From tleffler at uw.edu Fri Nov 3 19:55:47 2017 From: tleffler at uw.edu (Trevor Leffler) Date: Fri, 3 Nov 2017 19:55:47 +0000 Subject: [Dbix-class] How to deal with right joins while under test/sqlite Message-ID: Up until recently I've been exclusively using a mysql database. I've started writing tests using Test::DBIx::Class, which loads my schema into an in-memory sqlite database. I got it working after making a few minor code changes to my Result/ResultSet classes. For example, I had to replace the use of the mysql-specific utc_timestamp() function in a couple queries with a DateTime string. Now I'm getting a "RIGHT and FULL OUTER JOINs are not currently supported" error while running under test (sqlite). A Result class has a has_many relationship with a {join_type => 'right'}. I'm curious what are folks doing here. Are you... avoiding right join relationships and using alternate SQL to achieve the same queries? Doing db feature detection to "do the right thing"? Not using sqlite for unit testing? Thanks, --Trevor -------------- next part -------------- An HTML attachment was scrubbed... URL: From tleffler at uw.edu Tue Nov 7 19:28:30 2017 From: tleffler at uw.edu (Trevor Leffler) Date: Tue, 7 Nov 2017 19:28:30 +0000 Subject: [Dbix-class] How to deal with right joins while under test/sqlite In-Reply-To: References: Message-ID: Well... nevermind this. A little searching reminding me that I can replace right join queries with left join queries (and vice versa). I've replaced my handful of right join queries and have things working. Cheers, --Trevor From: dbix-class-bounces at lists.scsys.co.uk [mailto:dbix-class-bounces at lists.scsys.co.uk] On Behalf Of Trevor Leffler Sent: Friday, November 3, 2017 12:56 PM To: dbix-class at lists.scsys.co.uk Subject: [Dbix-class] How to deal with right joins while under test/sqlite Up until recently I've been exclusively using a mysql database. I've started writing tests using Test::DBIx::Class, which loads my schema into an in-memory sqlite database. I got it working after making a few minor code changes to my Result/ResultSet classes. For example, I had to replace the use of the mysql-specific utc_timestamp() function in a couple queries with a DateTime string. Now I'm getting a "RIGHT and FULL OUTER JOINs are not currently supported" error while running under test (sqlite). A Result class has a has_many relationship with a {join_type => 'right'}. I'm curious what are folks doing here. Are you... avoiding right join relationships and using alternate SQL to achieve the same queries? Doing db feature detection to "do the right thing"? Not using sqlite for unit testing? Thanks, --Trevor -------------- next part -------------- An HTML attachment was scrubbed... URL: From jgrg at sanger.ac.uk Tue Nov 7 19:35:47 2017 From: jgrg at sanger.ac.uk (James Gilbert) Date: Tue, 7 Nov 2017 19:35:47 +0000 Subject: [Dbix-class] How to deal with right joins while under test/sqlite In-Reply-To: References: Message-ID: <881D1A78-5A36-4CBD-9E76-77AFE43160E3@sanger.ac.uk> Indeed. I've never used a right or full join. Only ever an inner or left. James > On 7 Nov 2017, at 19:28, Trevor Leffler wrote: > > Well? nevermind this. A little searching reminding me that I can replace right join queries with left join queries (and vice versa). I?ve replaced my handful of right join queries and have things working. > > Cheers, --Trevor -- The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE. -------------- next part -------------- An HTML attachment was scrubbed... URL: From trsibley at uw.edu Tue Nov 7 20:48:08 2017 From: trsibley at uw.edu (Thomas Sibley) Date: Tue, 7 Nov 2017 12:48:08 -0800 Subject: [Dbix-class] view and is_nullable In-Reply-To: References: Message-ID: What I do in this situation is amend the generated result definition with another call to add_columns, like so: __PACKAGE__->add_columns(?+column_name" => { is_nullable => 0 }); That overrides the generated value, and dbicdump preserves your custom amendments to the class provided they?re at the bottom, outside of the delimited ?don?t change this? block. > On Oct 18, 2017, at 07:06 , Alex Povolotsky wrote: > > Hello > > I'm using a view to make code more readable; dbicdump sets is_nullable to 1 for all fields. > > Than I'm copying relation from one of the tables used in view, and > > "might_have/has_one" must not be on columns with is_nullable set to true (Billing::Schema::Result::BillDebt/at_id). This might indicate an incorrect use of those relationship helpers instead of belongs_to > > while that column is not nullable and join is not left. > > How do I make automated dbicdump's view column not nullable? > > Alex > > > _______________________________________________ > List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class > IRC: irc.perl.org#dbix-class > SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ > Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk From v.melnik at uplink.ua Tue Nov 28 09:56:11 2017 From: v.melnik at uplink.ua (Vladimir Melnik) Date: Tue, 28 Nov 2017 11:56:11 +0200 Subject: [Dbix-class] Result->id Message-ID: <20171128095611.GG18385@shagomer.uplink.tucha13.net> Dear colleagues, Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: __PACKAGE__->set_primary_key("id", "known_from"); But it always return one number - the content of the `id` column. What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? Thanks in advance for your time, dear colleagues. If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. -- V.Melnik From alexander.hartmaier at t-systems.at Tue Nov 28 12:46:17 2017 From: alexander.hartmaier at t-systems.at (Hartmaier Alexander) Date: Tue, 28 Nov 2017 13:46:17 +0100 Subject: [Dbix-class] Result->id In-Reply-To: <20171128095611.GG18385@shagomer.uplink.tucha13.net> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> Message-ID: <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> Hi, when you define your 'id' column using add_column it generates an accessor method named 'id' which overwrites the id method of DBIx::Class::PK. Best regards, Alex On 2017-11-28 10:56, Vladimir Melnik wrote: > Dear colleagues, > > Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? > > So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: > > __PACKAGE__->set_primary_key("id", "known_from"); > > But it always return one number - the content of the `id` column. > > What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? > > Thanks in advance for your time, dear colleagues. > > If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. > *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* T-Systems Austria GesmbH Rennweg 97-99, 1030 Wien Handelsgericht Wien, FN 79340b *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* Notice: This e-mail contains information that is confidential and may be privileged. If you are not the intended recipient, please notify the sender and then delete this e-mail immediately. *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* From v.melnik at uplink.ua Wed Nov 29 11:39:44 2017 From: v.melnik at uplink.ua (Vladimir Melnik) Date: Wed, 29 Nov 2017 13:39:44 +0200 Subject: [Dbix-class] Result->id In-Reply-To: <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> Message-ID: <20171129113944.GH18385@shagomer.uplink.tucha13.net> Thank you! That's why I can't decide should I rename the `id` column to `_id` or should I only change the 1st parameter of the `add_columns()` call from `id` to `_id`. Or should I make my own method (and call it, let's say, `PKs`) to do the same thing that `DBIx::Class::PK::id` does? It's always difficult to decide what way is better when all ways are "correct". :-) Being grateful for all your comments and suggestions regarding this topic, Vlad On Tue, Nov 28, 2017 at 01:46:17PM +0100, Hartmaier Alexander wrote: > Hi, > > when you define your 'id' column using add_column it generates an > accessor method named 'id' which overwrites the id method of > DBIx::Class::PK. > > Best regards, Alex > > > On 2017-11-28 10:56, Vladimir Melnik wrote: > >Dear colleagues, > > > >Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? > > > >So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: > > > > __PACKAGE__->set_primary_key("id", "known_from"); > > > >But it always return one number - the content of the `id` column. > > > >What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? > > > >Thanks in advance for your time, dear colleagues. > > > >If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. > > > > > > *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* > T-Systems Austria GesmbH Rennweg 97-99, 1030 Wien > Handelsgericht Wien, FN 79340b > *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* > Notice: This e-mail contains information that is confidential and may be privileged. > If you are not the intended recipient, please notify the sender and then > delete this e-mail immediately. > *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* > > _______________________________________________ > List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class > IRC: irc.perl.org#dbix-class > SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ > Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk -- V.Melnik From jgrg at sanger.ac.uk Wed Nov 29 13:53:29 2017 From: jgrg at sanger.ac.uk (James Gilbert) Date: Wed, 29 Nov 2017 13:53:29 +0000 Subject: [Dbix-class] Result->id In-Reply-To: <20171129113944.GH18385@shagomer.uplink.tucha13.net> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> <20171129113944.GH18385@shagomer.uplink.tucha13.net> Message-ID: <85EA2BBF-178A-476B-BC64-DB9674E36958@sanger.ac.uk> I