PostgreSQL Server Encoding sql_ascii attention
背景
群里一位兄弟问到的一个问题 :
” 我有一个postgresql，比较大，编码是sqlascii码，我想转换成有utf8 ，有方案可行吗? “
如果使用场景中未用到non-ASCII编码的字符, 那很幸运, 导出导入就可以了.
但是如果场景中使用了non-ASCII编码的字符, 那就没那么幸运了, 因为SQL_ASCII字符集不会对non-ASCII字符做合法性检验, 同时不做任何的编码转换, 如果client_encoding也设置了sql_ascii时，那么客户端上来是什么值就是什么值，而且不做字符合法性检测（因为都设置了sql_ascii后就无评判标准了，所以无法做校验），所以啥都能存进来(除了空字符0x00).
正文
初始化一个SQL_ASCII编码的数据库
pgdev@db-172-16-3-150-> initdb -D $PGDATA -E SQL_ASCII --locale=C -U postgres
pg_ctl start
-- 当前服务器编码
pgdev@db-172-16-3-150-> locale
LANG=en_US.utf8
LC_CTYPE="en_US.utf8"
LC_NUMERIC="en_US.utf8"
LC_TIME="en_US.utf8"
LC_COLLATE="en_US.utf8"
LC_MONETARY="en_US.utf8"
LC_MESSAGES="en_US.utf8"
LC_PAPER="en_US.utf8"
LC_NAME="en_US.utf8"
LC_ADDRESS="en_US.utf8"
LC_TELEPHONE="en_US.utf8"
LC_MEASUREMENT="en_US.utf8"
LC_IDENTIFICATION="en_US.utf8"
LC_ALL=
psql
postgres=# create role digoal login encrypted password 'digoal';
CREATE ROLE
postgres=# create database digoal owner digoal;
CREATE DATABASE
digoal=# \l
List of databases
Name | Owner | Encoding | Collate | Ctype | Access privileges
-----------+----------+-----------+---------+-------+-----------------------
digoal | digoal | SQL_ASCII | C | C |
postgres=# \c digoal digoal
digoal=> create schema digoal;
CREATE SCHEMA
digoal=> create table t(id int, info text);
CREATE TABLE
digoal=> insert into t values (1, '中国');
INSERT 0 1
插入中文没有问题, 当前插入的汉字是UTF8编码.
接下来把服务器改成gbk，插入GBK编码字符
digoal=> \q
pgdev@db-172-16-3-150-> export LANG=zh_CN.gbk
pgdev@db-172-16-3-150-> locale
LANG=zh_CN.gbk
LC_CTYPE="zh_CN.gbk"
LC_NUMERIC="zh_CN.gbk"
LC_TIME="zh_CN.gbk"
LC_COLLATE="zh_CN.gbk"
LC_MONETARY="zh_CN.gbk"
LC_MESSAGES="zh_CN.gbk"
LC_PAPER="zh_CN.gbk"
LC_NAME="zh_CN.gbk"
LC_ADDRESS="zh_CN.gbk"
LC_TELEPHONE="zh_CN.gbk"
LC_MEASUREMENT="zh_CN.gbk"
LC_IDENTIFICATION="zh_CN.gbk"
LC_ALL=
同时修改一下secureCRT的字符集，同样设置为GBK.
pgdev@db-172-16-3-150-> psql digoal digoal
psql (9.3devel)
Type "help" for help.
digoal=> insert into t values (2, '美国');
INSERT 0 1
digoal=> insert into t values (3, '德国');
INSERT 0 1
插入中文没有问题, 当前插入的汉字是GBK编码.
digoal=> select *,info::bytea from t;
id | info | info
----+--------+----------------
1 | 涓浗 | \xe4b8ade59bbd
2 | 美国 | \xc3c0b9fa
3 | 德国 | \xb5c2b9fa
由于’中国’ 是以UTF8编码插入的, 美国德国是以GBK编码插入的,
像这种保护混合编码的数值要导出并导入到另一个UTF8编码的数据库, GBK编码的必然会导致导入失败.
后面使用BYTEA插入UTF8字符集的数据库看看就知道了.
插入一些字节流, SQL_ASCII对non-ASCII也就是0-127以外的值都不做判断, 直接插入.
digoal=> insert into t values (4, E'\xe2\x82\xad');
INSERT 0 1
0x00为0-127范围内的值, SQL_ASCII编码的无效字节. 这个是被正常的检测出来了.
digoal=> insert into t values (4, E'\xe2\x82\x00');
ERROR: invalid byte sequence for encoding "SQL_ASCII": 0x00
其他字节流随便插入.
digoal=> insert into t values (4, E'\xe2\x82\x01');
INSERT 0 1
digoal=> insert into t values (4, E'\xe2\x82\xae');
INSERT 0 1
由于服务端是SQL_ASCII的, 所以虽然以下编码在UTF8中是非法的, 但是插入到SQL_ASCII的数据库还是没有问题.
digoal=> set client_encoding='UTF8';
SET
digoal=> insert into t values (4, E'\xe2\x82\xae');
INSERT 0 1
digoal=> insert into t values (4, E'\xe2\x82\x0e');
INSERT 0 1
digoal=> insert into t values (4, E'\x8f');
INSERT 0 1
查询结果 :
digoal=> select * from t;
id | info
----+--------
1 | 涓浗
2 | 美国
3 | 德国
4 | 鈧
4 | 鈧
4 | 鈧\x01
4 | 鈧
4 | 鈧
4 | 鈧\x0E
4 |
4 |
(11 rows)
如果数据库是UTF8字符集的话 :
postgres=# \l
List of databases
Name | Owner | Encoding | Collate | Ctype | Access privileges
-------------+----------+----------+---------+-------+-----------------------
digoal | digoal | UTF8 | C | C |
UTF8的非法字符将不允许插入.
