rspec users - Aug 2007 - undefined method `mock_model'' for [RSpec example]:#<Class:0x25550a8>

I am getting:
  undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>

with my view specs after upgrading to 1.0.8. Is anyone else seeing this?

To isolate the message I did the following setup:

 - create new rails project (with edge)
 - installed rspec and rspec_on_rails
 - ran rake spec
 - ran script/generate rspec_scaffold products
 - tried to run a builtin view spec

Zach

On 8/16/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> I am getting:
>   undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>
The fact that it says [RSpec example] means that it''s not the right
class for a view spec. The plugin knows how to create the right class
based on one of two things:

If the file is below /spec/views/

or

describe "...", :behaviour_type => :view do
  ...

If the spec you''re running satisfies either of these, you
shouldn''t be
seeing this.
>
> with my view specs after upgrading to 1.0.8. Is anyone else seeing this?
>
> To isolate the message I did the following setup:
>
>  - create new rails project (with edge)
>  - installed rspec and rspec_on_rails
>  - ran rake spec
>  - ran script/generate rspec_scaffold products
>  - tried to run a builtin view spec
When you say "builtin view spec," do you mean one that was generated
by running the rspec_scaffold generator?
>
> Zach
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 8/16/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > I am getting:
> >   undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>
>
> The fact that it says [RSpec example] means that it''s not the
right
> class for a view spec. The plugin knows how to create the right class
> based on one of two things:
>
> If the file is below /spec/views/
>
> or
>
> describe "...", :behaviour_type => :view do
>   ...
>
> If the spec you''re running satisfies either of these, you
shouldn''t be
> seeing this.
>
> >
> > with my view specs after upgrading to 1.0.8. Is anyone else seeing
this?
> >
> > To isolate the message I did the following setup:
> >
> >  - create new rails project (with edge)
> >  - installed rspec and rspec_on_rails
> >  - ran rake spec
> >  - ran script/generate rspec_scaffold products
> >  - tried to run a builtin view spec
>
> When you say "builtin view spec," do you mean one that was
generated
> by running the rspec_scaffold generator?
>
One that was generated by the rspec_scaffold generator,

Zach

On 8/16/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> On 8/16/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > > I am getting:
> > >   undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>
> >
> > The fact that it says [RSpec example] means that it''s not the
right
> > class for a view spec. The plugin knows how to create the right class
> > based on one of two things:
> >
> > If the file is below /spec/views/
> >
> > or
> >
> > describe "...", :behaviour_type => :view do
> >   ...
> >
> > If the spec you''re running satisfies either of these, you
shouldn''t be
> > seeing this.
> >
> > >
> > > with my view specs after upgrading to 1.0.8. Is anyone else
seeing this?
> > >
> > > To isolate the message I did the following setup:
> > >
> > >  - create new rails project (with edge)
> > >  - installed rspec and rspec_on_rails
> > >  - ran rake spec
> > >  - ran script/generate rspec_scaffold products
> > >  - tried to run a builtin view spec
> >
> > When you say "builtin view spec," do you mean one that was
generated
> > by running the rspec_scaffold generator?
> >
>
> One that was generated by the rspec_scaffold generator,
Sorry man, it should be working fine. I''m not having that experience
personally - is anybody else?

Zach - what platform are you on?
>
> Zach
>

On 8/16/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > On 8/16/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > > On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > > > I am getting:
> > > >   undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>
> > >
> > > The fact that it says [RSpec example] means that it''s
not the right
> > > class for a view spec. The plugin knows how to create the right
class
> > > based on one of two things:
> > >
> > > If the file is below /spec/views/
> > >
> > > or
> > >
> > > describe "...", :behaviour_type => :view do
> > >   ...
> > >
> > > If the spec you''re running satisfies either of these,
you shouldn''t be
> > > seeing this.
> > >
> > > >
> > > > with my view specs after upgrading to 1.0.8. Is anyone else
seeing this?
> > > >
> > > > To isolate the message I did the following setup:
> > > >
> > > >  - create new rails project (with edge)
> > > >  - installed rspec and rspec_on_rails
> > > >  - ran rake spec
> > > >  - ran script/generate rspec_scaffold products
> > > >  - tried to run a builtin view spec
> > >
> > > When you say "builtin view spec," do you mean one that
was generated
> > > by running the rspec_scaffold generator?
> > >
> >
> > One that was generated by the rspec_scaffold generator,
>
> Sorry man, it should be working fine. I''m not having that
experience
> personally - is anybody else?
>
> Zach - what platform are you on?
>
I tried on OSX and also Kubuntu Feisty. I think the rspec scaffold
generators need to be updated to include  ", :behaviour_type =>
:view".

