Comments on: Custom rake tasks in Merb: Data Backup and Import

By: jc

jc — Sun, 27 Sep 2009 10:13:16 +0000

adapter.send(:with_connection) do |connection|
adapter.execute(”INSERT INTO #{model_class.name.pluralize.snake_case} (#{fixture.keys.collect{|key| adapter.send(:quote_name, key.to_s)}.join(”,”)}) VALUES (#{values.collect {|value| connection.quote_value(value)}.join(”,”)})”)
end

By: jc

jc — Sun, 27 Sep 2009 09:30:07 +0000

This no longer works in datamapper 0.10. quote_column_name no longer exists on adapter. Supposedly its in the connection now, but I cant find it

By: Brian Stolz

Brian Stolz — Fri, 10 Jul 2009 16:36:30 +0000

I was having issues with join tables which have names such as “roles_users” because of the double pluralization.

To resolve this I changed:

INSERT INTO #{table.name.pluralize.snake_case}

to:

INSERT INTO #{table.storage_name}

This uses the proper table name instead.

By: jc

jc — Sun, 05 Apr 2009 08:51:41 +0000

Here’s what the timestamps look like in the YML file

:created_at: 2009-03-31T00:04:45-07:00
:updated_at: 2009-03-31T00:04:45-07:00

By: jc

jc — Sun, 05 Apr 2009 08:50:59 +0000

Looks like its a YAML bug. Tracked here:
http://redmine.ruby-lang.org/issues/show/752

It fails just loading the generated YAML file.
irb(main):014:0> YAML.load_file “config/data/development/Job.yml”
ArgumentError: time out of range
from /opt/local/lib/ruby/1.8/yaml.rb:133:in `utc’
from /opt/local/lib/ruby/1.8/yaml.rb:133:in `node_import’
from /opt/local/lib/ruby/1.8/yaml.rb:133:in `load’
from /opt/local/lib/ruby/1.8/yaml.rb:133:in `load’
from /opt/local/lib/ruby/1.8/yaml.rb:144:in `load_file’

By: mikong

mikong — Mon, 23 Mar 2009 02:06:39 +0000

I have created_at and updated_at in several places in the project that I used this. I didn’t have any problems. Can you give more details about your environment? The Merb version and DM version you’re using, what your database is, and maybe how your updated_at/created_at data looks like in your YAML file?

I like YAML because it has a nice format, and it’s quite easy to modify the data before loading it. I guess if I had your errors and couldn’t make YAML work, then I’d consider other solutions. Btw, it’s not exactly my solution. As mentioned above, I just translated Tobi’s work which I think is amazing coz it was an old article, and yet it just worked for my case.

Anyway, I hope you could give me the details of your environment so I can take a look at the problem.

By: jc

jc — Sun, 22 Mar 2009 23:00:34 +0000

Hrmmm.. the load task doesn’t work. It croaks on the created_at and updated_at columns. “Invalid time”

Seems like YAML is a very odd choice for this. Why not just use sql in and sql out? The data format conversions make things very unstable.

By: mikong

mikong — Mon, 02 Mar 2009 16:36:59 +0000

Hi Philippe,

I suggest you try dumping some data so you can see the yaml expected. Here’s an example data dump on my side:

--- - :word: burger :id: 1 - :word: food :id: 2

By: Philippe Rathé

Philippe Rathé — Sun, 01 Mar 2009 20:15:06 +0000

Can you tell me what your script expect as yaml syntax?
I get the following error:
rake aborted!
undefined method `keys’ for [”reason1″, {”name”=>”My own reason”}]:Array

With this yaml:
reason1:
name: My own reason

Thanks

By: Uri

Uri — Tue, 16 Dec 2008 20:27:02 +0000

Thanks!