看例子：

開兩個終端來對比：

在終端A:

[pgsql@localhost bin]$ ./psql psql (9.1.2) Type "help" for help.pgsql=# begin; BEGIN pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 2 (2 rows)pgsql=# insert into tab01 values(3,'3'); INSERT 0 1 pgsql=# insert into tab01 values(4,'4'); INSERT 0 1 pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 21885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 4 (4 rows)pgsql=#

此時的終端B:

[pgsql@localhost bin]$ ./psql psql (9.1.2) Type "help" for help.pgsql=# select xmin,xmax,cmin,cmax, * from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 2 (2 rows)pgsql=#

然后再在終端A進行提交：

pgsql=# commit; COMMIT pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 21885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 4 (4 rows)pgsql=#

此時，再在終端B進行觀察：

pgsql=# select xmin,xmax,cmin,cmax, * from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 21885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 4 (4 rows)pgsql=#

?繼續研究cmin是咋回事：

pgsql=# begin; BEGIN pgsql=# insert into tab01(id,cd) values(generate_series(5,6),'xx'); INSERT 0 2 pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+-----------+----1878 | 0 | 0 | 0 | 1 | 11884 | 0 | 0 | 0 | 999888777 | 21885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 41886 | 0 | 0 | 0 | 5 | xx1886 | 0 | 0 | 0 | 6 | xx (6 rows)pgsql=#

可以說cmin可理解為一個事務里，執行了幾次sql命令的順序。

那么cmax呢？在前面的基礎上繼續執行，居然沒有看到區別：

pgsql=# update tab01 set id=2 where cd = '2'; UPDATE 1 pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+----+----1878 | 0 | 0 | 0 | 1 | 11885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 41886 | 0 | 0 | 0 | 5 | xx1886 | 0 | 0 | 0 | 6 | xx1886 | 0 | 1 | 1 | 2 | 2 (6 rows)pgsql=# commit; COMMIT pgsql=# select xmin,xmax,cmin,cmax,* from tab01;xmin | xmax | cmin | cmax | id | cd ------+------+------+------+----+----1878 | 0 | 0 | 0 | 1 | 11885 | 0 | 0 | 0 | 3 | 31885 | 0 | 1 | 1 | 4 | 41886 | 0 | 0 | 0 | 5 | xx1886 | 0 | 0 | 0 | 6 | xx1886 | 0 | 1 | 1 | 2 | 2 (6 rows)pgsql=#

?經過反復折騰，終于發現，其實 cmin和 cmax就是一個東西：

看源代碼：

/* ----------------* heap_getsysattr** Fetch the value of a system attribute for a tuple.** This is a support routine for the heap_getattr macro. The macro* has already determined that the attnum refers to a system attribute.* ----------------*/ Datum heap_getsysattr(HeapTuple tup, int attnum, TupleDesc tupleDesc, bool *isnull) {Datum result;Assert(tup);/* Currently, no sys attribute ever reads as NULL. */*isnull = false;switch (attnum){case SelfItemPointerAttributeNumber:/* pass-by-reference datatype */result = PointerGetDatum(&(tup->t_self));break;case ObjectIdAttributeNumber:result = ObjectIdGetDatum(HeapTupleGetOid(tup));break;case MinTransactionIdAttributeNumber:result = TransactionIdGetDatum(HeapTupleHeaderGetXmin(tup->t_data));break;case MaxTransactionIdAttributeNumber:result = TransactionIdGetDatum(HeapTupleHeaderGetXmax(tup->t_data));break;case MinCommandIdAttributeNumber:case MaxCommandIdAttributeNumber:/** cmin and cmax are now both aliases for the same field, which* can in fact also be a combo command id. XXX perhaps we should* return the "real" cmin or cmax if possible, that is if we are* inside the originating transaction?*/result = CommandIdGetDatum(HeapTupleHeaderGetRawCommandId(tup->t_data));break;case TableOidAttributeNumber:result = ObjectIdGetDatum(tup->t_tableOid);break;default:elog(ERROR, "invalid attnum: %d", attnum);result = 0; /* keep compiler quiet */break;}return result; }

?

轉載于:https://www.cnblogs.com/gaojian/p/3165178.html

總結

以上是生活随笔為你收集整理的对PostgreSQL cmin和cmax的理解的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。