Zach

On 8/19/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> On 8/16/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > > On 8/16/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> > > > On 8/16/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> > > > > I am getting:
> > > > >   undefined method `mock_model'' for [RSpec
example]:#<Class:0x25550a8>
> > > >
> > > > The fact that it says [RSpec example] means that
it''s not the right
> > > > class for a view spec. The plugin knows how to create the
right class
> > > > based on one of two things:
> > > >
> > > > If the file is below /spec/views/
> > > >
> > > > or
> > > >
> > > > describe "...", :behaviour_type => :view do
> > > >   ...
> > > >
> > > > If the spec you''re running satisfies either of
these, you shouldn''t be
> > > > seeing this.
> > > >
> > > > >
> > > > > with my view specs after upgrading to 1.0.8. Is anyone
else seeing this?
> > > > >
> > > > > To isolate the message I did the following setup:
> > > > >
> > > > >  - create new rails project (with edge)
> > > > >  - installed rspec and rspec_on_rails
> > > > >  - ran rake spec
> > > > >  - ran script/generate rspec_scaffold products
> > > > >  - tried to run a builtin view spec
> > > >
> > > > When you say "builtin view spec," do you mean one
that was generated
> > > > by running the rspec_scaffold generator?
> > > >
> > >
> > > One that was generated by the rspec_scaffold generator,
> >
> > Sorry man, it should be working fine. I''m not having that
experience
> > personally - is anybody else?
> >
> > Zach - what platform are you on?
> >
>
> I tried on OSX and also Kubuntu Feisty. I think the rspec scaffold
> generators need to be updated to include  ", :behaviour_type =>
> :view".
I''d like to avoid that because it would be extra noise and it
shouldn''t be necessary. The generated specs work absolutely perfectly
for me as/is (I''m on OS X). There have been a couple of bugs related
to regexps and windows paths over time, but they''ve either been
resolved or await more feedback from users in the tracker.

I''m thinking about other ways to be explicit about this (besides the
verbose ":behaviour_type => :view". What if we added methods like:

describe_model
describe_view
describe_controller
describe_helper

Or support the first argument being a Symbol:

describe :model, "Thing" do

I''m not sure I like those, but I kind of dislike :behaviour_type =>
:foo as a default.

WDYT?

