如何解决如何优化 Oracle UNION ALL 的两个 3 秒查询，耗时超过 200 秒......即使使用 rownum
我正在努力进一步优化以下查询。到目前为止，我已经在优化查询方面取得了很多成功，但现在我正在寻求其他人的帮助以寻求想法。感谢您的帮助。
以下查询需要 200 多秒才能完成，但构成“UNION ALL”的两个单独的 WITH 子句查询（receiving_1 和 receive_2）每个都需要不到 4 秒。
- receiving_1 -- 3.4 秒内接收 58 次；成本 295（无 UNION）
- receiving_2 -- 3.8 秒内收到 0 条记录；成本 295（无 UNION）
- UNION ALL of 1+2 -- 58 recs in 203.1 secs;成本 300（UNION ALL）
我希望“rownum”能帮助优化“UNION ALL”，但它没有。此尝试基于 AskTom --> 文件 --> sqlTechniques.zip（61 到 72 之间）的注释
当你有两个分别以光速运行的查询时 但不是这样在一起 通常是一个混合的“CBO/RBO”问题 使用具有启动 CBO 的功能的 RBO Rownum 可以是一个临时修复，直到所有事情都成为 CBO
注意：避免此问题的一种优化是嵌入的替换参数或游标参数，它们极大地减少了查询中关键点的行数。但是，由于我需要针对只读数据库通过数据库链接运行此查询，因此此查询无法使用以下选项。
- dbms_sql over 数据库链接：查询太长； dbs_sql 不接受数据库链接上的 lob；此外，DBA 不允许在只读数据库中运行任何 PL/sql。
- 流水线函数：流水线函数在 19c 和 11g 之间的数据库链接上存在已知错误，因此失败
- sys_refcursor 通过 PL/sql：数据库链接不支持游标
- 在 PL/sql 包中设置参数值并在视图中使用它们：不允许在只读数据库中执行任何代码（我仍在窃听 DBA 以允许这样做。如果我不能，这可能是我最大的希望优化以下查询以作为视图运行。）
注意：为了用尽可能少的代码提供一个好的测试用例，下面的 sql（4,584 个字符，88 个换行符）是实际 sql（93,683 个字符，2,014 个换行符）的大大简化版本我正在开发中。最初的 sql 已经经过多轮优化，将运行时间从一个多小时缩短到 10 秒以下。这个最新的挑战是必须将 sql 重写为视图，而不是带有嵌入替换参数的 sql 文件。到目前为止，替换参数是优化的关键。我现在正在重构 sql 以用作 Oracle 11gR2 中的视图，以便通过来自 Oracle 19c 的数据库链接进行调用。显然，视图没有参数。查询视图时，驱动参数（账本、OU、期间）将是WHERE子句条件。
大图：正在开发的实际 sql 是 6 个共享大量重复代码的报表查询的合并。我的工作是将所有六个源报告查询转换为有效的摘录，以便导入新的数据库模式以进行协调。
with
gccs as (
select
gcc.code_combination_id
from apps.gl_code_combinations gcc
where 1=1
AND regexp_like(gcc.segment1,'^[0-9]{4}$') -- only include 4-digit numbers
AND gcc.segment1 BETWEEN to_char('1500') and to_char('1504')
AND regexp_like(gcc.segment3,'^[0-9]{4}$') -- only include 4-digit numbers
AND ( gcc.segment3 BETWEEN to_char('5000') and to_char('5999')
OR gcc.segment3 in ('6000','6002','6003','6004','6005','6006','6007','6008','6009','6010','6011','6012','6013','6014','6017','6018','6019','6020','6021','6022','6023','6024','6025','6026','6027','6028','6029','6030','6031','6032','6033','6034','6035','6036','6037','6038','6039','6040','6041','6042','6043','6044','6045','6046','6047','6048','6049','6050','6051','6052','6053','6054','6055','6056','6058','6060','6061','6062','6063','6064','6065','6066','6067','6068','6069','6070','6071','6072','6073','6074','6075','6076','6077','6084','6085','6086','6087','6088','6089','6090','6091','6092','6093','6094','6095','6096','6097','6098','6099','6100','6101','6102','6103','6104','6105','6106','6120','6121','6130','6131','6140','6150','6151','6155','6158','6162','6990','6991','6992','6993','6994','6995','6996','6997','6998'))
