Обсуждение: pl/perl function life and variable scope - concurrency problem?

 Greetings,

 Thanks to a lot of help on this list, I've managed to get my pl/perl
function working. However, I have an unexpected result. Here's a simple
way to reproduce this problem:

CREATE or REPLACE FUNCTION perltest(integer)
returns integer as '
  $MyInt = $MyInt + 1;
  return $MyInt;
' language plperlu;


Executing it produces:

chris=# select perltest(1);
 perltest
----------
        1
(1 row)

chris=# select perltest(1);
 perltest
----------
        2
(1 row)

chris=# select perltest(1);
 perltest
----------
        3
(1 row)

chris=# select perltest(1);
 perltest
----------
        4
(1 row)

 Doing the right thing and initialzing variables before using them
solves this. Should I worry about this? Are concurrent callings of this
function protected from each other?

 Any info would be much appreciated.

Cheers,

Chris


--
Christopher Murtagh
Enterprise Systems Administrator
ISR / Web Communications Group
McGill University
Montreal, Quebec
Canada

Tel.: (514) 398-3122
Fax:  (514) 398-2017

On Wed, Nov 12, 2003 at 03:05:34PM -0500, Christopher Murtagh wrote:

> CREATE or REPLACE FUNCTION perltest(integer)
> returns integer as '
>   $MyInt = $MyInt + 1;
>   return $MyInt;
> ' language plperlu;

Use a local variable:

 CREATE or REPLACE FUNCTION perltest(integer)
 returns integer as '
   my $MyInt = $MyInt + 1;
   return $MyInt;
 ' language plperlu;

--
Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
"The problem with the future is that it keeps turning into the present"
(Hobbes)

Christopher Murtagh <christopher.murtagh@mcgill.ca> writes:

>  Thanks to a lot of help on this list, I've managed to get my pl/perl
> function working. However, I have an unexpected result. Here's a simple
> way to reproduce this problem:
>
> CREATE or REPLACE FUNCTION perltest(integer)
> returns integer as '
>   $MyInt = $MyInt + 1;
>   return $MyInt;
> ' language plperlu;

There's a reason Perl has "my" variables.  Use them.  ;)

-Doug

Or in other words... code to "strict"

J


Doug McNaught wrote:

>Christopher Murtagh <christopher.murtagh@mcgill.ca> writes:
>
>
>
>> Thanks to a lot of help on this list, I've managed to get my pl/perl
>>function working. However, I have an unexpected result. Here's a simple
>>way to reproduce this problem:
>>
>>CREATE or REPLACE FUNCTION perltest(integer)
>>returns integer as '
>>  $MyInt = $MyInt + 1;
>>  return $MyInt;
>>' language plperlu;
>>
>>
>
>There's a reason Perl has "my" variables.  Use them.  ;)
>
>-Doug
>
>---------------------------(end of broadcast)---------------------------
>TIP 9: the planner will ignore your desire to choose an index scan if your
>      joining column's datatypes do not match
>
>

--
Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
Postgresql support, programming shared hosting and dedicated hosting.
+1-503-222-2783 - jd@commandprompt.com - http://www.commandprompt.com
Editor-N-Chief - PostgreSQl.Org - http://www.postgresql.org

On Wed, 2003-11-12 at 15:44, Joshua D. Drake wrote:
> Or in other words... code to "strict"

 :-)

 Yes, I do normally. I was simply running a test and came across this
and it caught me by surprise. For the record, my pl/perl function is
this:

CREATE OR REPLACE FUNCTION htdig(text, text)
RETURNS text[] AS '

 my $SearchTerms = $_[0];
 my $HtDigDB = $_[1];
 my @Result = {};
 my $Line = '''';

 #open HTDIG, "/usr/local/htdig/bin/htsearch  ''config=" . $HtDigDB . ";words=" . $SearchTerms .
"'';matchesperpage=1000;";
 open HTDIG, "/bin/cat /home/postgres/" . $SearchTerms . " |";

 while(<HTDIG>) {

    $Line = $_;
    $Line =~ s/^h[^0-9]*//;
    chomp($Line);
    push @Result, $Line;
 }

 close HTDIG;

 return qq/{/ . (join qq/,/, @Result) . qq/}/;
' LANGUAGE plperlu;


I started writing this in C and realized that this was going to be a
couple hundred lines of code - compared to the 23 lines in Perl,
including test lines. All this function needs to do is to take a pipe
from the htsearch and grab integers in the URL that it returns. Then
I'll write a pl/pgSQL wrapper that returns this array as a set.

Pretty sweet, and I couldn't have done it without the help of this list.

Thanks!

Cheers,

Chris

--
Christopher Murtagh
Enterprise Systems Administrator
ISR / Web Communications Group
McGill University
Montreal, Quebec
Canada

Tel.: (514) 398-3122
Fax:  (514) 398-2017