On 3/1/06 6:34 PM, Ron Savage wrote:
> I've gained 'create view' privilege since I sent you that code, and have
> activated the commented-out references to views in list-tables.t, and those
> aspects of the tests now pass also.
Great :) Are you set up with SVN yet?
-John
On 3/1/06 10:44 PM, Ron Savage wrote:
> Pursuant to my attempts to generate Perl code under Oracle, I had to make a
> number of patches to this module.
Ah, I didn't know you were editing Rose::DB::Object modules yet. I'll add
them to the Oracle branch.
Having just taken a casual look at your cha
On 3/2/06 12:41 AM, Ron Savage wrote:
> I wasn't intending to work on it. I just wanted to fiddle it to see what
> happened with my source code generator.
Ah, okay. I added it to the branch anyway, though. At the very least,
it'll give me good practice doing merges in SVN :)
> For the record, w
On 3/2/06 10:14 PM, Ron Savage wrote:
> Both Rose::DB::Object::Metadata::Auto and
> Rose::DB::Object::Metadata::Auto::MySQL are missing an 'l' from:
>
> no primary key coumns found
Thanks, fixed. (It was in Informix and SQLite too. Copy/paste...)
-John
-
On 3/2/06 10:48 PM, Ron Savage wrote:
> I'm writing auto_retrieve_primary_key_column_names in this module, and the
> problem is the call $schema = $self->select_schema($db).
>
> Oracle needs to to return uc $username.
You can do whatever you want in Rose::DB::Object::Metadata::Auto::Oracle.
Just
On 3/3/06 12:11 AM, Ron Savage wrote:
> But how come it doesn't ask me to authenticate when I upload a changed file?
My guess is that it somehow found your ssh keys or you SF.net login name in
a browser cookie or some crazy thing. Also, after I authenticate using even
the regular svn command-line
On 3/7/06 7:31 PM, Ron Savage wrote:
> package Monash::Rose::MorseCode::Base::DB;
> ...
> Rose::DB -> use_private_registry();
> Rose::DB -> default_domain('windows');
> Rose::DB->register_db(...)
>
> o At the start of the first package I have:
> Rose::DB -> use_private_registry();
> sort of as per
Thanks, applied.
-John
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new codi
On 3/8/06 6:06 PM, Nilson Santos Figueiredo Junior wrote:
> Today I was wandering... is there anyone implementing a Catalyst model
> based on Rose::DB::Object? I think this would be something nice to
> have, both for Rose::DB::Object and Catalyst.
>
> This might be a nice project for someone to pl
On 3/8/06 10:11 PM, Nilson Santos Figueiredo Junior wrote:
> It's usually simple glue code and without it, things get less coherent.
> You get one nice thing, though: $c->model('Artist') (where $c is the
> Catalyst context object) returns the appropriate object for the
> "Artist" model.
What's the
On 3/8/06 11:27 PM, Nilson Santos Figueiredo Junior wrote:
> On 3/9/06, John Siracusa <[EMAIL PROTECTED]> wrote:
>> Well, anyone could read the Rose::DB and RDBO tutorials and have all the
>> information they need, I think.
>
> If this is really true, I think I'
I'm adding a Rose::DB::Object::Helpers module to house all of the "extra"
methods that people might want, but that I don't want crowding
Rose::DB::Object itself. Example usage:
package MyDBObject;
use base 'Rose::DB::Object';
use Rose::DB::Object::Helpers qw(load_or_insert load_specu
No one suggested any helper methods, so I shipped with just the two that I
created.
0.69 (03.12.2006) - John Siracusa <[EMAIL PROTECTED]>
* Added Rose::DB::Object::MixIn and Rose::DB::Object::Helpers.
* Made insert() and update() part of Rose::DB::Object's public API.
*
On 3/14/06 4:17 AM, Michael Lackhoff wrote:
> Is it possible to either define a default sort_by clause in the
> relationship definition
See Uwe's earlier response for how to do this.
> or when the relationship method is actually used
> ($company->products({sort_by => 'name'}) or something
That's
On 3/14/06 4:17 AM, Michael Lackhoff wrote:
> So woudn't it be great if it was possible to maintain only the class
> definition and have a 'create_sql_from_class' method that gave the
> sql to create a new table according to the actual class definition.