David

>
> Zach
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 8/19/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On 8/19/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > On 8/16/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > > On 8/16/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > > > On 8/16/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> > > > > On 8/16/07, Zach Dennis <zach.dennis at
gmail.com> wrote:
> > > > > > I am getting:
> > > > > >   undefined method `mock_model'' for
[RSpec example]:#<Class:0x25550a8>
> > > > >
> > > > > The fact that it says [RSpec example] means that
it''s not the right
> > > > > class for a view spec. The plugin knows how to create
the right class
> > > > > based on one of two things:
> > > > >
> > > > > If the file is below /spec/views/
> > > > >
> > > > > or
> > > > >
> > > > > describe "...", :behaviour_type => :view
do
> > > > >   ...
> > > > >
> > > > > If the spec you''re running satisfies either of
these, you shouldn''t be
> > > > > seeing this.
> > > > >
> > > > > >
> > > > > > with my view specs after upgrading to 1.0.8. Is
anyone else seeing this?
> > > > > >
> > > > > > To isolate the message I did the following setup:
> > > > > >
> > > > > >  - create new rails project (with edge)
> > > > > >  - installed rspec and rspec_on_rails
> > > > > >  - ran rake spec
> > > > > >  - ran script/generate rspec_scaffold products
> > > > > >  - tried to run a builtin view spec
> > > > >
> > > > > When you say "builtin view spec," do you mean
one that was generated
> > > > > by running the rspec_scaffold generator?
> > > > >
> > > >
> > > > One that was generated by the rspec_scaffold generator,
> > >
> > > Sorry man, it should be working fine. I''m not having
that experience
> > > personally - is anybody else?
> > >
> > > Zach - what platform are you on?
> > >
> >
> > I tried on OSX and also Kubuntu Feisty. I think the rspec scaffold
> > generators need to be updated to include  ", :behaviour_type
=>
> > :view".
>
> I''d like to avoid that because it would be extra noise and it
> shouldn''t be necessary. The generated specs work absolutely
perfectly
> for me as/is (I''m on OS X). There have been a couple of bugs
related
> to regexps and windows paths over time, but they''ve either been
> resolved or await more feedback from users in the tracker.
>
> I''m thinking about other ways to be explicit about this (besides
the
> verbose ":behaviour_type => :view". What if we added methods
like:
>
> describe_model
> describe_view
> describe_controller
> describe_helper
>
I like these. They are explicit.
> Or support the first argument being a Symbol:
>
> describe :model, "Thing" do
>
> I''m not sure I like those, but I kind of dislike :behaviour_type
=>
> :foo as a default.
>
> WDYT?
>
> David
>
>
> >
> > Zach
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

+1 for those

On 19/08/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> I''m thinking about other ways to be explicit about this (besides
the
> verbose ":behaviour_type => :view". What if we added methods
like:
>
> describe_model
> describe_view
> describe_controller
> describe_helper
-- 
MK

> I''d like to avoid that because it would be extra noise and it
> shouldn''t be necessary. The generated specs work absolutely
perfectly
> for me as/is (I''m on OS X). There have been a couple of bugs
related
> to regexps and windows paths over time, but they''ve either been
> resolved or await more feedback from users in the tracker.
> 
> I''m thinking about other ways to be explicit about this (besides
the
> verbose ":behaviour_type => :view". What if we added methods
like:
> 
> describe_model
> describe_view
> describe_controller
> describe_helper
> 
> Or support the first argument being a Symbol:
> 
> describe :model, "Thing" do
I''m not clear why we need any new syntax at all - isn''t this
just a bug
related to some configurations, if it works for David but not for Zach 
(both on OS X)?

Jay
> 
> I''m not sure I like those, but I kind of dislike :behaviour_type
=>
> :foo as a default.
> 
> WDYT?
> 
> David
> 
> 
>> Zach
>> _______________________________________________
>> rspec-users mailing list
>> rspec-users at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/rspec-users
>>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users

On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org>
wrote:
> > I''d like to avoid that because it would be extra noise and it
> > shouldn''t be necessary. The generated specs work absolutely
perfectly
> > for me as/is (I''m on OS X). There have been a couple of bugs
related
> > to regexps and windows paths over time, but they''ve either
been
> > resolved or await more feedback from users in the tracker.
> >
> > I''m thinking about other ways to be explicit about this
(besides the
> > verbose ":behaviour_type => :view". What if we added
methods like:
> >
> > describe_model
> > describe_view
> > describe_controller
> > describe_helper
> >
> > Or support the first argument being a Symbol:
> >
> > describe :model, "Thing" do
>
> I''m not clear why we need any new syntax at all - isn''t
this just a bug
> related to some configurations, if it works for David but not for Zach
> (both on OS X)?
>
Good point. Let''s figure out why it doesn''t work for Zach
first.

Zach: What''s the relative path of the spec? Did you move it after it
was created?

