表函数 - S3 - 《Apache Doris v2.0 中文文档》

S3

Name

SinceVersion 1.2

description

S3表函数（table-valued-function,tvf），可以让用户像访问关系表格式数据一样，读取并访问 S3 兼容的对象存储上的文件内容。目前支持csv/csv_with_names/csv_with_names_and_types/json/parquet/orc文件格式。

语法

s3(
  "uri" = "..",
  "s3.access_key" = "...",
  "s3.secret_key" = "...",
  "s3.region" = "...",
  "format" = "csv",
  "keyn" = "valuen",
  ...
  );

参数说明

S3 tvf中的每一个参数都是一个 "key"="value" 对。访问S3相关参数：

uri： (必填) 访问S3的uri，S3表函数会根据 use_path_style 参数来决定是否使用 path style 访问方式，默认为 virtual-hosted style 方式
s3.access_key： (必填)
s3.secret_key： (必填)
s3.region： (选填)。如果Minio服务设置了其他的region，那么必填，否则默认使用us-east-1。
s3.session_token： (选填)
use_path_style：(选填) 默认为false 。S3 SDK 默认使用 virtual-hosted style 方式。但某些对象存储系统可能没开启或没支持virtual-hosted style 方式的访问，此时我们可以添加 use_path_style 参数来强制使用 path style 方式。比如 minio默认情况下只允许path style访问方式，所以在访问minio时要加上use_path_style=true。

注意：uri目前支持三种schema：http://, https:// 和 s3://

如果使用http://或https://, 则会根据 ‘use_path_style’ 参数来决定是否使用’path style’方式访问s3
如果使用s3://, 则都使用 ‘virtual-hosted style’ 方式访问s3, ‘use_path_style’参数无效。
如果uri路径不存在或文件都是空文件，s3 tvf将返回空集合

详细使用案例可以参考最下方 Best Practice。

文件格式参数：

format：(必填) 目前支持 csv/csv_with_names/csv_with_names_and_types/json/parquet/orc
column_separator：(选填) 列分割符, 默认为,。
line_delimiter：(选填) 行分割符，默认为\n。
compress_type: (选填) 目前支持 UNKNOWN/PLAIN/GZ/LZO/BZ2/LZ4FRAME/DEFLATE。默认值为 UNKNOWN, 将会根据 uri 的后缀自动推断类型。

下面6个参数是用于json格式的导入，具体使用方法可以参照：Json Load
read_json_by_line： (选填) 默认为 "true"
strip_outer_array： (选填) 默认为 "false"
json_root： (选填) 默认为空
jsonpaths： (选填) 默认为空
num_as_string： (选填) 默认为 false
fuzzy_parse： (选填) 默认为 false

SinceVersion dev下面2个参数是用于csv格式的导入
trim_double_quotes：布尔类型，选填，默认值为 false，为 true 时表示裁剪掉 csv 文件每个字段最外层的双引号
skip_lines：整数类型，选填，默认值为0，含义为跳过csv文件的前几行。当设置format设置为 csv_with_names 或 csv_with_names_and_types 时，该参数会失效

其他参数：

path_partition_keys：（选填）指定文件路径中携带的分区列名，例如/path/to/city=beijing/date=”2023-07-09”, 则填写path_partition_keys="city,date"，将会自动从路径中读取相应列名和列值进行导入。

Example

读取并访问 S3 兼容的对象存储上的csv格式文件

select * from s3("uri" = "http://127.0.0.1:9312/test2/student1.csv",
                "s3.access_key"= "minioadmin",
                "s3.secret_key" = "minioadmin",
                "format" = "csv",
                "use_path_style" = "true") order by c1;

可以配合desc function使用

MySQL [(none)]> Desc function s3("uri" = "http://127.0.0.1:9312/test2/student1.csv",
                 "s3.access_key"= "minioadmin",
                 "s3.secret_key" = "minioadmin",
                 "format" = "csv",
                 "use_path_style" = "true");

