** |
|
18
|
|
|
* @var \Level23\Druid\InputFormats\InputFormatInterface|null |
|
19
|
|
|
*/ |
|
20
|
|
|
protected ?InputFormatInterface $inputFormat = null; |
|
21
|
|
|
|
|
22
|
|
|
/** |
|
23
|
|
|
* Specify that we use JSON as input format. |
|
24
|
|
|
* |
|
25
|
|
|
* @param FlattenSpec|null $flattenSpec Specifies flattening configuration for nested JSON data. See |
|
26
|
|
|
* flattenSpec for more info. |
|
27
|
|
|
* @param array<string,bool>|null $features JSON parser features supported by Jackson library. Those features |
|
28
|
|
|
* will be applied when parsing the input JSON data. |
|
29
|
|
|
* |
|
30
|
|
|
* @see https://github.com/FasterXML/jackson-core/wiki/JsonParser-Features |
|
31
|
|
|
*/ |
|
32
|
1 |
|
public function jsonFormat(?FlattenSpec $flattenSpec = null, ?array $features = null): self |
|
33
|
|
|
{ |
|
34
|
1 |
|
$this->inputFormat = new JsonInputFormat($flattenSpec, $features); |
|
35
|
|
|
|
|
36
|
1 |
|
return $this; |
|
37
|
|
|
} |
|
38
|
|
|
|
|
39
|
|
|
/** |
|
40
|
|
|
* Specify that we use CSV as input format. |
|
41
|
|
|
* |
|
42
|
|
|
* @param string[]|null $columns Specifies the columns of the data. The columns should be in the same |
|
43
|
|
|
* order with the columns of your data. |
|
44
|
|
|
* @param string|null $listDelimiter A custom delimiter for multi-value dimensions. |
|
45
|
|
|
* @param bool|null $findColumnsFromHeader If this is set, the task will find the column names from the header |
|
46
|
|
|
* row. Note that skipHeaderRows will be applied before finding column |
|
47
|
|
|
* names from the header. For example, if you set skipHeaderRows to 2 |
|
48
|
|
|
* and findColumnsFromHeader to true, the task will skip the first two |
|
49
|
|
|
* lines and then extract column information from the third line. |
|
50
|
|
|
* columns will be ignored if this is set to true. |
|
51
|
|
|
* @param int $skipHeaderRows If this is set, the task will skip the first skipHeaderRows rows. |
|
52
|
|
|
*/ |
|
53
|
1 |
|
public function csvFormat( |
|
54
|
|
|
?array $columns = null, |
|
55
|
|
|
?string $listDelimiter = null, |
|
56
|
|
|
?bool $findColumnsFromHeader = null, |
|
57
|
|
|
int $skipHeaderRows = 0 |
|
58
|
|
|
): self { |
|
59
|
1 |
|
$this->inputFormat = new CsvInputFormat($columns, $listDelimiter, $findColumnsFromHeader, $skipHeaderRows); |
|
60
|
|
|
|
|
61
|
1 |
|
return $this; |
|
62
|
|
|
} |
|
63
|
|
|
|
|
64
|
|
|
/** |
|
65
|
|
|
* Specify that we use TSV as input format. |
|
66
|
|
|
* |
|
67
|
|
|
* @param array<string>|null $columns Specifies the columns of the data. The columns should be in the |
|
68
|
|
|
* same order with the columns of your data. |
|
69
|
|
|
* @param string|null $delimiter A custom delimiter for data values. |
|
70
|
|
|
* @param string|null $listDelimiter A custom delimiter for multi-value dimensions. |
|
71
|
|
|
* @param bool|null $findColumnsFromHeader If this is set, the task will find the column names from the |
|
72
|
|
|
* header row. Note that skipHeaderRows will be applied before |
|
73
|
|
|
* finding column names from the header. For example, if you set |
|
74
|
|
|
* skipHeaderRows to 2 and findColumnsFromHeader to true, the task |
|
75
|
|
|
* will skip the first two lines and then extract column |
|
76
|
|
|
* information from the third line. columns will be ignored if |
|
77
|
|
|
* this is set to true. |
|
78
|
|
|
* @param int $skipHeaderRows If this is set, the task will skip the first skipHeaderRows |
|
79
|
|
|
* rows. |
|
80
|
|
|
*/ |
|
81
|
1 |
|
public function tsvFormat( |
|
82
|
|
|
?array $columns = null, |
|
83
|
|
|