It's planned. I'm going to use SQL::Transla
On 3/14/06 8:12 AM, Uwe Voelker wrote:
>> Can you think of a good way to pass custom manager_args to individual calls
>> to a relationship accessor? Maybe a new read-only accessor type where
>> arguments can't possible be meant to set values?
>
> Is it possible to make a second accessor for the s
On 3/15/06 6:02 AM, Michael Lackhoff wrote:
> To my understanding the two definitions should be equivalent but with
> the second I get the error message:
> Don't know how to auto-join relationship 'gruppe' of type 'many to
> one' at C:/Perl/site/lib/Rose/DB/Object/MakeMethods/Generic.pm line
> 3330
On 3/15/06, Michael Lackhoff <[EMAIL PROTECTED]> wrote:
> Now everything seems broken:
> First I got some test failures in db-object-loader-5
> I installed anyway but before made a backup.copy of the
> Rose/DB/Object directory.
> Now I always get an error message:
> Cannot create method PVonline::T
On 3/15/06 8:38 AM, Michael Lackhoff wrote:
> Now I always get an error message:
> Cannot create method PVonline::Therapeuten::anmeldename - method
> already exists
Just to let the list know, this turned out to be due to a silly mistake in
Michael's code that I shan't elaborate on in an effort to
This update fixes a significant bug and is recommended for all users.
0.691 (03.16.2006) - John Siracusa <[EMAIL PROTECTED]>
* Fixed a bug that prevented the Manager from correctly handling
many-to-one relationships in some situations. (Reported by
Michael Lackhoff)
File
On 3/22/06, Sean Davis <[EMAIL PROTECTED]> wrote:
> I would like to do a large amount of read->parse->store operations. In
> short, I have a large text file with one record per line. To speed things
> up, I have seen others "reuse" objects rather than allocating new ones for
> each record. Can I
On 3/22/06, Sean Davis <[EMAIL PROTECTED]> wrote:
> Why is the type for "contents" a scalar?
Because that's the default when there is not a more specific RDBO
column type mapped to whatever DBI returns for the column type.
> I would like to store binary data in that column, which I can do with st
On 3/22/06 6:45 PM, Sean Davis wrote:
> Just tried it and Base64 encoding works fine. I can presumably inflate and
> deflate within my RDBO class. Thanks for the idea.
Yeah, using column triggers...but if you're feeling energetic, you could
also make a proper "blob" column class and associated m
On 3/22/06 8:38 PM, Daniel Pittman wrote:
> Daniel Pittman <[EMAIL PROTECTED]> writes:
>> Sean Davis <[EMAIL PROTECTED]> writes:
>>> Just tried it and Base64 encoding works fine. I can presumably
>>> inflate and deflate within my RDBO class. Thanks for the idea.
>>
>> Presumably that works becau
On 3/22/06 8:48 PM, John Siracusa wrote:
> Looking here:
>
> http://www.postgresql.org/docs/8.1/interactive/datatype-binary.html
>
> it seems that there's already an ASCII-ified format for bytea columns.
Furthermore, bytea data seems to round-trip just fine without
On 3/22/06 9:43 PM, Sean Davis wrote:
> No error. I just got truncated insertion (presumably from a \000?).
Ah, yeah, it does seem to truncate on null bytes:
my $o1 = Foo->new(b => "\001\000\003");
$o1->save;
my $o2 = Foo->new(id => $o1->id);
$o2->load;
print unpack('H*', $o1->b), "
On 3/24/06, Michael Lackhoff <[EMAIL PROTECTED]> wrote:
> Say, I have two tables: vendors and products. Now I want a list of all the
> vendors that have no products yet. This can be seen from the products table if
> there are no records with the vendor_id of the vendor in question. Is this
> possib
Ha, Cees is a faster typist I guess... :)
-John
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group brea
On 3/25/06 10:44 AM, Michael Lackhoff wrote:
>> $subselect =
>> 'NOT EXISTS (SELECT * FROM products p WHERE p.vendor_id = id)';
>>
>> $vendors = Vendor::Manager->get_vendors(clauses => [ $subselect ]);
>
> This doesn't work for me (result list is empty but should have two
> entries
/showfiles.php?group_id=147570
Full change lists are below.