AND regexp_like(gcc.segment6,'^[0-9]{4}$') -- only include 4-digit numbers
AND gcc.segment6 BETWEEN to_char('3000') and to_char('4999')
) -- select count(*) code_combination_id from gccs;,base_query as (
SELECT
gl.LEDGER_ID,hou.organization_id,gper.PERIOD_NAME,XAL.AE_HEADER_ID,XAL.AE_LINE_NUM,GJB.STATUS BATCH_STATUS,GJH.JE_CATEGORY,XAL.ACCOUNTING_CLASS_CODE,GJH.JE_SOURCE,GJH.JE_HEADER_ID,GJL.JE_LINE_NUM
FROM apps.gl_ledgers gl
join apps.hr_operating_units hou
on hou.set_of_books_id = gl.ledger_id
join APPS.GL_JE_LInes GJL
on exists (select null from gccs where gccs.CODE_COMBINATION_ID = GJL.CODE_COMBINATION_ID)
AND (nvl(gjl.stat_amount,0) <> 0 OR (NVL (gjl.accounted_dr,0) - NVL (gjl.accounted_cr,0)) <> 0)
join APPS.GL_PERIODS gper
on GJL.PERIOD_NAME = gper.PERIOD_NAME
join APPS.GL_JE_HEADERS GJH
on GJH.JE_HEADER_ID = GJL.JE_HEADER_ID
AND GJH.PERIOD_NAME = gper.PERIOD_NAME
AND GJH.LEDGER_ID = gl.ledger_id
AND GJH.JE_SOURCE = 'Cost Management'
join APPS.GL_JE_BATCHES GJB
on GJB.JE_BATCH_ID = GJH.JE_BATCH_ID
join APPS.GL_IMPORT_REFERENCES GIR
on GIR.JE_HEADER_ID = GJL.JE_HEADER_ID
AND GIR.JE_LINE_NUM = GJL.JE_LINE_NUM
join APPS.XLA_AE_LInes XAL
on XAL.GL_SL_LINK_TABLE = GIR.GL_SL_LINK_TABLE
AND XAL.GL_SL_LINK_ID = GIR.GL_SL_LINK_ID
AND XAL.LEDGER_ID = gl.LEDGER_ID
),receiving_1 as (
select distinct
bq.LEDGER_ID,bq.organization_id,bq.period_name,bq.AE_HEADER_ID,bq.AE_LINE_NUM
FROM base_query bq
WHERE bq.BATCH_STATUS = 'P'
AND UPPER(bq.JE_CATEGORY) = UPPER('Receiving')
),receiving_2 as (
select distinct
bq.LEDGER_ID,bq.AE_LINE_NUM
FROM base_query bq
WHERE bq.ACCOUNTING_CLASS_CODE = 'INTRA'
AND bq.BATCH_STATUS = 'P'
AND UPPER(bq.JE_CATEGORY) = UPPER('Receiving')
),receiving_all as (
-- the use of "rownum" below sometimes helps to optimize "union all",but does not help in this query
select rownum,x.* from receiving_1 x -- 58 recs in 3.4 secs cost 295 (no UNION)
union all -- 58 recs in 203.1 secs cost 300 (UNION ALL)
select rownum,x.* from receiving_2 x -- 0 recs in 3.8 secs cost 295 (no UNION)
)
select count(*)
from receiving_all
where ledger_id = 2021
and organization_id = 82
and period_name = 'SEP-2020';
用于接收_1 部分查询的 AUTOTRACE"
COUNT(*)
----------
58
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%cpu)| Pstart| Pstop |
-------------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 35 | 295 (1)| | |
| 1 | SORT AGGREGATE | | 1 | 35 | | | |
| 2 | VIEW | | 1 | 35 | 295 (1)| | |
