Problems with enclosing in ticks.

Question

Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production
PL/SQL Release 11.2.0.3.0 - Production
CORE     11.2.0.3.0     Production
TNS for Solaris: Version 11.2.0.3.0 - Production
NLSRTL Version 11.2.0.3.0 - Production
I got help recently using tick marks.
It's complaining about the YES in the constraint towards the bottom.
Previously, I was just using execute immediate, but was corrected to use the Qliteral...as seen here, to correct for issues with tick marks.
I guess I still don't understand what I'm doing wrong.
set serveroutput on;
declare
  v_count     number;
BEGIN
  select count(*)
   into v_count
   from all_tables
   where table_name = 'NRV_ADMIN_REGIONS_WILLY'
   and owner = 'FS_NRIS_FSVEG';
   IF (v_count >= 1 )
    then dbms_output.put_line('TABLE EXISTS'); 
    else
execute immediate   --Qliteral begins here
Q'{CREATE TABLE FS_NRIS_FSVEG.NRV_ADMIN_REGIONS_WILLY
  CN           VARCHAR2(32 BYTE) CONSTRAINT NRV_ADMIN_REG_STAT_CK NOT NULL,
  REGION_ID    VARCHAR2(2 BYTE)                 NOT NULL,
  REGION_NAME  VARCHAR2(50 BYTE)                NOT NULL,
  STATUS       VARCHAR2(3 BYTE)
TABLESPACE USERS_NRIS_FSVEG
RESULT_CACHE (MODE DEFAULT)
PCTUSED    0
PCTFREE    10
INITRANS   1
MAXTRANS   255
STORAGE    (
            INITIAL          80K
            NEXT             1M
            MINEXTENTS       1
            MAXEXTENTS       UNLIMITED
            PCTINCREASE      0
            BUFFER_POOL      DEFAULT
            FLASH_CACHE      DEFAULT
            CELL_FLASH_CACHE DEFAULT
LOGGING
NOCOMPRESS
NOCACHE
NOPARALLEL
MONITORING;
CREATE UNIQUE INDEX FS_NRIS_FSVEG.NRV_ADMIN_REG_PK ON FS_NRIS_FSVEG.NRV_ADMIN_REGIONS_WILLY
(CN)
LOGGING
TABLESPACE USERS_NRIS_FSVEG
PCTFREE    10
INITRANS   2
MAXTRANS   255
STORAGE    (
            INITIAL          80K
            NEXT             1M
            MINEXTENTS       1
            MAXEXTENTS       UNLIMITED
            PCTINCREASE      0
            BUFFER_POOL      DEFAULT
            FLASH_CACHE      DEFAULT
            CELL_FLASH_CACHE DEFAULT
NOPARALLEL;
CREATE OR REPLACE TRIGGER FS_NRIS_FSVEG.NRV_ADMIN_REGION_INS
BEFORE INSERT
ON FS_NRIS_FSVEG.NRV_ADMIN_REGIONS_WILLY
REFERENCING OLD AS OLD NEW AS NEW
FOR EACH ROW
DECLARE
BEGIN
   if :new.cn is null
   then
      :new.cn := SYS_GUID();
   end if;
END;
ALTER TABLE FS_NRIS_FSVEG.NRV_ADMIN_REGIONS_WILLY ADD (
  CONSTRAINT NRV_ADMIN_REGION_STATUS
  CHECK (STATUS IN  ('YES','NO'))
  ENABLE VALIDATE,
  CONSTRAINT NRV_ADMIN_REG_PK
  PRIMARY KEY
  (CN)
  USING INDEX FS_NRIS_FSVEG.NRV_ADMIN_REG_PK
  ENABLE VALIDATE);
GRANT DELETE, INSERT, SELECT, UPDATE ON FS_NRIS_FSVEG.NRV_ADMIN_REGIONS_WILLY TO NRV_DATA_COLLECTOR}';   --Qliteral ends here
end if;
EXCEPTION WHEN OTHERS THEN dbms_output.put_line(sqlerrm);
END;
/

Answer 1

SaadL wrote:
for Yes and No you need to enter single quote twice
''YES'', ''NO''Tried that...failed. As well as even doing actual quotes...failed.
kendenny wrote:
I don't see a }' at the end of your quoted string. You need that.It's down there... look to the far left, it's annotated too.
Justin Cave wrote:
First, you would need a separate call to EXECUTE IMMEDIATE for each DDL statement. You'd need one for the CREATE TABLE statement, one for the CREATE INDEX statement, etc. You can't just combine a bunch of DDL statements together like this.
Second, what, exactly, does "it's complaining" mean? I'm guessing that you are getting some sort of error. What error are you getting? What is the error stack? Are you getting a compilation error or a runtime error? Are you getting an error from the database? Or from your front end?
JustinYou are right...my apologies....I know better than to not include details.
The only error I'm getting, with no details, is ORA-00911: invalid character. But it doesn't indicate any line or anything.
And I didn't realize that I could only put one DDL per execution. I will try that and report.
3360 wrote:
odie_63 wrote:
First thing that comes to mind : why dynamic code? You're getting yourself problems where there shouldn't be any.This has already been discussed in the OP's other thread, e.g.
{message:id=10694055}
This seems to be an attempt to meet this requirement
My code has to be able to be run multiple times without any kind of failure. It can ONLY write the error to the log.In some weird upgrade process that seems both unusual and highly unreliable.When I have the scripts working correctly, it actually works really well. Here's why... We have multiple companies that all reside on one large oracle instance that is managed by a team of DBA's. My job as a development DBA is to process the code that comes in from the developers, and deal with them at our level. when it's time to update our application, I take all the code that they have turned in to me, along with my own adhoc requests like this, and I make one huge script. That script has to be able to be rerun. The biggest caviet is create table scripts. Because if our dev time is 8 weeks, we've built the table and have had 8 weeks of inserts and play time in the dev environment...so to just drop and recreate the table is not possible due to data loss. So my scripts have to be able to say, build the table...but if it exists, move on with no action. Views, procedures and other objects have no weight on them, as they don't store data.
Anyway...My packaged up script will call multiple objects. meanwhile, the dba team is getting the same thing from many teams at my level from the other companies. They have to be able to have their own script, on the night of deployment, fire off all of our scripts without any issues. So they spool the scripts to utilize their own respective log files, and upon completion ship those out to us. It makes it flow really well actually. The problem for me is that I'm not used to writing pl/sql code. Once I get these quarks figured out, it will be smooth sailing.
Great information. Thank you all.