-John
Rose::DB::Object:
0.70 (04.04.2006) - John Siracusa <[EMAIL PROTECTED]>
* Added support for the interval data type.
* Added support for big integer and serial columns via Math::BigInt.
* Added explicit suppo
I screwed up the Rose::DB 0.67 release: stray "._*" files snuck into the tar
archive yet again! Grrr. Anyway, 0.671 was just released. (I repaired the
0.67 package on SF.net, but I had to upload a new version to CPAN.)
-John
---
This SF.Ne
On 4/5/06 1:54 AM, Lucian Dragus wrote:
> Thanks for this new release :)
> Unfortuantely, the "._*" files are present in Rose-DB-Object-0.70, too :(
Blah, sorry. Version 0.701 released to CPAN and SF.net.
http://sourceforge.net/project/showfiles.php?group_id=147570
-John
---
and Rose::DateTime 0.521 released too. Can you guess why? Sigh.
-John
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join
On 4/8/06 9:02 AM, Michael Lackhoff wrote:
> At last I tried to install the new versions. I get this Test result
> (with 0.521):
>
> C:\Perl\bin\perl.exe "-MExtUtils::Command::MM" "-e"
> "test_harness(0, 'bl
> ib\lib', 'blib\arch')" t\basic.t t\locale.t t\parser.t t\pod.t
> t\basic.ok
> After digging a little bit, I saw that when updating vendor,
> it first deletes all products for this vendor and tries to add the new
> ones. So far, so good. But all the objects retrieved using products()
> method are in state "in_db" which means that they will be updated
> (using SQL UPDATE) in
On 4/14/06 2:24 AM, Lucian Dragus wrote:
> I have a small patch to Rose::DB::Pg, in refine_dbi_column_info(), to
> recognize the float4 (real) columns as float types.
Thanks, applied.
-John
---
This SF.Net email is sponsored by xPML, a groun
On 14 Apr 2006 11:21:50 -0400, Ted Zlatanov <[EMAIL PROTECTED]> wrote:
> On 14 Mar 2006, [EMAIL PROTECTED] wrote:
>> Can you think of a good way to pass custom manager_args to individual calls
>> to a relationship accessor? Maybe a new read-only accessor type where
>> arguments can't possible be m
On 14 Apr 2006 11:25:05 -0400, Ted Zlatanov <[EMAIL PROTECTED]> wrote:
> On 9 Mar 2006, [EMAIL PROTECTED] wrote:
>> package MyDBObject;
>>
>> use base 'Rose::DB::Object';
>> use Rose::DB::Object::Helpers qw(load_or_insert load_speculative);
>>
>> My question is, what methods should be in the helpe
On 4/14/06, Jesse Brown <[EMAIL PROTECTED]> wrote:
> As a side note, I've noticed that mysql 'bigint' gets converted to a 20 byte
> scalar, which is about as correct as it can get on 32bit machines. However, on
> perls with 64 bit integers, it should just be an int type. (more and more
> common wit
(04.14.2006) - John Siracusa <[EMAIL PROTECTED]>
* Correctly clear the "in the database" state flag when a
speculative load fails because the object was not found.
(Reported indirectly by Svilen Ivanov)
* Changed how multi-table queries are formulated for SQLite i
On 4/14/06 4:56 PM, Jesse Brown wrote:
> Some of my code relies on 64 bit ints (very large file transfers, etc), and I
> use the following to make sure...
>
> # We require 64bit integers in order to work properly
> # (large file offsets and sizes).
> BEGIN {
> use Config;
> die "You do not hav
) - John Siracusa <[EMAIL PROTECTED]>
* Many reference-count bugs fixed, including several that could cause
database handles to leak.
* Fixed a bug that could cause "empty" sub-objects to be created when
processing certain with_objects parameters.
* Impro
On 4/24/06 5:52 PM, Clayton Scott wrote:
> My schema has a table "address". When using the default ConventionManager
> and autogenerating classes the coresponding class will have it's last
> 's' dropped
> to become "Addres".