Aslak
> Jay
> >
> > I''m not sure I like those, but I kind of dislike
:behaviour_type =>
> > :foo as a default.
> >
> > WDYT?
> >
> > David
> >
> >
> >> Zach
> >> _______________________________________________
> >> rspec-users mailing list
> >> rspec-users at rubyforge.org
> >> http://rubyforge.org/mailman/listinfo/rspec-users
> >>
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 8/20/2007 8:20 AM, aslak hellesoy wrote:
> On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
>>> I''d like to avoid that because it would be extra noise and
it
>>> shouldn''t be necessary. The generated specs work
absolutely perfectly
>>> for me as/is (I''m on OS X). There have been a couple of
bugs related
>>> to regexps and windows paths over time, but they''ve either
been
>>> resolved or await more feedback from users in the tracker.
>>>
>>> I''m thinking about other ways to be explicit about this
(besides the
>>> verbose ":behaviour_type => :view". What if we added
methods like:
>>>
>>> describe_model
>>> describe_view
>>> describe_controller
>>> describe_helper
>>>
>>> Or support the first argument being a Symbol:
>>>
>>> describe :model, "Thing" do
>> I''m not clear why we need any new syntax at all -
isn''t this just a bug
>> related to some configurations, if it works for David but not for Zach
>> (both on OS X)?
>>
> 
> Good point. Let''s figure out why it doesn''t work for Zach
first.
> 
> Zach: What''s the relative path of the spec? Did you move it after
it
> was created?
If Zach could create a screencast or precise step-by-step, I can attempt 
to reproduce it tonight on OS X, Windows XP, Cygwin, and/or Ubuntu 6.06.

Jay

aslak hellesoy wrote:
> On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
>>> I''d like to avoid that because it would be extra noise and
it
>>> shouldn''t be necessary. The generated specs work
absolutely perfectly
>>> for me as/is (I''m on OS X). There have been a couple of
bugs related
>>> to regexps and windows paths over time, but they''ve either
been
>>> resolved or await more feedback from users in the tracker.
>>>
>>> I''m thinking about other ways to be explicit about this
(besides the
>>> verbose ":behaviour_type => :view". What if we added
methods like:
>>>
>>> describe_model
>>> describe_view
>>> describe_controller
>>> describe_helper
>>>
>>> Or support the first argument being a Symbol:
>>>
>>> describe :model, "Thing" do
>> I''m not clear why we need any new syntax at all -
isn''t this just a bug
>> related to some configurations, if it works for David but not for Zach
>> (both on OS X)?
>>
> 
> Good point. Let''s figure out why it doesn''t work for Zach
first.
> 
> Zach: What''s the relative path of the spec? Did you move it after
it
> was created?
If Zach could create a screencast or precise step-by-step, I can attempt 
to reproduce it tonight on OS X, Windows XP, Cygwin, and/or Ubuntu 6.06.

Jay

On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org>
wrote:
> aslak hellesoy wrote:
> > On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> >>> I''d like to avoid that because it would be extra
noise and it
> >>> shouldn''t be necessary. The generated specs work
absolutely perfectly
> >>> for me as/is (I''m on OS X). There have been a couple
of bugs related
> >>> to regexps and windows paths over time, but they''ve
either been
> >>> resolved or await more feedback from users in the tracker.
> >>>
> >>> I''m thinking about other ways to be explicit about
this (besides the
> >>> verbose ":behaviour_type => :view". What if we
added methods like:
> >>>
> >>> describe_model
> >>> describe_view
> >>> describe_controller
> >>> describe_helper
> >>>
> >>> Or support the first argument being a Symbol:
> >>>
> >>> describe :model, "Thing" do
> >> I''m not clear why we need any new syntax at all -
isn''t this just a bug
> >> related to some configurations, if it works for David but not for
Zach
> >> (both on OS X)?
> >>
> >
> > Good point. Let''s figure out why it doesn''t work for
Zach first.
> >
> > Zach: What''s the relative path of the spec? Did you move it
after it
> > was created?
>
> If Zach could create a screencast or precise step-by-step, I can attempt
> to reproduce it tonight on OS X, Windows XP, Cygwin, and/or Ubuntu 6.06.
Thanks for offering to do this Jay - that''s great.
>
> Jay
>
>
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

