Question

考虑使用以下属性进行表名测试

+-------+------------+---------------------------+----------+
| id    | test_case  | file_name                 | coverage |
+-------+------------+---------------------------+----------+
|  8645 | test case1 | /vendor/src/kmod/vendor.c |       32 |
| 12456 | test case4 | /vendor/src/kmod/vendor.c |       28 |
| 20258 | test case3 | /vendor/src/kmod/vendor.c |       30 |
+-------+------------+---------------------------+----------+

如同在此表中每次进入表中的新插入时，它必须检查新覆盖值是否大于现有的最小覆盖值，如果为真，我需要更新具有现有最小覆盖值的新条目，以便我可以只保留表中最高的3个覆盖率值。

考虑一个新的条目覆盖= 29因为它大于min（coverage）= 28然后我需要更新覆盖值29和测试用例名称。

使用python-mysql我这样做请让我知道python mysql查询/代码来执行以下操作。

我使用python

从列表中向表中插入值

for x,y in zip(out2,out4):

    cur.execute("insert into testing(test_case,file_name,coverage) values('test case8',%s,%s) on duplicate file_name like "%s" update coverage=case when values(coverage) then values(coverage) else coverage end",(x,y,x))
    db.commit()

所以无论何时出现新值，都必须检查min（coverage）值。如果新值大于它必须更新的现有min（coverage），否则它需要忽略该循环

Answer 1

我认为这对于插件来说太复杂了。在mysql中我会使用一个过程，但你可能想在前端执行此操作。请注意，如果发生某些事件，我不确定你想要什么。

drop table if exists t;
create table t( id    int , test_case varchar(20), file_name varchar(30), coverage int);

drop procedure if exists t;
delimiter $$
create procedure t(inid    int , intest_case varchar(20), infile_name varchar(30), incoverage int)
begin
    declare cnt int;
    declare minid int;
    declare minvoverage int;
    declare covexists int;
    set cnt = (select count(*) from t where file_name = infile_name);
    set @mincoverage = (select min(coverage) from t where file_name = infile_name);
    set minid = (select min(id) from t where file_name = infile_name and coverage = @mincoverage);
    if cnt < 3  then
        insert into t values (inid,intest_case,infile_name, incoverage);
    else
      if @mincoverage <> incoverage then #retain existsing detail
        if not exists(select 1 from t where file_name = infile_name and coverage = incoverage) then #retain existing detail
            update t
                set test_case = intest_case, coverage = incoverage
                where file_name = infile_name and
                    id = minid;
        end if;
        end if;
    end if;
end$$
delimiter ;


call t(8645 , 'test case1' , '/vendor/src/kmod/vendor.c' ,       32 );
call t(12456, 'test case2' , '/vendor/src/kmod/vendor.c' ,       28 );
call t(20258, 'test case3' , '/vendor/src/kmod/vendor.c' ,       30 );
call t(30000, 'test case4' , '/vendor/src/kmod/vendor.c' ,       27 );

select * from t;

结果

+-------+------------+---------------------------+----------+
| id    | test_case  | file_name                 | coverage |
+-------+------------+---------------------------+----------+
|  8645 | test case1 | /vendor/src/kmod/vendor.c |       32 |
| 12456 | test case4 | /vendor/src/kmod/vendor.c |       27 |
| 20258 | test case3 | /vendor/src/kmod/vendor.c |       30 |
+-------+------------+---------------------------+----------+
3 rows in set (0.00 sec)

如果当前值大于现有最小值，则更新mysql中的表（如果不忽略该更新）

1 个答案: