DBD::DBM and hashrefs

Hello,

I have a question about using DBD::DBM in MLDBM mode: does anyone know
whether it works with hashrefs as values, or just arrayrefs?

Reading the docs and trying the module leads me to believe it only
works with data like:

$Tied_hash{ $id } =3D [ $val1, $val2 ];

But all the data in my DB file looks like:

$Tied_hash{ $id } =3D { key1 =3D> $val1, key2 =3D> $val2 };

.... and I'd like to use DBD::DBM to access it (i.e. "SELECT key1, key2
from foo WHERE id=3D?") but as far as I can tell this is not possible.

Am I right?

Thanks for any help

RT

Re: DBD::DBM and hashrefs

On Thu, 13 Oct 2005 10:11:41 -0700, Ryan Tate wrote:

Hi Ryan

> $Tied_hash{ $id } =3D { key1 =3D> $val1, key2 =3D> $val2 };

But a hash is just an array of key/value pairs. Can't you just try:

$T{$id} =3D [\%hash];

What happens?
--
Cheers
Ron Savage, ron@savage.net.au on 14/10/2005
http://savage.net.au/index.html
Let the record show: Microsoft is not an Australian company

Re: DBD::DBM and hashrefs

> -----Original Message-----
> From: Ryan Tate [mailto:lists@ryantate.com]
> Subject: DBD::DBM and hashrefs
>
> Hello,
>
> I have a question about using DBD::DBM in MLDBM mode: does anyone know
> whether it works with hashrefs as values, or just arrayrefs?
>
> Reading the docs and trying the module leads me to believe it only
> works with data like:
>
> $Tied_hash{ $id } = [ $val1, $val2 ];
>
> But all the data in my DB file looks like:
>
> $Tied_hash{ $id } = { key1 => $val1, key2 => $val2 };

I believe you can specify hash storage by passing the appropriate DBM flags when you do DBI->connect(). Use the flags your DBM implementation expects and pass them as specified in the DBD::DBM docs.

--
Jeff

Re: DBD::DBM and hashrefs

On Thu, 13 Oct 2005 17:22:07 -0700, Ron Savage wrote:

>Can't you just try:
>
>$T{$id} =3D [\%hash];

The tied hash already exists in the other format and is used by
multiple scripts, so changing the format is not something I want to
do.

The only reason I'm even using DBD::DBM is I want to keep using the
DBM file as I have it, if I were ready to change the whole data
storage format I would just go to a proper relational database
already.

Thanks nevertheless.

RT

Re: DBD::DBM and hashrefs

On Fri, 14 Oct 2005 11:41:45 -0700, jeff wrote:

>I believe you can specify hash storage by passing the appropriate DBM flag=
s
>when you do DBI->connect(). Use the flags your DBM implementation expects =
and
>pass them as specified in the DBD::DBM docs.

I have been over the docs several times, as well as the source code,
and am knee-deep in writing a patch (on the offhand suggestion of Tim
Bunce on the Class::DBI list).

If DBD::DBM supports hashrefs as values the same way it supports
arrayrefs as values, I am _really_ missing something.

Needless to say, I saw no such flags. If they exist I would love to
see them, so I can chuck this patch.

RT

Re: DBD::DBM and hashrefs

I've blabbered about this enough, here's some example code I ran just now.

The script:

----------start----------
use strict;
use warnings;

use DBI;
use Data::Dumper;

#Test of DBD::DBM with MLDBM
#MLDBM file format is $Dbm{ $id } =3D {key =3D> $val, key2 =3D> $val2 };

my $dbh =3D DBI->connect('dbi:DBM:mldbm=3DStorable;type=3DDB_File;f_dir= 3D/=
home/sappy/public_html');
$dbh->{RaiseError} =3D 1;
$dbh->{dbm_store_metadata} =3D 0;

#Except for 'id', each column in dbm_cols corresponds
#to keys from the hashrefs in the MLDBM file
$dbh->{dbm_cols} =3D 'id,path,publication_name,regex_headlines';


my $statement =3D $dbh->prepare('SELECT
id,path,publication_name,regex_headlines FROM sites WHERE id=3D?');
$statement->execute('id_services-http-users-r-ryantate-cgi-b in-se_ho_admin-=
23362-1058765796-14\
70540404-3232250416');
my @row =3D $statement->fetchrow_array;
print Dumper(@row);

----------end----------

What I expected as output was an array of four defined, simple scalar
values, corresponding to $id ("id_services ..."), $Dbm{$id}{path}
("/index.html"), $Dbm{$id}{publication_name} ("Dude's Site"), etc.

What I got as output from this test script was $id, a hashref
comprising the entirety of $Dbm{$id}, and two undefined values. More
precisely:

----------start----------
$VAR1 =3D 'id_services-http-users-r-ryantate-cgi-bin-se_ho_admin-23362 -1058=
765796-1470540404-323\
2250416';
$VAR2 =3D {
'match_join_exclude_target' =3D> '',
'match_num_year' =3D> [
'3'
],
'search_target_exact_phrase_atom_handling' =3D> 'Match whole exact phrase=
',
---snip---
'path' =3D> '/index.html',
---snip---
'publication_name' =3D> 'Dude's Site',
---snip---
'regex_headlines' =3D> '

[A-Za-z]+?, ([A-Za-z]+?) (\\d\\d?),
(\\d\\d\\d\\d)