| 3 | COUNT | | | | | | |
| 4 | VIEW | | 1 | 35 | 295 (1)| | |
| 5 | HASH UNIQUE | | 1 | 266 | 295 (1)| | |
| 6 | nesTED LOOPS | | 1 | 266 | 294 (1)| | |
| 7 | nesTED LOOPS | | 1 | 266 | 294 (1)| | |
| 8 | nesTED LOOPS | | 1 | 237 | 236 (1)| | |
| 9 | nesTED LOOPS | | 1 | 214 | 178 (1)| | |
| 10 | nesTED LOOPS | | 1 | 207 | 176 (1)| | |
| 11 | nesTED LOOPS | | 1 | 183 | 176 (1)| | |
| 12 | nesTED LOOPS | | 1 | 152 | 173 (1)| | |
| 13 | nesTED LOOPS | | 1 | 148 | 173 (1)| | |
| 14 | nesTED LOOPS | | 1 | 123 | 172 (1)| | |
| 15 | nesTED LOOPS | | 1 | 101 | 169 (1)| | |
| 16 | nesTED LOOPS | | 1 | 97 | 169 (1)| | |
| 17 | nesTED LOOPS | | 1 | 53 | 167 (1)| | |
| 18 | SORT UNIQUE | | 1 | 21 | 22 (0)| | |
| 19 | TABLE ACCESS BY INDEX ROWID| GL_CODE_COMBINATIONS | 1 | 21 | 22 (0)| | |
| 20 | INDEX RANGE SCAN | GL_CODE_COMBINATIONS_N1 | 117 | | 8 (0)| | |
| 21 | TABLE ACCESS BY INDEX ROWID | GL_JE_LInes | 16 | 512 | 144 (0)| | |
| 22 | INDEX RANGE SCAN | GL_JE_LInes_N1 | 319 | | 4 (0)| | |
| 23 | TABLE ACCESS BY INDEX ROWID | GL_JE_HEADERS | 1 | 44 | 2 (0)| | |
| 24 | INDEX UNIQUE SCAN | GL_JE_HEADERS_U1 | 1 | | 1 (0)| | |
| 25 | INDEX UNIQUE SCAN | GL_LEDGERS_U2 | 1 | 4 | 0 (0)| | |
| 26 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 22 | 3 (0)| | |
| 27 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK1 | 30 | | 1 (0)| | |
| 28 | INDEX RANGE SCAN | GL_PERIODS_U1 | 1 | 25 | 1 (0)| | |
| 29 | INDEX UNIQUE SCAN | HR_ORGANIZATION_UNITS_PK | 1 | 4 | 0 (0)| | |
| 30 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 31 | 3 (0)| | |
| 31 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK2 | 2 | | 1 (0)| | |
| 32 | INDEX UNIQUE SCAN | HR_ALL_ORGANIZATION_UNTS_TL_PK | 1 | 24 | 0 (0)| | |
| 33 | TABLE ACCESS BY INDEX ROWID | GL_JE_BATCHES | 1 | 7 | 2 (0)| | |
| 34 | INDEX UNIQUE SCAN | GL_JE_BATCHES_U1 | 1 | | 1 (0)| | |
| 35 | TABLE ACCESS BY INDEX ROWID | GL_IMPORT_REFERENCES | 105 | 2415 | 58 (0)| | |
| 36 | INDEX RANGE SCAN | GL_IMPORT_REFERENCES_N1 | 105 | | 3 (0)| | |
| 37 | PARTITION LIST ALL | | 1 | | 57 (0)| 1 | 19 |
| 38 | INDEX RANGE SCAN | XLA_AE_LInes_N4 | 1 | | 57 (0)| 1 | 19 |
| 39 | TABLE ACCESS BY LOCAL INDEX ROWID | XLA_AE_LInes | 1 | 29 | 58 (0)| 1 | 1 |
-------------------------------------------------------------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
Statistics
----------------------------------------------------------
99 recursive calls
0 db block gets
3500320 consistent gets