Here is a link to the .mov:

  http://www.continuousthinking.com/movies/RSpec_mockModelError.mov

In doing this I realized something. I only get the error when I try to
run an individual spec within the spec/views/<model_name>/ directory.
Thanks for the suggestion Jay,

Zach



On 8/20/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> > aslak hellesoy wrote:
> > > On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org>
wrote:
> > >>> I''d like to avoid that because it would be extra
noise and it
> > >>> shouldn''t be necessary. The generated specs work
absolutely perfectly
> > >>> for me as/is (I''m on OS X). There have been a
couple of bugs related
> > >>> to regexps and windows paths over time, but
they''ve either been
> > >>> resolved or await more feedback from users in the
tracker.
> > >>>
> > >>> I''m thinking about other ways to be explicit
about this (besides the
> > >>> verbose ":behaviour_type => :view". What if
we added methods like:
> > >>>
> > >>> describe_model
> > >>> describe_view
> > >>> describe_controller
> > >>> describe_helper
> > >>>
> > >>> Or support the first argument being a Symbol:
> > >>>
> > >>> describe :model, "Thing" do
> > >> I''m not clear why we need any new syntax at all -
isn''t this just a bug
> > >> related to some configurations, if it works for David but not
for Zach
> > >> (both on OS X)?
> > >>
> > >
> > > Good point. Let''s figure out why it doesn''t
work for Zach first.
> > >
> > > Zach: What''s the relative path of the spec? Did you move
it after it
> > > was created?
> >
> > If Zach could create a screencast or precise step-by-step, I can
attempt
> > to reproduce it tonight on OS X, Windows XP, Cygwin, and/or Ubuntu
6.06.
>
> Thanks for offering to do this Jay - that''s great.
>
> >
> > Jay
> >
> >
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

Several problems here:

First, rspec_scaffold must be given a *singularised* name, in your
case ''product''. (This is Rails being finicky, not RSpec).

Second, after running rspec_scaffold you must run rake db:migrate

Third, before you can run specs with ruby or spec, you must create the
test database. This can be done with rake spec or rake db:test:prepare

HTH,
Aslak

On 8/21/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> Here is a link to the .mov:
>
>   http://www.continuousthinking.com/movies/RSpec_mockModelError.mov
>
> In doing this I realized something. I only get the error when I try to
> run an individual spec within the spec/views/<model_name>/ directory.
> Thanks for the suggestion Jay,
>
> Zach
>
>
>
> On 8/20/07, David Chelimsky <dchelimsky at gmail.com> wrote:
> > On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org> wrote:
> > > aslak hellesoy wrote:
> > > > On 8/20/07, Jay Levitt <lists-rspec at shopwatch.org>
wrote:
> > > >>> I''d like to avoid that because it would be
extra noise and it
> > > >>> shouldn''t be necessary. The generated specs
work absolutely perfectly
> > > >>> for me as/is (I''m on OS X). There have been
a couple of bugs related
> > > >>> to regexps and windows paths over time, but
they''ve either been
> > > >>> resolved or await more feedback from users in the
tracker.
> > > >>>
> > > >>> I''m thinking about other ways to be
explicit about this (besides the
> > > >>> verbose ":behaviour_type => :view".
What if we added methods like:
> > > >>>
> > > >>> describe_model
> > > >>> describe_view
> > > >>> describe_controller
> > > >>> describe_helper
> > > >>>
> > > >>> Or support the first argument being a Symbol:
> > > >>>
> > > >>> describe :model, "Thing" do
> > > >> I''m not clear why we need any new syntax at all
- isn''t this just a bug
> > > >> related to some configurations, if it works for David
but not for Zach
> > > >> (both on OS X)?
> > > >>
> > > >
> > > > Good point. Let''s figure out why it
doesn''t work for Zach first.
> > > >
> > > > Zach: What''s the relative path of the spec? Did you
move it after it
> > > > was created?
> > >
> > > If Zach could create a screencast or precise step-by-step, I can
attempt
> > > to reproduce it tonight on OS X, Windows XP, Cygwin, and/or
Ubuntu 6.06.
> >
> > Thanks for offering to do this Jay - that''s great.
> >
> > >
> > > Jay
> > >
> > >
> > > _______________________________________________
> > > rspec-users mailing list
> > > rspec-users at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/rspec-users
> > >
> > _______________________________________________
> > rspec-users mailing list
> > rspec-users at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/rspec-users
> >
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 8/20/07, aslak hellesoy <aslak.hellesoy at gmail.com>
wrote:
> Several problems here:
>
> First, rspec_scaffold must be given a *singularised* name, in your
> case ''product''. (This is Rails being finicky, not RSpec).
ok
> Second, after running rspec_scaffold you must run rake db:migrate
This I didn''t do, but doing this makes no difference on the mock model
error. This does fix the issue if I run the spec from RAILS_ROOT, but
not if I run the spec from within the spec/views/products directory.
> Third, before you can run specs with ruby or spec, you must create the
> test database. This can be done with rake spec or rake db:test:prepare
The database was already in existence, so I left that part omitted in the video