postgres=# insert into t values (4, E'\x8f');
ERROR: invalid byte sequence for encoding "UTF8": 0x8f
postgres=# insert into t values (2, E'\xc3\xc0\xb9\xfa');
ERROR: invalid byte sequence for encoding "UTF8": 0xc3 0xc0
postgres=# insert into t values (2, E'\xb5\xc2\xb9\xfa');
ERROR: invalid byte sequence for encoding "UTF8": 0xb5
即使将客户端编码改成SQL_ASCII, 插入到数据库时也是会报错的.
postgres=# set client_encoding='SQL_ASCII';
SET
postgres=# insert into t values (2, E'\xc3\xc0\xb9\xfa');
ERROR: invalid byte sequence for encoding "UTF8": 0xc3 0xc0
postgres=# insert into t values (2, E'\xb5\xc2\xb9\xfa');
ERROR: invalid byte sequence for encoding "UTF8": 0xb5
postgres=# insert into t values (2, E'\xc3\xc0\xb9\xfa');
ERROR: 22021: invalid byte sequence for encoding "UTF8": 0xc3 0xc0
LOCATION: report_invalid_encoding, wchar.c:2015
所以如果存在混合编码的数据插入到SQL_ASCII数据库后, 要把数据正常的导入到UTF8字符集的数据库中, 是一件非常困难的事情.
需将数据导出, 并做好相应的转码再导入到数据库中.
再说说client_encoding，我们也已经看到了，实际上在数据库中存在两层编码校验，一层是client_encoding，另一层是database的encoding。
连接到数据库编码=SQL_ASCII的数据库
db=# \l+
List of databases
Name | Owner | Encoding | Collate | Ctype | Access privileges | Size | Tablespace | Description
-----------+----------+-----------+------------+------------+-----------------------+---------+------------+--------------------------------------------
db | postgres | SQL_ASCII | C | C | | 7335 kB | pg_default |
将client_encoding设置为sql_ascii，现在除了0x00，你可以肆无忌惮的输入
db=# set client_encoding='SQL_ASCII';
SET
db=# select E'\xee';
?column?
----------
(1 row)
将client_encoding设置为其他字符集，你又受到约束了
db=# set client_encoding='GBK';
SET
db=# select E'\xee';
ERROR: invalid byte sequence for encoding "GBK": 0xee
postgres=# set client_encoding='UTF8';
SET
postgres=# select E'\xee';
ERROR: invalid byte sequence for encoding "UTF8": 0xee
convert函数观察编码合法性
postgres=# \df *convert*
List of functions
Schema | Name | Result data type | Argument data types | Type
------------+--------------+------------------+---------------------+--------
pg_catalog | convert | bytea | bytea, name, name | normal
pg_catalog | convert_from | text | bytea, name | normal
pg_catalog | convert_to | bytea | text, name | normal
(3 rows)
postgres数据库编码为UTF8
postgres=# select convert_from('\xee'::bytea,'UTF8');
ERROR: invalid byte sequence for encoding "UTF8": 0xee
库级编码校验报错
postgres=# select convert_from('\xee'::bytea,'sql_ascii');
ERROR: invalid byte sequence for encoding "UTF8": 0xee
postgres=# set client_encoding='SQL_ASCII';
SET
postgres=# select convert_from('\xee'::bytea,'sql_ascii');
ERROR: invalid byte sequence for encoding "UTF8": 0xee
两个字符集直接转换
postgres=# select convert('\xee'::bytea,'sql_ascii','sql_ascii');
convert
---------
\xee
(1 row)
小结
1. 如果数据库中需要存储非ASCII字符, 那么不推荐数据库使用SQL_ASCII字符集.
2. 使用SQL_ASCII字符集必须了解的几种情况 :
The SQL_ASCII setting behaves considerably differently from the other settings.
When the server character set is SQL_ASCII, the server interprets byte values 0-127 according to the ASCII standard, while byte values 128-255 are taken as uninterpreted characters.
No encoding conversion will be done when the setting is SQL_ASCII.
Thus, this setting is not so much a declaration that a specific encoding is in use, as a declaration of ignorance about the encoding.
In most cases, if you are working with any non-ASCII data, it is unwise to use the SQL_ASCII setting because PostgreSQL will be unable to help you by converting or validating non-ASCII characters.
PostgreSQL will allow superusers to create databases with SQL_ASCII encoding even when LC_CTYPE is not C or POSIX.
As noted above, SQL_ASCII does not enforce that the data stored in the database has any particular encoding, and so this choice poses risks of locale-dependent misbehavior.
Using this combination of settings is deprecated and may someday be forbidden altogether.
If the client character set is defined as SQL_ASCII, encoding conversion is disabled, regardless of the server's character set.
Just as for the server, use of SQL_ASCII is unwise unless you are working with all-ASCII data.
参考
1. http://www.postgresql.org/docs/9.2/static/multibyte.html
2. PostgreSQL WHY ERROR: invalid byte sequence for encoding “UTF8”
http://blog.163.com/digoal@126/blog/static/163877040201211281407682/