Caught NameError: undefined local variable or method `attributes' for
YAML::YPath:Class
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/util/ann_attr.rb:51:in
`serializable_attributes'
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:192:in
`manageable?'
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:222:in
`manageable_classes'
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:218:in
`manageable_classes'
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:239:in
`manage_classes'
/usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/main.rb:192:in
`setup'
-Matt
on 27.12.2007 14:42
on 28.12.2007 06:25
On Dec 27, 8:32 am, Matthew B Gardner <weat...@speakeasy.net> wrote: > /usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:239:in > `manage_classes' > /usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/main.rb:192:in `setup' I though I already fixed this, but maybe George hasn't had the chance to apply the patch. The solution is simply that #attributes should be #instance_attributes. T.
on 28.12.2007 12:21
Hello -- On Thursday 27 December 2007 23:47, Trans wrote: > > /usr/lib/ruby/gems/1.8/gems/og-0.50.0/lib/og/manager.rb:218:in > T. Thanks for the reply -- I fixed the issue there and then another that popped up in sql.rb as well. However, more problems arose once I got my application running -- searching for data in a table was giving incorrect results and saving an object was trying to save it to its parent's table. Also, the parent table had fields which it shouldn't have...fields declared only in another child (not the same child that I encountered the saving problem with). Are these problems related to the pending patch? Thanks again, Matt
on 28.12.2007 15:30
Hi, [...] > saving an object was trying to save it to its parent's table. Also, the > parent table had fields which it shouldn't have...fields declared only in > another child (not the same child that I encountered the saving problem > with). Are these problems related to the pending patch? if you are using STI (single table inheritance, `is SingleTableInherited` I think it is now), that is the correct behaviour. If not, it's a bug. Jo
on 05.01.2008 07:33
Hello -- > Thanks for the reply -- I fixed the issue there and then another that > popped up in sql.rb as well. However, more problems arose once I got my > application running -- searching for data in a table was giving incorrect > results and saving an object was trying to save it to its parent's table. > Also, the parent table had fields which it shouldn't have...fields declared > only in another child (not the same child that I encountered the saving > problem with). Are these problems related to the pending patch? > I'm still encountering the above issues, although I'm also dealing with more issues now. I deleted my tables to see if Og just needed to rebuild them...that didn't work and actually created more problems. RootObject is a parent of Account, but RootObject is adding and deleting Account's columns and an Account table isn't being created at all. Character is also a child of RootObject and RootObject is taking two of the columns in Character as its own. Here's a printout -- none of the columns printed should be part of RootObject: INFO: Created table `ogrootobject`. INFO: Adding field 'its_id' to '`ogrootobject`' INFO: Adding field 'its_update' to '`ogrootobject`' INFO: Adding field 'bts_id' to '`ogrootobject`' INFO: Adding field 'bts_update' to '`ogrootobject`' INFO: Adding field 'logins' to '`ogrootobject`' INFO: Adding field 'type' to '`ogrootobject`' INFO: Adding field 'password' to '`ogrootobject`' INFO: Adding field 'created_host' to '`ogrootobject`' INFO: Adding field 'last_host' to '`ogrootobject`' INFO: Adding field 'color' to '`ogrootobject`' INFO: Adding field 'hints' to '`ogrootobject`' INFO: Adding field 'linemode' to '`ogrootobject`' INFO: Adding field 'compact' to '`ogrootobject`' INFO: Adding field 'pagebreak' to '`ogrootobject`' INFO: Removing field 'pagebreak' from '`ogrootobject`' INFO: Removing field 'logins' from '`ogrootobject`' INFO: Removing field 'created_host' from '`ogrootobject`' INFO: Removing field 'compact' from '`ogrootobject`' INFO: Removing field 'type' from '`ogrootobject`' INFO: Removing field 'its_id' from '`ogrootobject`' INFO: Removing field 'last_host' from '`ogrootobject`' INFO: Removing field 'its_update' from '`ogrootobject`' INFO: Removing field 'hints' from '`ogrootobject`' INFO: Removing field 'password' from '`ogrootobject`' INFO: Removing field 'bts_id' from '`ogrootobject`' INFO: Removing field 'linemode' from '`ogrootobject`' INFO: Removing field 'color' from '`ogrootobject`' INFO: Removing field 'bts_update' from '`ogrootobject`' INFO: Adding field 'aliases' to '`ogrootobject`' INFO: Adding field 'account_oid' to '`ogrootobject`' Does this point to something I should be doing that I'm not? My application is pretty much paralyzed until I get this issue fixed, so any help is very much appreciated. Thanks, Matt
on 06.01.2008 12:35
Hi, > Does this point to something I should be doing that I'm not? My > application is pretty much paralyzed until I get this issue > fixed, so any help is very much appreciated. like I wrote before, if you are using STI, this is the correct behaviour. That it dropps fields, tells me that you must be using `:evolve_schema => :full`, which hmm.. might not be good with STI. If you don't use STI, it's a bug and Trans should fix it. :P class Model1 attr_accessor :chunky, String end class Model2 < Model1 attr_accessor :bacon, String end Og.start Model2.table # => 'ogmodel2' If this prints ogmodel1, there's a bug. The other Og guys here should fix it. :P Jo