>
> I expected that using RDBO::ConventionManager::Null would prevent tha
On 4/25/06, Clayton Scott <[EMAIL PROTECTED]> wrote:
> What kind of problems can I expect to run into if I reimplement the
> plural_to_singular and the other related methods to do the equivalent of:
>
> sub plural_to_singular
> {
> my($self, $word) = @_;
> return $word;
> }
>
> I ask because I'
tion of the Rose::DB::Object
documentation.
Files sent to CPAN and on SF.net
http://sourceforge.net/project/showfiles.php?group_id=147570
(I screwed up the 0.721 release by forgetting to expunge a few uses of
Clone::PP. Sorry about that...)
-John
0.722 (04.27.2006) - John Siracusa <[EMAIL PR
On 4/29/06 12:20 PM, Guillermo Roditi wrote:
> It seems to run fine but then I have no output or anything? is it supposed
> to ouput to stdout? or a file?
It depends on which methods you call.
> --
> #! /usr/bin/perl -w
> use strict;
> use warnings;
> use diagnostics;
> use Ro
On 4/30/06 2:24 PM, Michael Lackhoff wrote:
> This script works perfectly well when run from the command line:
> use PVonline::Therapeuten;
>
> sub groups_of_user {
>return map {$_->name} PVonline::Therapeuten->new(nr=>1)->gruppen;
> }
>
> print groups_of_user();
> # prints the names of
On 5/1/06 3:06 AM, Michael Lackhoff wrote:
> So the bottom line is that there was a recent bugfix that is
> important for many to many relationships.
I don't remember that one, but there have been so many updates that I've
kind of lost track. Anyway, I'm glad the current version works correctly :
On 5/1/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
Is there any reason why I wouldn't want to use Rose::DB::Object::Cached for
everything instead of just using Rose::DB:: Object as a base for my
database table classes?
I think the description section from the POD contains some very good
rea
On 5/1/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
Hi, I am having a little trouble with Rose::DB::Object.
I keep getting the error below and I am not entirely sure what it means...
Dos it mean that it couldnt guess the foreign class name ?
Are you using the Loader? If so, can you provide
On 5/1/06 7:09 PM, Guillermo Roditi wrote:
> Is it because there is a circular relationship??? where every user has many
> props and many props have one user? it seems what heppens is that its
> trying to make a user object and so it makes the prop type object and then
> it tries to make a user o
On 5/1/06 7:53 PM, Guillermo Roditi wrote:
> use ing works just dandy, its when i relationship() the modules that it
> breaks : /
As I suspected, it was a timing issue for the circular dependencies. I'll
work on making the Loader account for these automatically in its generated
code. In the mean
On 5/2/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
I added it to the wiki. and I also reverted some spam.
http://rose.sourceforge.net/wiki/index.php/Generating_Schema_Graphs
Thanks! The wiki spam is out of control. Anyone know how I can fight
it using mediawiki? Blocking the spammer IP a
On 5/3/06, Michael Lackhoff <[EMAIL PROTECTED]> wrote:
groups => {
type => 'many to many',
map_class => 'Groupmappings',
},
but how can I negate this, so I get all the groups, the user is not a
member of?
You're going to have to do a Manager query (wrapped in a met
On 5/3/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
No such Cantella::PaperKiller::DB::UserProp where
id = 153
Are you sure it's connecting to teh same db as the Manager query?
Also, can you set $Rose::DB::Object::Debug = 1 before making that
load() call to see the SQL it's running?
-John
On 5/3/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
another question though. any reason why there is no 'time' type for columns?
Because I haven't gotten around to it yet :)
-John
---
Using Tomcat but need to do more? Need to support web
On 5/3/06, Guillermo Roditi <[EMAIL PROTECTED]> wrote:
I have a table which includes a blob and a field for a hash of that blob.
I'd like to make atrigger that automatically updates the hash every time the
blob is changed, so i figured i would do it with a trigger. I figured the
best way to do it
On 5/3/06 5:58 PM, Guillermo Roditi wrote:
> Normally I would just create an aggregator class that managed the dispatching
> of the right object types and queried all tables and merged the results
That sounds like a fine solution to me.
