coComments related to laurentschneider

I am amazed by your solution.

2009-03-26T20:29:29.634+01:00

I am amazed by your solution. I did read about it in your book, but had no clue it was working in 11.2. I would definitely vote for yours :)

Maybe XML format or otherwise

2009-03-25T07:28:59.673+01:00

Maybe XML format or otherwise InterMedia

@carri

2009-01-19T20:02:00.560+01:00

@carrie b yes, username=password and it is hardcoded everywhere :(

really? could you copy paste y

2009-01-18T18:34:39.188+01:00

really? could you copy paste your input/output and provide version?

the OPATCH options are identic

2008-07-16T21:59:02.862+02:00

the OPATCH options are identical. I cannot confirm if the dummy patch exists on Windows.

yes, quite special :) SQL>

2008-07-16T07:12:38.701+02:00

yes, quite special :) SQL> with x$dual as (select '1' from dual) select * from x$dual; ORA-32031: illegal reference of a query name in WITH clause

Fun, but it works only in one

2008-07-15T11:51:14.526+02:00

Fun, but it works only in one direction :mrgreen: with dual as (select 'x' from v$instance) select * from dual; with dual as (select 'x' from v$instance) select * from dual * ERROR at line 1: ORA-32035: unreferenced query name defined in WITH clause

Well, I cannot test it it as I

2007-12-19T21:24:21.313+01:00

Well, I cannot test it it as I have full privilege right now, but as I remember print was not working that well... Thanks for the tip, I keep it in case I have to use a castrated version of toad again

SQL> set mark html on table

2007-11-08T03:16:19.178+01:00

SQL> set mark html on table "><img src=/images/laurentschneider.png></table><br><table" SQL> select * from dual; D X

I will have a look

2007-08-21T20:04:27.300+02:00

I will have a look

Firstly, I find that it is mor

2007-03-23T16:47:49.770+01:00

Firstly, I find that it is more difficult to understand what the code is supposed to do, and it is much more easy (for most of us) to put an error in the connect by as by reusing YOUR stragg function. Secondly, I think connect by will not scale well A little test (well, it is just a test, it is not supposed to prove anything) set timi on lin 9999 trims on newp none pages 0 termout off spool /tmp/stragg.txt select trunc(object_id/100),stragg(object_name) from all_objects group by trunc(object_id/100) -- having stragg(object_name)='XYZ'; spool off spool /tmp/connect.txt select o, max(sys_connect_by_path(object_name, ' ' )) scbp from (select trunc(object_id/100) o, object_name, row_number() over (partition by trunc(object_id/100) order by object_name) rn from all_objects ) start with rn = 1 connect by prior rn = rn-1 and prior o = o group by o -- having max(sys_connect_by_path(ename, ' ' ))='XYZ' / spool off the first query finished in 5 secs. After 3 min, the second one ended with ORA-30004: when using SYS_CONNECT_BY_PATH function, cannot have seperator as part of column value (why?) If I replace all_objects by user_objects the first try takes 0.03 sec and the second one takes 0.35. So when used with such a complex view as ALL_OBJECTS, connect by artifice is not very attractive to me.

why? I like your stragg

2007-03-23T16:13:40.124+01:00

why? I like your stragg