0 physical reads
116 redo size
346 bytes sent via sql*Net to client
4388 bytes received via sql*Net from client
2 sql*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
用于接收_2 部分查询的 AUTOTRACE：
COUNT(*)
----------
0
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%cpu)| Pstart| Pstop |
-------------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 35 | 295 (1)| | |
| 1 | SORT AGGREGATE | | 1 | 35 | | | |
| 2 | VIEW | | 1 | 35 | 295 (1)| | |
| 3 | COUNT | | | | | | |
| 4 | VIEW | | 1 | 35 | 295 (1)| | |
| 5 | HASH UNIQUE | | 1 | 279 | 295 (1)| | |
| 6 | nesTED LOOPS | | 1 | 279 | 294 (1)| | |
| 7 | nesTED LOOPS | | 1 | 279 | 294 (1)| | |
| 8 | nesTED LOOPS | | 1 | 272 | 292 (1)| | |
| 9 | nesTED LOOPS | | 1 | 230 | 234 (1)| | |
| 10 | nesTED LOOPS | | 1 | 207 | 176 (1)| | |
| 11 | nesTED LOOPS | | 1 | 183 | 176 (1)| | |
| 12 | nesTED LOOPS | | 1 | 152 | 173 (1)| | |
| 13 | nesTED LOOPS | | 1 | 148 | 173 (1)| | |
| 14 | nesTED LOOPS | | 1 | 126 | 170 (1)| | |
| 15 | nesTED LOOPS | | 1 | 122 | 170 (1)| | |
| 16 | nesTED LOOPS | | 1 | 97 | 169 (1)| | |
| 17 | nesTED LOOPS | | 1 | 53 | 167 (1)| | |
| 18 | SORT UNIQUE | | 1 | 21 | 22 (0)| | |
| 19 | TABLE ACCESS BY INDEX ROWID| GL_CODE_COMBINATIONS | 1 | 21 | 22 (0)| | |
| 20 | INDEX RANGE SCAN | GL_CODE_COMBINATIONS_N1 | 117 | | 8 (0)| | |
| 21 | TABLE ACCESS BY INDEX ROWID | GL_JE_LInes | 16 | 512 | 144 (0)| | |
| 22 | INDEX RANGE SCAN | GL_JE_LInes_N1 | 319 | | 4 (0)| | |
| 23 | TABLE ACCESS BY INDEX ROWID | GL_JE_HEADERS | 1 | 44 | 2 (0)| | |
| 24 | INDEX UNIQUE SCAN | GL_JE_HEADERS_U1 | 1 | | 1 (0)| | |
| 25 | INDEX RANGE SCAN | GL_PERIODS_U1 | 1 | 25 | 1 (0)| | |
| 26 | INDEX UNIQUE SCAN | GL_LEDGERS_U2 | 1 | 4 | 0 (0)| | |
| 27 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 22 | 3 (0)| | |
| 28 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK1 | 30 | | 1 (0)| | |
| 29 | INDEX UNIQUE SCAN | HR_ORGANIZATION_UNITS_PK | 1 | 4 | 0 (0)| | |
| 30 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 31 | 3 (0)| | |
| 31 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK2 | 2 | | 1 (0)| | |
| 32 | INDEX UNIQUE SCAN | HR_ALL_ORGANIZATION_UNTS_TL_PK | 1 | 24 | 0 (0)| | |
| 33 | TABLE ACCESS BY INDEX ROWID | GL_IMPORT_REFERENCES | 105 | 2415 | 58 (0)| | |
| 34 | INDEX RANGE SCAN | GL_IMPORT_REFERENCES_N1 | 105 | | 3 (0)| | |
| 35 | PARTITION LIST ALL | | 1 | 42 | 58 (0)| 1 | 19 |
| 36 | TABLE ACCESS BY LOCAL INDEX ROWID | XLA_AE_LInes | 1 | 42 | 58 (0)| 1 | 19 |