> but I'd like to make something that can more closely iterf
On 5/3/06 7:41 PM, Guillermo Roditi wrote:
> SQL definitions are at the bottom if i am not clear enough
>
> yeah, i am basically trying to do a view, but myISAM doesnt support it [...]
>
> I baiscally want the different types of property vlues to be as transparent
> to the interface as possib
On 5/5/06 6:25 AM, Michael Lackhoff wrote:
> I would like to copy objects from a default record/object. I tried it
> with something like:
>
> use MyClass;
> use Clone qw(clone);
>
> my $obj = MyClass->new(nr=>1)->load;
> my $obj2 = clone($obj);
> $obj2->id(undef);
> $obj2->price($a_changed_value)
On 5/5/06 5:29 PM, Michael Lackhoff wrote:
> Works great but also makes greedy. How about a related method that
> also deletes the primary key and sets the db.
Maybe I'll add it to Rose::DB::Object::Helpers. In fact, I may move clone()
there too. (Remember, it's currently not part of the public
Bug fixes and some new helper methods. Files sent to CPAN and on SF.net:
http://sourceforge.net/project/showfiles.php?group_id=147570
0.723 (05.07.2006) - John Siracusa <[EMAIL PROTECTED]>
* Fixed a bug that caused Rose::DB::Object::Cached objects
to stay in the cache after
For Michael Lackhoff :
Yes, this also means that clone() is no longer in Rose::DB::Object proper.
You'll now have to add this to your base class:
use Rose::DB::Object::Helpers qw(clone);
...or, if you want to use clone_and_reset(), a I suspect you might, then do
this instead:
use Rose::
On 5/10/06 3:20 AM, Uwe Voelker wrote:
> initilzed -> initialized
Fixed, thanks.
-John
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easie
On 5/10/06 3:31 AM, Uwe Voelker wrote:
> from the documentation about inheritance:
>
> # Change the "start" column into a datetime column.
> __PACKAGE__->meta->delete_column('start');
> __PACKAGE__->meta->add_column(start => { type => 'datetime' });
>
> Why not add a method change_column()
(I think this bounced the first two times...apologies if it's a triple!)
On 5/10/06, Uwe Voelker <[EMAIL PROTECTED]> wrote:
> Between "swap" and "exchange" I would prefer "exchange" too. But maybe
> "replace"?
Yeah, I like that better too. Of course, it occurs to me now that you
can already just
On 5/10/06, Jesse Brown <[EMAIL PROTECTED]> wrote:
When adding objects to a relationship via $object->add_({ blah
=> 1, etc}), I seem unable to get back the list of objects that were created
and added to the relationship, which seems like it would be VERY convenient?
It also seems that the behav
On 5/10/06, Jesse Brown <[EMAIL PROTECTED]> wrote:
I guess the question I should ask is: Why NOT return the newly created
objects?
I just like to know the reasoning before I change the API :) The code
examples are exactly what I was looking for, thanks. The only thing I
remain unsure of is t
On 5/10/06 6:03 PM, Jesse Brown wrote:
> Re: the scalar context return value - IS it available elsewhere? (the number
> of items added in that call) - total number of items, yes, but new ones?
Yeah, because you're making the call! :) Example:
$o->add_widgets(@widgets); # $count = @widgets
On 5/10/06 7:20 PM, LU, ZHENYUAN [AG/1000] wrote:
> I am using Oracle.
Unfortunately, Rose::DB does not yet support Oracle. What you're using is
the "generic" interface that Rose::DB falls back to when it encounters an
unknown database.
Ron Savage is working on adding Oracle support to Rose::DB
On 5/10/06 9:57 PM, Jesse Brown wrote:
> If you would like me to implement, just point me to the
> place the add_ code is defined and I'll send you a patch :)
Thanks, but it's a simple change. I'll do it myself when I make a decision.
As always, policy is more trouble than implementation.
-John
On 5/11/06 3:19 AM, Michael Lackhoff wrote:
> You can get the free express edition here:
> http://www.oracle.com/technology/software/products/database/xe/index.html
Sure, for Linux or Windows. I have a Mac (and not an x86 one...yet).
