Details
-
Type:
Task
-
Status: Open
-
Priority:
Minor
-
Resolution: Unresolved
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
Description
Implement some new functions (maybe UDF),
1)return a JSON object with token tree of SQL parser, example
SELECT SQL_TOKEN_TREE(
"SELECT * FROM (SELECT 1 AS a) AS b WHERE a=@var"
)
Result
{Query:[select, *, from, [select, var<1>, as, const<a>], as, const<b>, where, var<2>, =, var<3>], vars:{1:'1', 2:'a',3:'@var','@var':'qwert'}}
Or something more intelligent since i don't know how mysql parser works
2)SQL_TOKEN_SQL(sql tree)
Result a string removing consts with a willchar (like prepared statments)
SELECT * FROM TABLE WHERE A=1234
should return
SELECT * FROM TABLE WHERE A=?
this is something like percona tools, but it run inside server with server parse and not a third part parser:
$pt-fingerprint --query "/* -- S++ SU ABORTABLE -- spd_user: rspadim */SELECT SQL_SMALL_RESULT SQL_CACHE DISTINCT centro_atividade FROM est_dia WHERE unidade_id=1001 AND item_id=67 AND item_id_red=573"
SELECT DISTINCT centro_atividade FROM est_dia WHERE unidade_id=? AND item_id=? AND item_id_red=?
3)Add a SQL_PARSER_OPTIMIZE(QUERY)
it parse the query and execute optimizations (opt_range and others)
example
SELECT * FROM TABLE WHERE A=1234 AND 0=0
rewrite to
SELECT * FROM TABLE WHERE A=1234
Gliffy Diagrams
Attachments
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
