[Catalyst] DBIx::Class and managing schemas

[Jess Robinson]

Or you could look at the versioning branch of DBIx::Class, which will 
handle all that for you. See VERSIONING_SKETCH in the -current branch for 
a summary of what it will be able to do.

Jess


On Wed, 14 Jun 2006, Daniel McBrearty wrote:

> thanks!
>
> On 6/13/06, Richard Jolly <richardjolly at mac.com> wrote:
>> On 12 Jun 2006, at 20:53, Daniel McBrearty wrote:
>>
>>> I'm currently migrating engoi from the current mysql schema to pg.
>>> Along the way, the schema is changing shape quite drastically, for
>>> various reasons.
>>>
>>> Here's how I'm doing it:
>>>
>>> 1. writing the new pg schema in a single file (pg.sql)
>>> 2. writing a migrate script that gets data out of the old db and into
>>> the new.
>>> 3. testing, updating the cat app
>>> 4. repeat for a new chunk of original schema.
>>>
>>> Everything uses DBIx::Class now (can't believe I was rolling my own
>>> ORM a year back ...)
>>>
>>> My question : given that this new schema will not be set in stone (it
>>> will get added to as we add new functions, though hopefully we will
>>> not have to rehash the whole thing again) - am I making life harder
>>> than needs be for future?
>>>
>>> What I can see happening for future is this :
>>>
>>> 1. dev a new feature, along with a scripts to add to and maybe
>>> populate the schema as needed.
>>> 2. when ready use scripts to update the prod server.
>>>
>>> I just had a speed read of the cookbook for DBIx, and saw some stuff
>>> about managing schema versioning. I'm just wondering if my current way
>>> of working is going to be something I'm going to regret down the line
>>> ... to ge honest I don't have too much time to read all the docs, and
>>> I'm not sure if I'd get all teh implications on a first reading anyhow
>>> ...
>>
>> Take a look at DBIx::Migration. That would give you more management of
>> updates, and, importantly, rollbacks.  It is not a magic module - it
>> basically just manages a collection of update and rollback sql scripts
>> that you have to write.
>>
>> It could be used in conjunction with the DBIC method. The
>> DBIx::Class::Manual::Cookbook method relies on sqlt-diff (from
>> SQL::Translator) to produce a "diff" between two schemas as a
>> collection of update table statements. These diffs could be used as a
>> basis for the DBIx::Migration sql scripts. The schema files themselves
>> can be generated from DBIC.
>>
>> Richard
>>
>>
>> _______________________________________________
>> List: Catalyst at lists.rawmode.org
>> Listinfo: http://lists.rawmode.org/mailman/listinfo/catalyst
>> Searchable archive: http://www.mail-archive.com/catalyst@lists.rawmode.org/
>> Dev site: http://dev.catalyst.perl.org/
>>
>
>
> -- 
> Daniel McBrearty
> email : danielmcbrearty at gmail.com
> www.engoi.com : the multi - language vocab trainer
> BTW : 0873928131
>
> _______________________________________________
> List: Catalyst at lists.rawmode.org
> Listinfo: http://lists.rawmode.org/mailman/listinfo/catalyst
> Searchable archive: http://www.mail-archive.com/catalyst@lists.rawmode.org/
> Dev site: http://dev.catalyst.perl.org/
>
>