-John
I added the features that have been discussed on the mailing list
recently. I'm still fighting with SF.net to get the files up there,
but I uploaded to CPAN (and, of course, checked into SVN).
0.724 (05.11.2006) - John Siracusa <[EMAIL PROTECTED]>
* Added the replace_column
On 5/11/06, LU, ZHENYUAN [AG/1000] <[EMAIL PROTECTED]> wrote:
How about the second problem I mentioned which is not specific to Oracle?
In list_tables sub, $table_info->{'TABLE_NAME'} is used, but it will not
work if people setup FetchHashKeyName => 'NAME_lc' in connect_options.
I'll work on t
The beginnings of Oracle support will go out in the next release of
Rose::DB. I can't test the Oracle code myself because I don't have
access to an Oracle database (or a machine that can run the free
version or Oracle). I'm looking for volunteers to test the code.
The Oracle code was written by
On 5/11/06, LU, ZHENYUAN [AG/1000] <[EMAIL PROTECTED]> wrote:
The problem involving my setup ( FetchHashKeyName => 'NAME_lc' ) is still
not solved.
Oops, sorry, I forgot that the new Oracle changes have their own
list_tables() method that I needed to fix :) Do "svn up" and then try
your list_t
On 5/11/06, LU, ZHENYUAN [AG/1000] <[EMAIL PROTECTED]> wrote:
Yeah, the list_tables() works now.
Good! :)
But I still can't use the Loader.
Although Oracle support is starting to creep into Rose::DB, there is
zero Oracle support in Rose::DB::Object right now. If anything works
with Oracle
Oracle support creeping in slowly, plus that FetchHashKeyName fix.
Sent to CPAN and SF.net:
http://sourceforge.net/project/showfiles.php?group_id=147570
0.673 (05.12.2006) - John Siracusa <[EMAIL PROTECTED]>
* Added partial support for Oracle databases. (Code by Ron Savage)
* F
On 5/13/06 11:19 PM, John Siracusa wrote:
> You need a build of Scalar::Util that supports weak references. IIRC, some
> builds bundled with certain Linux distributions are broken.
Er, make that "some Windows Perl distributions
On 5/16/06, Michael Lackhoff <[EMAIL PROTECTED]> wrote:
DBD::SQLite has the SQLite sources included, so a newer version of the driver
means a newer version of SQLite -- and hopefully fewer bugs.
Ah.
Can you reproduce the ORDER BY bugs with a recent version of sqlite3 (v3.3.5)?
I don't know,
On 5/14/06, Ron Savage <[EMAIL PROTECTED]> wrote:
I have DBD::SQLite V 1.09 installed. V 1.12 won't install under Windows. After
installing V 1.11, I re-ran the Rose::DB::Object install. The errors above
disappeared.
Yeah, I've been debating requiring DBD::SQLite 1.11 or later. SQLite
in gener
On 5/13/06 8:23 PM, Ron Savage wrote:
> I'm getting many errors (under Win2K and Perl V 5.8.7) like:
>
> # Failed test 'use Rose::DB::Object;'
> # in t/column-triggers.t at line 10.
> # Tried to use 'Rose::DB::Object'.
> # Error: Weak references are not implemented in the version of p
On 5/17/06 4:52 PM, Michael Drozdov wrote:
> On a minor note,
> local $Rose::DB::Object::Debug = 1
> doesn't print query (2).
That's because it uses the Manager to do that one. There are separate debug
variables for the Manager and Object. Set both to see everything:
local $Rose::DB::Objec
On 5/17/06 4:52 PM, Michael Drozdov wrote:
> In other words, it remembers the parent_id=26 generated by query (1),
> and misses child_id=27, generated by query (4).
This was a bug. It's fixed in SVN now, and the fix will go out in the next
RDBO release. Thanks for the report! :)
-John
-
Bug fixes...
0.725 (05.17.2006) - John Siracusa <[EMAIL PROTECTED]>
* Fixed a bug that caused self-referential many-to-many relationships to
fail during cascaded save() operations. (Reported by Michael Drozdov)
* The test suite now requires DBD::SQLite version 1.11 or
I screwed up the skip() count in the new test for the many-to-many bug fix.