always give my id columns a longer name incorporating the entity, such as "id_person". If you only call them "id", you will have many tables with column "id", making it hard to see how the tables are supposed to link together when looking at the schema. James > On 29 Nov 2017, at 11:39, Vladimir Melnik wrote: > > Thank you! > > That's why I can't decide should I rename the `id` column to `_id` or should I only change the 1st parameter of the `add_columns()` call from `id` to `_id`. Or should I make my own method (and call it, let's say, `PKs`) to do the same thing that `DBIx::Class::PK::id` does? > > It's always difficult to decide what way is better when all ways are "correct". :-) > > Being grateful for all your comments and suggestions regarding this topic, > Vlad -- The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE. From racke at linuxia.de Wed Nov 29 13:57:36 2017 From: racke at linuxia.de (Stefan Hornburg (Racke)) Date: Wed, 29 Nov 2017 14:57:36 +0100 Subject: [Dbix-class] Result->id In-Reply-To: <20171129113944.GH18385@shagomer.uplink.tucha13.net> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> <20171129113944.GH18385@shagomer.uplink.tucha13.net> Message-ID: On 11/29/2017 12:39 PM, Vladimir Melnik wrote: > Thank you! > > That's why I can't decide should I rename the `id` column to `_id` or should I only change the 1st parameter of the `add_columns()` call from `id` to `_id`. Or should I make my own method (and call it, let's say, `PKs`) to do the same thing that `DBIx::Class::PK::id` does? > > It's always difficult to decide what way is better when all ways are "correct". :-) I rather use products_id as PK for table products, it avoids this conflict and is also easier to distinguish in a query / linking table etc Just my two cents .. Regards Racke > > Being grateful for all your comments and suggestions regarding this topic, > Vlad > > On Tue, Nov 28, 2017 at 01:46:17PM +0100, Hartmaier Alexander wrote: >> Hi, >> >> when you define your 'id' column using add_column it generates an >> accessor method named 'id' which overwrites the id method of >> DBIx::Class::PK. >> >> Best regards, Alex >> >> >> On 2017-11-28 10:56, Vladimir Melnik wrote: >>> Dear colleagues, >>> >>> Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? >>> >>> So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: >>> >>> __PACKAGE__->set_primary_key("id", "known_from"); >>> >>> But it always return one number - the content of the `id` column. >>> >>> What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? >>> >>> Thanks in advance for your time, dear colleagues. >>> >>> If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. >>> >> >> >> >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> T-Systems Austria GesmbH Rennweg 97-99, 1030 Wien >> Handelsgericht Wien, FN 79340b >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> Notice: This e-mail contains information that is confidential and may be privileged. >> If you are not the intended recipient, please notify the sender and then >> delete this e-mail immediately. >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> >> _______________________________________________ >> List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class >> IRC: irc.perl.org#dbix-class >> SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ >> Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk > -- Ecommerce and Linux consulting + Perl and web application programming. Debian and Sympa administration. Provisioning with Ansible. From billcrawford1970 at gmail.com Wed Nov 29 14:04:42 2017 From: billcrawford1970 at gmail.com (Will Crawford) Date: Wed, 29 Nov 2017 14:04:42 +0000 Subject: [Dbix-class] Result->id In-Reply-To: <20171129113944.GH18385@shagomer.uplink.tucha13.net> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> <20171129113944.GH18385@shagomer.uplink.tucha13.net> Message-ID: If it's a third-party product ID (which the "known from" suggests), or it's non-unique for other reasons, having it named just "id" is misleading at best :) Maybe go with something like "their_id"? On 29 November 2017 at 11:39, Vladimir Melnik wrote: > Thank you! > > That's why I can't decide should I rename the `id` column to `_id` or should I only change the 1st parameter of the `add_columns()` call from `id` to `_id`. Or should I make my own method (and call it, let's say, `PKs`) to do the same thing that `DBIx::Class::PK::id` does? > > It's always difficult to decide what way is better when all ways are "correct". :-) > > Being grateful for all your comments and suggestions regarding this topic, > Vlad > > On Tue, Nov 28, 2017 at 01:46:17PM +0100, Hartmaier Alexander wrote: >> Hi, >> >> when you define your 'id' column using add_column it generates an >> accessor method named 'id' which overwrites the id method of >> DBIx::Class::PK. >> >> Best regards, Alex >> >> >> On 2017-11-28 10:56, Vladimir Melnik wrote: >> >Dear colleagues, >> > >> >Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? >> > >> >So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: >> > >> > __PACKAGE__->set_primary_key("id", "known_from"); >> > >> >But it always return one number - the content of the `id` column. >> > >> >What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? >> > >> >Thanks in advance for your time, dear colleagues. >> > >> >If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. >> > >> >> >> >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> T-Systems Austria GesmbH Rennweg 97-99, 1030 Wien >> Handelsgericht Wien, FN 79340b >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> Notice: This e-mail contains information that is confidential and may be privileged. >> If you are not the intended recipient, please notify the sender and then >> delete this e-mail immediately. >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> >> _______________________________________________ >> List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class >> IRC: irc.perl.org#dbix-class >> SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ >> Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk > > -- > V.Melnik > > _______________________________________________ > List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class > IRC: irc.perl.org#dbix-class > SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ > Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk From gjungwirth at sipwise.com Wed Nov 29 17:27:38 2017 From: gjungwirth at sipwise.com (Gerhard Jungwirth) Date: Wed, 29 Nov 2017 18:27:38 +0100 Subject: [Dbix-class] Result->id In-Reply-To: <20171129113944.GH18385@shagomer.uplink.tucha13.net> References: <20171128095611.GG18385@shagomer.uplink.tucha13.net> <2611fffe-e4ee-0d1a-b55e-9f9c20dc92c5@t-systems.at> <20171129113944.GH18385@shagomer.uplink.tucha13.net> Message-ID: AFAIK it should not be neccessary to rename the actual DB column. It is sufficient, to rename its accessor in the schema definition i.e. what you call it in add_columns() On 2017-11-29 12:39, Vladimir Melnik wrote: > Thank you! > > That's why I can't decide should I rename the `id` column to `_id` or should I only change the 1st parameter of the `add_columns()` call from `id` to `_id`. Or should I make my own method (and call it, let's say, `PKs`) to do the same thing that `DBIx::Class::PK::id` does? > > It's always difficult to decide what way is better when all ways are "correct". :-) > > Being grateful for all your comments and suggestions regarding this topic, > Vlad > > On Tue, Nov 28, 2017 at 01:46:17PM +0100, Hartmaier Alexander wrote: >> Hi, >> >> when you define your 'id' column using add_column it generates an >> accessor method named 'id' which overwrites the id method of >> DBIx::Class::PK. >> >> Best regards, Alex >> >> >> On 2017-11-28 10:56, Vladimir Melnik wrote: >>> Dear colleagues, >>> >>> Is `id()` should return a content of the `id` column if it presents, even if we have a multi-column PK (so it's supposed to return a list of PK values as it described in the `DBIx::Class::Row` manual)? >>> >>> So, I have a table whose PKs are `id` and `known_from`, I have the following in the Result-based class definition: >>> >>> __PACKAGE__->set_primary_key("id", "known_from"); >>> >>> But it always return one number - the content of the `id` column. >>> >>> What overrides what: an internal method and a column-name-based method? If the column-name-based one has higher priority, should I consider renaming the `id` field to some crappy `_id`? Or would it be a better way to call it as '_id' when I'm doing `__PACKAGE__->add_columns("id", ...)`? If yes, would it be possible to ask `DBIx::Class::Schema::Loader` to do that each time? >>> >>> Thanks in advance for your time, dear colleagues. >>> >>> If anyone have experince with multi-column PK with DBIx::Class and with temporal (also as bi-temporal and multi-temporal) - please share your wisdom. >>> >> >> >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> T-Systems Austria GesmbH Rennweg 97-99, 1030 Wien >> Handelsgericht Wien, FN 79340b >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> Notice: This e-mail contains information that is confidential and may be privileged. >> If you are not the intended recipient, please notify the sender and then >> delete this e-mail immediately. >> *"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"*"* >> >> _______________________________________________ >> List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class >> IRC: irc.perl.org#dbix-class >> SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/ >> Searchable Archive: http://www.grokbase.com/group/dbix-class at lists.scsys.co.uk