?string $delimiter = null, |
|
84
|
|
|
?string $listDelimiter = null, |
|
85
|
|
|
?bool $findColumnsFromHeader = null, |
|
86
|
|
|
int $skipHeaderRows = 0 |
|
87
|
|
|
): self { |
|
88
|
1 |
|
$this->inputFormat = new TsvInputFormat( |
|
89
|
|
|
$columns, |
|
90
|
|
|
$delimiter, |
|
91
|
|
|
$listDelimiter, |
|
92
|
|
|
$findColumnsFromHeader, |
|
93
|
|
|
$skipHeaderRows |
|
94
|
|
|
); |
|
95
|
|
|
|
|
96
|
1 |
|
return $this; |
|
97
|
|
|
} |
|
98
|
|
|
|
|
99
|
|
|
/** |
|
100
|
|
|
* Specify that we use ORC as input format. |
|
101
|
|
|
* |
|
102
|
|
|
* To use the ORC input format, load the Druid Orc extension ( druid-orc-extensions). |
|
103
|
|
|
* |
|
104
|
|
|
* @param FlattenSpec|null $flattenSpec Specifies flattening configuration for nested ORC data. See flattenSpec |
|
105
|
|
|
* for more info. |
|
106
|
|
|
* @param bool|null $binaryAsString Specifies if the binary orc column which is not logically marked as a |
|
107
|
|
|
* string should be treated as a UTF-8 encoded string. Default is false. |
|
108
|
|
|
*/ |
|
109
|
1 |
|
public function orcFormat(?FlattenSpec $flattenSpec = null, ?bool $binaryAsString = null): self |
|
110
|
|
|
{ |
|
111
|
1 |
|
$this->inputFormat = new OrcInputFormat($flattenSpec, $binaryAsString); |
|
112
|
|
|
|
|
113
|
1 |
|
return $this; |
|
114
|
|
|
} |
|
115
|
|
|
|
|
116
|
|
|
/** |
|
117
|
|
|
* Specify that we use Parquet as input format. |
|
118
|
|
|
* |
|
119
|
|
|
* To use the Parquet input format load the Druid Parquet extension (druid-parquet-extensions). |
|
120
|
|
|
* |
|
121
|
|
|
* @param FlattenSpec|null $flattenSpec Define a flattenSpec to extract nested values from a Parquet file. Note |
|
122
|
|
|
* that only 'path' expression are supported ('jq' is unavailable). |
|
123
|
|
|
* @param bool|null $binaryAsString Specifies if the bytes parquet column which is not logically marked as a |
|
124
|
|
|
* string or enum type should be treated as a UTF-8 encoded string. |
|
125
|
|
|
*/ |
|
126
|
1 |
|
public function parquetFormat(?FlattenSpec $flattenSpec = null, ?bool $binaryAsString = null): self |
|
127
|
|
|
{ |
|
128
|
1 |
|
$this->inputFormat = new ParquetInputFormat($flattenSpec, $binaryAsString); |
|
129
|
|
|
|
|
130
|
1 |
|
return $this; |
|
131
|
|
|
} |
|
132
|
|
|
|
|
133
|
|
|
/** |
|
134
|
|
|
* Specify that we use Protobuf as input format. |
|
135
|
|
|
* |
|
136
|
|
|
* You need to include the druid-protobuf-extensions as an extension to use the Protobuf input format. |
|
137
|
|
|
* |
|
138
|
|
|
* @param array<string,string> $protoBytesDecoder Specifies how to decode bytes to Protobuf record. See below for |
|
139
|
|
|
* an example. |
|
140
|
|
|
* @param FlattenSpec|null $flattenSpec Define a flattenSpec to extract nested values from a Parquet |
|
141
|
|
|
* file. Note that only 'path' expression are supported ('jq' is |
|
142
|
|
|
* unavailable). |
|
143
|
|
|
* |
|
144
|
|
|
* Example $protoBytesDecoder value: |
|
145
|
|
|
* ``` |
|
146
|
|
|
* [ |
|
147
|
|
|
* "type" => "file", |
|
148
|
|
|
* "descriptor" => "file:///tmp/metrics.desc", |
|
149
|
|
|
* "protoMessageType" => "Metrics" |
|
150
|
|
|
* ] |
|
151
|
|
|
* ``` |
|
152
|
|
|
* |
|
153
|
|
|
* @see https://druid.apache.org/docs/latest/ingestion/data-formats.html#protobuf |
|
154
|
|
|
*/ |
|
155
|
1 |
|
public function protobufFormat(array $protoBytesDecoder, ?FlattenSpec $flattenSpec = null): self |
|
156
|
|
|
{ |
|
157
|
1 |
|
$this->inputFormat = new ProtobufInputFormat($protoBytesDecoder, $flattenSpec); |
|
158
|
|
|
|
|
159
|
1 |
|
return $this; |
|
160
|
|
|
} |
|
161
|
|
|
} |
level23 /
druid-client