Keywords

s3, table-valued-function, tvf

Best Practice

不同url schema的写法 http:// 、https:// 使用示例：

// 注意URI bucket写法以及use_path_style参数设置，http同理。
// 由于设置了"use_path_style"="true", 所以将采用path style方式访问s3。
select * from s3(
    "uri" = "https://endpoint/bucket/file/student.csv",
    "s3.access_key"= "ak",
    "s3.secret_key" = "sk",
    "format" = "csv",
    "use_path_style"="true");
// 注意URI bucket写法以及use_path_style参数设置，http同理。
// 由于设置了"use_path_style"="false", 所以将采用virtual-hosted style方式访问s3。
select * from s3(
    "uri" = "https://bucket.endpoint/bucket/file/student.csv",
    "s3.access_key"= "ak",
    "s3.secret_key" = "sk",
    "format" = "csv",
    "use_path_style"="false");
// 阿里云oss和腾讯云cos采用virtual-hosted style方式访问s3。
// OSS
select * from s3(
    "uri" = "http://example-bucket.oss-cn-beijing.aliyuncs.com/your-folder/file.parquet",
    "s3.access_key"= "ak",
    "s3.secret_key" = "sk",
    "s3.region" = "oss-cn-beijing",
    "format" = "parquet",
    "use_path_style" = "false");
// COS
select * from s3(
    "uri" = "https://example-bucket.cos.ap-hongkong.myqcloud.com/your-folder/file.parquet",
    "s3.access_key"= "ak",
    "s3.secret_key" = "sk",
    "s3.region" = "ap-hongkong",
    "format" = "parquet",
    "use_path_style" = "false");

s3:// 使用示例：

// 注意URI bucket写法, 无需设置use_path_style参数。
// 将采用virtual-hosted style方式访问s3。
select * from s3(
    "uri" = "s3://bucket.endpoint/file/student.csv",
    "s3.access_key"= "ak",
    "s3.secret_key" = "sk",
    "format" = "csv");

csv format 由于S3 table-valued-function事先并不知道table schema，所以会先读一遍文件来解析出table schema。

csv 格式: S3 table-valued-function 读取S3上的文件并当作csv文件来处理，读取文件中的第一行用于解析table schema。文件第一行的列个数n将作为table schema的列个数，table schema的列名则自动取名为c1, c2, ..., cn ，列类型都设置为 String, 举例:

student1.csv文件内容为：

1,ftw,12
2,zs,18
3,ww,20

使用S3 tvf

MySQL [(none)]> select * from s3("uri" = "http://127.0.0.1:9312/test2/student1.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv",
->                 "use_path_style" = "true") order by c1;
+------+------+------+
| c1   | c2   | c3   |
+------+------+------+
| 1    | ftw  | 12   |
| 2    | zs   | 18   |
| 3    | ww   | 20   |
+------+------+------+

可以配合 desc function S3() 来查看table schema

MySQL [(none)]> Desc function s3("uri" = "http://127.0.0.1:9312/test2/student1.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv",
->                 "use_path_style" = "true");
+-------+------+------+-------+---------+-------+
| Field | Type | Null | Key   | Default | Extra |
+-------+------+------+-------+---------+-------+
| c1    | TEXT | Yes  | false | NULL    | NONE  |
| c2    | TEXT | Yes  | false | NULL    | NONE  |
| c3    | TEXT | Yes  | false | NULL    | NONE  |
+-------+------+------+-------+---------+-------+

csv_with_names format csv_with_names格式：解析文件的第一行作为table schema的列个数和列名，列类型则都设置为 String, 举例：

student_with_names.csv文件内容为

id,name,age
1,ftw,12
2,zs,18
3,ww,20

使用S3 tvf

MySQL [(none)]> select * from s3("uri" = "http://127.0.0.1:9312/test2/student_with_names.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv_with_names",
->                 "use_path_style" = "true") order by id;
+------+------+------+
| id   | name | age  |
+------+------+------+
| 1    | ftw  | 12   |
| 2    | zs   | 18   |
| 3    | ww   | 20   |
+------+------+------+

同样配合desc function S3() 可查看table schema

MySQL [(none)]> Desc function s3("uri" = "http://127.0.0.1:9312/test2/student_with_names.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv_with_names",
->                 "use_path_style" = "true");
+-------+------+------+-------+---------+-------+
| Field | Type | Null | Key   | Default | Extra |
+-------+------+------+-------+---------+-------+
| id    | TEXT | Yes  | false | NULL    | NONE  |
| name  | TEXT | Yes  | false | NULL    | NONE  |
| age   | TEXT | Yes  | false | NULL    | NONE  |
+-------+------+------+-------+---------+-------+

csv_with_names_and_types foramt

csv_with_names_and_types格式：目前暂不支持从csv文件中解析出column type。使用该format时，S3 tvf会解析文件的第一行作为table schema的列个数和列名，列类型则都设置为 String，同时将忽略该文件的第二行。

student_with_names_and_types.csv文件内容为

id,name,age
INT,STRING,INT
1,ftw,12
2,zs,18
3,ww,20

使用S3 tvf

MySQL [(none)]> select * from s3("uri" = "http://127.0.0.1:9312/test2/student_with_names_and_types.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv_with_names_and_types",
->                 "use_path_style" = "true") order by id;
+------+------+------+
| id   | name | age  |
+------+------+------+
| 1    | ftw  | 12   |
| 2    | zs   | 18   |
| 3    | ww   | 20   |
+------+------+------+

