RE: PL/SQL Interpreter oddity - bug or "expected"?

From: Matthew Parker <dimensional.dba_at_comcast.net>
Date: Sat, 17 Mar 2018 01:44:21 -0700
Message-ID: <010101d3bdcc$25929da0$70b7d8e0$_at_comcast.net>



From a Oracle SQL Reserved Word prospective you might think so but when you go to the documentation on Reserved Words, “loop” is not listed.  

https://docs.oracle.com/database/121/SQLRF/ap_keywd001.htm#SQLRF55621  

Also in testing words it appears to error on the listed Reserved Words, but some other words that are also fundamental to SQL such as “merge” do not error either and are not in the key word list.    

Matthew Parker

Chief Technologist

Dimensional DBA

Oracle Gold Partner

425-891-7934 (cell)

D&B 047931344

CAGE 7J5S7 Dimensional.dba_at_comcast.net <mailto:Dimensional.dba_at_comcast.net>

 <http://www.linkedin.com/pub/matthew-parker/6/51b/944/> View Matthew Parker's profile on LinkedIn

www.dimensionaldba.com <http://www.dimensionaldba.com/>    

From: oracle-l-bounce_at_freelists.org <oracle-l-bounce_at_freelists.org> On Behalf Of Stefan Knecht Sent: Friday, March 16, 2018 10:14 PM
To: oracle-l-freelists <oracle-l_at_freelists.org> Subject: PL/SQL Interpreter oddity - bug or "expected"?  

Ran into this recently, and personally I feel that this should not compile. What do you guys think?  

Simple case in point:  

create or replace package foo as

procedure bar;

end;

/
 

Package created.    

create or replace package body foo as

procedure bar

is

begin

  begin

    null;

  end loop;

end;

end;

/
 

Package body created.  

show errors  

No errors.  

SQL> exec foo.bar  

PL/SQL procedure successfully completed.    

What is happening is that Oracle interprets the "end loop" as being "end anonymous_block_name". And in my opinion, with a keyword like "loop" that should not be happening. In essence, the above is interpreted as:  

create or replace package body foo as

procedure bar

is

begin

  begin

    null;

  end anon_block_name;

end;

end;

/
   

If we switch the "end loop" to an "end if" (which is a similar flow control keyword) it reacts (in my opinion) correctly:  

create or replace package body foo as

procedure bar

is

begin

  begin

    null;

  end if;

end;

end;

/
 

Warning: Package Body created with compilation errors.  

SQL> show errors

Errors for PACKAGE BODY FOO:  

LINE/COL ERROR

  • -----------------------------------------------------------------

7/7 PLS-00103: Encountered the symbol "IF" when expecting one of the

         following:

         ; <an identifier> <a double-quoted delimited-identifier>

         The symbol "IF" was ignored.  

SQL>     Would you agree that that should not compile in the first place?      

Stefan          

-- 


//
zztat - The Next-Gen Oracle Performance Monitoring and Reaction Framework! Visit us at zztat.net <http://zztat.net> | Support our Indiegogo campaign at igg.me/at/zztat <http://igg.me/at/zztat> | _at_zztat_oracle -- http://www.freelists.org/webpage/oracle-l
Received on Sat Mar 17 2018 - 09:44:21 CET

Original text of this message