COPY table_name [ ( column_name [, ...] ) ]
FROM { 'filename' | STDIN }
[ [ WITH ] ( option [, ...] ) ]
COPY { table_name [ ( column_name [, ...] ) ] | ( query ) }
TO { 'filename' | STDOUT }
[ [ WITH ] ( option [, ...] ) ]
where option can be one of:
FORMAT format_name
OIDS [ boolean ]
DELIMITER 'delimiter_character'
NULL 'null_string'
HEADER [ boolean ]
QUOTE 'quote_character'
ESCAPE 'escape_character'
FORCE_QUOTE { ( column_name [, ...] ) | * }
FORCE_NOT_NULL ( column_name [, ...] )
ENCODING 'encoding_name'
FORMAT:输入/输出格式(text, csv, binary),默认的是text。
DELIMITER:输入/输出文本的分割符,文本文件默认的是tab,csv默认的逗号,此选项对二进制文件无效。另:分隔符只能为单字符。
NULL:代表空值的字符。
HEADER:是否包含第一行,只对csv文件有效。
copy导入的时候默认将整个文件的内容copy到整个表里,除此之外,可以指定导入的表的字段,导出的时候除了支持(所有字段/部分字段)导出外,还支持自定义查询语句导出。因为整个的导入导出操作是在一个事务里完成的,所以速度比insert要快。
bytea是二进制字符串,类似于varchar/text,但其存储单位不是字符而是字节,它可以用来存储一个字节序列,其中可以包含0字节(zero octet)以及其他一些不可打印(non-printable,值在[32, 126]范围之外的)字符。在用户看来这些就是原始的字节(raw bytes),不像varchar/tex,bytea没有什么字面意义。在postgresql中还有OID类型,用来专门操作大对象,二者的对比如下表所示:
Characteristic |
BYTEA |
OID |
Max. allowed space |
1 GB |
2 GB |
Data access |
As a whole |
Stream-style |
Storage |
In defined table |
In pg_largeobject system table |
Data manipulation |
Using SQL and escaping sequnces |
Only within transaction block by special functions |
Loading |
Preload |
On demand |
如图片等文件,不能直接read之后存,因为不是postgresql支持的标准二进制格式,所以不支持。如果要存的话,需要转一步:
xxd -p /home/user/myimage.png | tr -d '\n' > /tmp/image.hex
echo "
-- CREATE TABLE hexdump (hex text);
DELETE FROM hexdump;
COPY hexdump FROM '/tmp/image.hex';
-- CREATE TABLE bindump (binarydump bytea);
DELETE FROM bindump;
INSERT INTO bindump (binarydump) (SELECT decode(hex, 'hex') FROM hexdump limit 1);
" | psql mydatabase
其中,xxd的作用就是将一个文件以十六进制的形式显示出来。图片文件可以以二进制的形式存储,但这些二进制的字节数据不能写到文件里,需要用能识别这些字节数据的工具。比如在python中可以这样:
from PIL import Image
import psycopg2
import StringIO
if __name__ == "__main__":
conn = psycopg2.connect(host="127.0.0.1", user="postgres", password="", database="postgres")
cur = conn.cursor()
query_sql = "select binarydump from bindump limit 1"
cur.execute(query_sql)
rcd = cur.fetchone()
img_data = rcd[0]
img = Image.open(StringIO.StringIO(img_data))
img.save("1.png","PNG")
在SQL语句中,可以用decode函数,将普通的字节码数据转为标准的bytea类型,在python中,可以借助psycopg2.Binary(buffer)。
import psycopg2
if __name__ == "__main__":
conn = psycopg2.connect(host="127.0.0.1", user="postgres", password="", database="postgres")
cur = conn.cursor()
img_buffer = None
with open("1.png") as reader:
img_buffer = reader.read()
insert_sql = "insert into bindump(binarydump) values(%s)"
params = (psycopg2.Binary(img_buffer),)
cur.execute(insert_sql, params)
conn.commit()
再来个sql函数,如果当前表里已经有了该图片就更新,没有就插入:
CREATE OR REPLACE FUNCTION update_img(img_id character varying,img_name character varying,img_data bytea)
RETURNS void AS
$BODY$
declare
img_exists boolean;
begin
select exists(select 1 from my_img where imgid=img_id) into img_exists;
if(img_exists) then
update my_img
set imgname = img_name,
imgdata = img_data
where imgid = img_id;
else
insert into my_img values(img_id, img_name, img_data);
end if;
end;
$BODY$
LANGUAGE plpgsql VOLATILE
COST 100;
ALTER FUNCTION update_img(character varying, character varying, bytea)
OWNER TO postgres;
http://www.postgresql.org/docs/9.2/static/sql-copy.html
http://initd.org/psycopg/docs/module.html
http://dba.stackexchange.com/questions/1742/how-to-insert-file-data-into-a-postgresql-bytea-column
https://www.microolap.com/products/connectivity/postgresdac/help/tipsandtricks_byteavsoid.htm