| 37 | INDEX RANGE SCAN | XLA_AE_LInes_N4 | 1 | | 57 (0)| 1 | 19 |
| 38 | INDEX UNIQUE SCAN | GL_JE_BATCHES_U1 | 1 | | 1 (0)| | |
| 39 | TABLE ACCESS BY INDEX ROWID | GL_JE_BATCHES | 1 | 7 | 2 (0)| | |
-------------------------------------------------------------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
Statistics
----------------------------------------------------------
99 recursive calls
0 db block gets
3451912 consistent gets
0 physical reads
0 redo size
345 bytes sent via sql*Net to client
4385 bytes received via sql*Net from client
2 sql*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
AUTOTRACE for UNION ALL 包含 request_1 和 request_2 的完整查询
COUNT(*)
----------
58
Execution Plan
----------------------------------------------------------
----------------------------------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%cpu)| Pstart| Pstop |
----------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 35 | 300 (1)| | |
| 1 | TEMP TABLE TRANSFORMATION | | | | | | |
| 2 | LOAD AS SELECT | SYS_TEMP_0FD9D6767_FC3F7B00 | | | | | |
| 3 | nesTED LOOPS | | 1 | 279 | 294 (1)| | |
| 4 | nesTED LOOPS | | 1 | 279 | 294 (1)| | |
| 5 | nesTED LOOPS | | 1 | 237 | 236 (1)| | |
| 6 | nesTED LOOPS | | 1 | 214 | 178 (1)| | |
| 7 | nesTED LOOPS | | 1 | 207 | 176 (1)| | |
| 8 | nesTED LOOPS | | 1 | 183 | 176 (1)| | |
| 9 | nesTED LOOPS | | 1 | 152 | 173 (1)| | |
| 10 | nesTED LOOPS | | 1 | 148 | 173 (1)| | |
| 11 | nesTED LOOPS | | 1 | 126 | 170 (1)| | |
| 12 | nesTED LOOPS | | 1 | 122 | 170 (1)| | |
| 13 | nesTED LOOPS | | 1 | 78 | 168 (1)| | |
| 14 | nesTED LOOPS | | 1 | 53 | 167 (1)| | |
| 15 | SORT UNIQUE | | 1 | 21 | 22 (0)| | |
| 16 | TABLE ACCESS BY INDEX ROWID| GL_CODE_COMBINATIONS | 1 | 21 | 22 (0)| | |
| 17 | INDEX RANGE SCAN | GL_CODE_COMBINATIONS_N1 | 117 | | 8 (0)| | |
| 18 | TABLE ACCESS BY INDEX ROWID | GL_JE_LInes | 16 | 512 | 144 (0)| | |
| 19 | INDEX RANGE SCAN | GL_JE_LInes_N1 | 319 | | 4 (0)| | |
| 20 | INDEX RANGE SCAN | GL_PERIODS_U1 | 1 | 25 | 1 (0)| | |
| 21 | TABLE ACCESS BY INDEX ROWID | GL_JE_HEADERS | 1 | 44 | 2 (0)| | |
| 22 | INDEX UNIQUE SCAN | GL_JE_HEADERS_U1 | 1 | | 1 (0)| | |
| 23 | INDEX UNIQUE SCAN | GL_LEDGERS_U2 | 1 | 4 | 0 (0)| | |
| 24 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 22 | 3 (0)| | |
| 25 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK1 | 30 | | 1 (0)| | |
| 26 | INDEX UNIQUE SCAN | HR_ORGANIZATION_UNITS_PK | 1 | 4 | 0 (0)| | |