同样配合desc function S3() 可查看table schema

MySQL [(none)]> Desc function s3("uri" = "http://127.0.0.1:9312/test2/student_with_names_and_types.csv",
->                 "s3.access_key"= "minioadmin",
->                 "s3.secret_key" = "minioadmin",
->                 "format" = "csv_with_names_and_types",
->                 "use_path_style" = "true");
+-------+------+------+-------+---------+-------+
| Field | Type | Null | Key   | Default | Extra |
+-------+------+------+-------+---------+-------+
| id    | TEXT | Yes  | false | NULL    | NONE  |
| name  | TEXT | Yes  | false | NULL    | NONE  |
| age   | TEXT | Yes  | false | NULL    | NONE  |
+-------+------+------+-------+---------+-------+

json format

json 格式：json格式涉及到较多的可选参数，各个参数的意义可以参考：Json Load。 S3 tvf查询json格式文件时根据 json_root 和 jsonpaths 参数定位到一个json对象，将该对象的中的key 作为table schema的列名，列类型都设置为String。举例：

data.json文件

[{"id":1, "name":"ftw", "age":18}]
[{"id":2, "name":"xxx", "age":17}]
[{"id":3, "name":"yyy", "age":19}]

使用S3 tvf查询

MySQL [(none)]> select * from s3(
    "uri" = "http://127.0.0.1:9312/test2/data.json",
    "s3.access_key"= "minioadmin",
    "s3.secret_key" = "minioadmin",
    "format" = "json",
    "strip_outer_array" = "true",
    "read_json_by_line" = "true",
    "use_path_style"="true");
+------+------+------+
| id   | name | age  |
+------+------+------+
| 1    | ftw  | 18   |
| 2    | xxx  | 17   |
| 3    | yyy  | 19   |
+------+------+------+
MySQL [(none)]> select * from s3(
    "uri" = "http://127.0.0.1:9312/test2/data.json",
    "s3.access_key"= "minioadmin",
    "s3.secret_key" = "minioadmin",
    "format" = "json",
    "strip_outer_array" = "true",
    "jsonpaths" = "[\"$.id\", \"$.age\"]",
    "use_path_style"="true");
+------+------+
| id   | age  |
+------+------+
| 1    | 18   |
| 2    | 17   |
| 3    | 19   |
+------+------+

parquet format

parquet 格式：S3 tvf支持从parquet文件中解析出table schema的列名、列类型。举例：

MySQL [(none)]> select * from s3(
    "uri" = "http://127.0.0.1:9312/test2/test.snappy.parquet",
    "s3.access_key"= "minioadmin",
    "s3.secret_key" = "minioadmin",
    "format" = "parquet",
    "use_path_style"="true") limit 5;
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+
| p_partkey | p_name                                   | p_mfgr         | p_brand  | p_type                  | p_size | p_container | p_retailprice | p_comment           |
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+
|         1 | goldenrod lavender spring chocolate lace | Manufacturer#1 | Brand#13 | PROMO BURNISHED COPPER  |      7 | JUMBO PKG   |           901 | ly. slyly ironi     |
|         2 | blush thistle blue yellow saddle         | Manufacturer#1 | Brand#13 | LARGE BRUSHED BRASS     |      1 | LG CASE     |           902 | lar accounts amo    |
|         3 | spring green yellow purple cornsilk      | Manufacturer#4 | Brand#42 | STANDARD POLISHED BRASS |     21 | WRAP CASE   |           903 | egular deposits hag |
|         4 | cornflower chocolate smoke green pink    | Manufacturer#3 | Brand#34 | SMALL PLATED BRASS      |     14 | MED DRUM    |           904 | p furiously r       |
|         5 | forest brown coral puff cream            | Manufacturer#3 | Brand#32 | STANDARD POLISHED TIN   |     15 | SM PKG      |           905 |  wake carefully     |
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+

MySQL [(none)]> desc function s3(
    "uri" = "http://127.0.0.1:9312/test2/test.snappy.parquet",
    "s3.access_key"= "minioadmin",
    "s3.secret_key" = "minioadmin",
    "format" = "parquet",
    "use_path_style"="true");
+---------------+--------------+------+-------+---------+-------+
| Field         | Type         | Null | Key   | Default | Extra |
+---------------+--------------+------+-------+---------+-------+
| p_partkey     | INT          | Yes  | false | NULL    | NONE  |
| p_name        | TEXT         | Yes  | false | NULL    | NONE  |
| p_mfgr        | TEXT         | Yes  | false | NULL    | NONE  |
| p_brand       | TEXT         | Yes  | false | NULL    | NONE  |
| p_type        | TEXT         | Yes  | false | NULL    | NONE  |
| p_size        | INT          | Yes  | false | NULL    | NONE  |
| p_container   | TEXT         | Yes  | false | NULL    | NONE  |
| p_retailprice | DECIMAL(9,0) | Yes  | false | NULL    | NONE  |
| p_comment     | TEXT         | Yes  | false | NULL    | NONE  |
+---------------+--------------+------+-------+---------+-------+

orc format

orc 格式：和parquet format使用方法一致, 将format参数设置为orc。

MySQL [(none)]> select * from s3(
    "uri" = "http://127.0.0.1:9312/test2/test.snappy.orc",
    "s3.access_key"= "minioadmin",
    "s3.secret_key" = "minioadmin",
    "format" = "orc",
    "use_path_style"="true") limit 5;
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+
| p_partkey | p_name                                   | p_mfgr         | p_brand  | p_type                  | p_size | p_container | p_retailprice | p_comment           |
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+
|         1 | goldenrod lavender spring chocolate lace | Manufacturer#1 | Brand#13 | PROMO BURNISHED COPPER  |      7 | JUMBO PKG   |           901 | ly. slyly ironi     |
|         2 | blush thistle blue yellow saddle         | Manufacturer#1 | Brand#13 | LARGE BRUSHED BRASS     |      1 | LG CASE     |           902 | lar accounts amo    |
|         3 | spring green yellow purple cornsilk      | Manufacturer#4 | Brand#42 | STANDARD POLISHED BRASS |     21 | WRAP CASE   |           903 | egular deposits hag |
|         4 | cornflower chocolate smoke green pink    | Manufacturer#3 | Brand#34 | SMALL PLATED BRASS      |     14 | MED DRUM    |           904 | p furiously r       |
|         5 | forest brown coral puff cream            | Manufacturer#3 | Brand#32 | STANDARD POLISHED TIN   |     15 | SM PKG      |           905 |  wake carefully     |
+-----------+------------------------------------------+----------------+----------+-------------------------+--------+-------------+---------------+---------------------+

avro format

avro 格式：S3 tvf支持从avro文件中解析出table schema的列名、列类型。举例：

select * from s3(
         "uri" = "http://127.0.0.1:9312/test2/person.avro",
         "ACCESS_KEY" = "ak",
         "SECRET_KEY" = "sk",
         "FORMAT" = "avro");
+--------+--------------+-------------+-----------------+
| name   | boolean_type | double_type | long_type       |
+--------+--------------+-------------+-----------------+
| Alyssa |            1 |     10.0012 | 100000000221133 |
| Ben    |            0 |    5555.999 |      4009990000 |
| lisi   |            0 | 5992225.999 |      9099933330 |
+--------+--------------+-------------+-----------------+

uri包含通配符

uri可以使用通配符来读取多个文件。注意：如果使用通配符要保证各个文件的格式是一致的(尤其是csv/csv_with_names/csv_with_names_and_types算做不同的格式)，S3 tvf用第一个文件来解析出table schema。如下两个csv文件：

// file1.csv
1,aaa,18
2,qqq,20
3,qwe,19
// file2.csv
5,cyx,19
6,ftw,21

可以在uri上使用通配符来导入。

MySQL [(none)]> select * from s3(
        "uri" = "http://127.0.0.1:9312/test2/file*.csv",
        "s3.access_key"= "minioadmin",
        "s3.secret_key" = "minioadmin",
        "format" = "csv",
        "use_path_style"="true");
+------+------+------+
| c1   | c2   | c3   |
+------+------+------+
| 1    | aaa  | 18   |
| 2    | qqq  | 20   |
| 3    | qwe  | 19   |
| 5    | cyx  | 19   |
| 6    | ftw  | 21   |
+------+------+------+

配合 insert into 和 cast 使用 S3 tvf

// 创建doris内部表
CREATE TABLE IF NOT EXISTS ${testTable}
    (
        id int,
        name varchar(50),
        age int
    )
    COMMENT "my first table"
    DISTRIBUTED BY HASH(id) BUCKETS 32
    PROPERTIES("replication_num" = "1");
// 使用S3插入数据
insert into ${testTable} (id,name,age)
select cast (id as INT) as id, name, cast (age as INT) as age
from s3(
    "uri" = "${uri}",
    "s3.access_key"= "${ak}",
    "s3.secret_key" = "${sk}",
    "format" = "${format}",
    "strip_outer_array" = "true",
    "read_json_by_line" = "true",
    "use_path_style" = "true");