0.726 (05.17.2006) - John Siracusa <[EMAIL PROTECTED]>
* Fixed incorrect skip count in t/spot-check-07.t
-John
---
Using Tomcat but need to do more? N
On 5/18/06, Jonathan Vanasco <[EMAIL PROTECTED]> wrote:
so my question is simply - is there a way for me to explicitly set
the db handle in rosedb objects midstream?
Sure, there are several ways. The usual way is to register different
data sources in your Rose::DB sublcass:
package My::DB
On 5/18/06, John Siracusa <[EMAIL PROTECTED]> wrote:
sub save
{
my $self = shift; # shift okay here; load() is the only special case
$self->db(My::DB->new('writer')) unless($self->db->type eq 'writer');
shift->SUPER::save(@_); # Call superclass
On 5/18/06, Ted Zlatanov <[EMAIL PROTECTED]> wrote:
On 10 May 2006, [EMAIL PROTECTED] wrote:
Oh well, I think replace_column() is more clear anyway. I'll add it
in the next version.
"replace" still has connotations of a permanent change. Why not
"remap"?
Too late, I already added it as rep
On 5/18/06, Ted Zlatanov <[EMAIL PROTECTED]> wrote:
On 10 May 2006, [EMAIL PROTECTED] wrote:
* Return a reference to an array of added objects
* Return the first added object.
> Hm, I'm not sure I like those much either. What I'm worried about is this:
$widget = $o->add_widgets($some_widget
On 5/19/06 9:26 PM, Jonathan Vanasco wrote:
> i need to do something so that
>
> sub init_db { NewApp::RoseDB->new }
>
> isn't actually a database connect
It isn't. Calling new() on a Rose::DB-derived object won't connect to the
database. It'll just create the object.
> in my very awkward set
What version of perl? Can anyone else reproduce this on Windows?
-John
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM W
On 5/22/06, Jonathan Vanasco <[EMAIL PROTECTED]> wrote:
I'm not sure how I can either swap a DBH into the ObjectManager
class, or just ignore it altogether (which I would not be averse to
doing )
Manager methods take a db parameter, If one is not passed, then one
will be created by calling the
On 5/22/06 7:07 PM, Jonathan Vanasco wrote:
> On May 22, 2006, at 5:34 PM, John Siracusa wrote:
>> Manager methods take a db parameter, If one is not passed, then one
>> will be created by calling the init_db() method of the object class.
>> So, you should be able to pass i
On 5/23/06 2:36 AM, [EMAIL PROTECTED] wrote:
> this is pretty generic 3nf 'crap' tables -- stuff like
>
> table filetype: id | name
> table accounttype: id | name
>
> about 10 tables, ~200 id:name pairigs total. bs stuff. using straight
> sql in the framework i built though, i learned pre-cachi
On 5/23/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
I'm wondering what the normal 'layout' is for files / packages under
mod_perl
Mod-perl or not, the most flexible way to handle things is to make a
bunch of (trivial) subclasses first to hold all your customizations
down the road, then bas
On 5/23/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
but what I've realized is that do handle the cached, I'd need something
like-
MyObj
MyObj::Manager
MyObj::Cached
MyObj::Cached::Manager
The benefits of having a Manager for a ::Cached class are dubious.
Unless you plan to save() (or reme
On 5/23/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
On May 23, 2006, at 3:52 PM, John Siracusa wrote:
The benefits of having a Manager for a ::Cached class are dubious.
Unless you
plan to save() (or remember() or forget()) some of the objects returned by
the Manager, it doesn&#x
On 5/24/06, Ethan Rowe <[EMAIL PROTECTED]> wrote:
For a class created by the loader, I found that I needed to go in and
adjust the configuration by hand (this isn't a big deal as long as I
know it's a necessary step).
It shouldn't be necessary. The loader should detect and configure the
sequenc
On 5/24/06, Cees Hek <[EMAIL PROTECTED]> wrote:
I noticed that when I try to set a boolean value to 0 it successfully
sets the value to false, but then fails to save the changes to the DB
when I call $object->save. This doesn't happen when you set the value
to 'f' or 'n' or any other value that
201 - 300 of 1026 matches
Mail list logo