| 27 | TABLE ACCESS BY INDEX ROWID | HR_ORGANIZATION_informatION | 1 | 31 | 3 (0)| | |
| 28 | INDEX RANGE SCAN | HR_ORGANIZATION_informatIO_FK2 | 2 | | 1 (0)| | |
| 29 | INDEX UNIQUE SCAN | HR_ALL_ORGANIZATION_UNTS_TL_PK | 1 | 24 | 0 (0)| | |
| 30 | TABLE ACCESS BY INDEX ROWID | GL_JE_BATCHES | 1 | 7 | 2 (0)| | |
| 31 | INDEX UNIQUE SCAN | GL_JE_BATCHES_U1 | 1 | | 1 (0)| | |
| 32 | TABLE ACCESS BY INDEX ROWID | GL_IMPORT_REFERENCES | 105 | 2415 | 58 (0)| | |
| 33 | INDEX RANGE SCAN | GL_IMPORT_REFERENCES_N1 | 105 | | 3 (0)| | |
| 34 | PARTITION LIST ALL | | 1 | | 57 (0)| 1 | 19 |
| 35 | INDEX RANGE SCAN | XLA_AE_LInes_N4 | 1 | | 57 (0)| 1 | 19 |
| 36 | TABLE ACCESS BY LOCAL INDEX ROWID | XLA_AE_LInes | 1 | 42 | 58 (0)| 1 | 1 |
| 37 | SORT AGGREGATE | | 1 | 35 | | | |
| 38 | VIEW | | 2 | 70 | 6 (34)| | |
| 39 | UNION-ALL | | | | | | |
| 40 | COUNT | | | | | | |
| 41 | VIEW | | 1 | 35 | 3 (34)| | |
| 42 | HASH UNIQUE | | 1 | 44 | 3 (34)| | |
| 43 | VIEW | | 1 | 44 | 2 (0)| | |
| 44 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6767_FC3F7B00 | 1 | 74 | 2 (0)| | |
| 45 | COUNT | | | | | | |
| 46 | VIEW | | 1 | 35 | 3 (34)| | |
| 47 | HASH UNIQUE | | 1 | 61 | 3 (34)| | |
| 48 | VIEW | | 1 | 61 | 2 (0)| | |
| 49 | TABLE ACCESS FULL | SYS_TEMP_0FD9D6767_FC3F7B00 | 1 | 74 | 2 (0)| | |
----------------------------------------------------------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
Statistics
----------------------------------------------------------
766 recursive calls
83076 db block gets
270487689 consistent gets
81660 physical reads
676 redo size
346 bytes sent via sql*Net to client
4582 bytes received via sql*Net from client
2 sql*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
解决方法
由于我不知道底层表、它们的结构和它们的数据，我只能猜测。话虽如此，我会尝试以下操作（不知道这些步骤是否有帮助！）：
-
从
中删除 rownumreceiving_all -
将主查询中的
where子句添加到receiving_all中的每个子查询。优化器可能会推断出这一点，但我不确定。 -
将
union all中的union更改为receiving_all，并从distinct和receiving_1中删除receiving_2 -
receiving_2似乎包含在receiving_2中？ -
如果 4. 不是这种情况，我会将两者合并为一个带有
的查询WHERE UPPER(bq.JE_CATEGORY) = UPPER('Receiving') AND ( (<conditions for 1>) OR (<conditions for 2>) ) -
如果可以避免
UPPER(bq.JE_CATEGORY)，我会尝试。函数阻止索引访问。 -
如果您可以创建表，我会将
gccs的值展开到两个表中，一个用于segment_1，另一个用于segment_2。 -
如果 7. 不是一个选项，我会将提示
select /*+ materialize */ gcc.code_combination_id添加到gcc。
我很想知道哪个建议有任何积极影响...
版权声明：本文内容由互联网用户自发贡献，该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容， 请发送邮件至 dio@foxmail.com 举报，一经查实，本站将立刻删除。