If I run specs from RAILS_ROOT then everything works, but not if I''m
not in RAILS_ROOT. This is less of an issue I originally thought, but
the directory thing is a minor irritation, although I can make sure to
run specs from the RAILS_ROOT.

thx,

Zach

On 8/20/07, Zach Dennis <zach.dennis at gmail.com>
wrote:
> On 8/20/07, aslak hellesoy <aslak.hellesoy at gmail.com> wrote:
> > Several problems here:
> >
> > First, rspec_scaffold must be given a *singularised* name, in your
> > case ''product''. (This is Rails being finicky, not
RSpec).
>
> ok
>
> > Second, after running rspec_scaffold you must run rake db:migrate
>
> This I didn''t do, but doing this makes no difference on the mock
model
> error. This does fix the issue if I run the spec from RAILS_ROOT, but
> not if I run the spec from within the spec/views/products directory.
Ah - now THAT makes sense. This won''t work on any system at all. RSpec
looks for /spec/views/ in the path to know that it''s a view spec. If
you''re in the view spec directory, it doesn''t get the
information it
needs. Make sense?
> > Third, before you can run specs with ruby or spec, you must create the
> > test database. This can be done with rake spec or rake db:test:prepare
>
> The database was already in existence, so I left that part omitted in the
video
>
> If I run specs from RAILS_ROOT then everything works, but not if
I''m
> not in RAILS_ROOT. This is less of an issue I originally thought, but
> the directory thing is a minor irritation, although I can make sure to
> run specs from the RAILS_ROOT.
We''d have to change how rspec figures out what behaviour_type to use
to reduce the irritation. Any suggestions?
>
> thx,
>
> Zach
> _______________________________________________
> rspec-users mailing list
> rspec-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/rspec-users
>

On 8/20/07, David Chelimsky <dchelimsky at gmail.com>
wrote:
> On 8/20/07, Zach Dennis <zach.dennis at gmail.com> wrote:
> > On 8/20/07, aslak hellesoy <aslak.hellesoy at gmail.com> wrote:
> > > Several problems here:
> > >
> > > First, rspec_scaffold must be given a *singularised* name, in
your
> > > case ''product''. (This is Rails being finicky,
not RSpec).
> >
> > ok
> >
> > > Second, after running rspec_scaffold you must run rake db:migrate
> >
> > This I didn''t do, but doing this makes no difference on the
mock model
> > error. This does fix the issue if I run the spec from RAILS_ROOT, but
> > not if I run the spec from within the spec/views/products directory.
>
> Ah - now THAT makes sense. This won''t work on any system at all.
RSpec
> looks for /spec/views/ in the path to know that it''s a view spec.
If
> you''re in the view spec directory, it doesn''t get the
information it
> needs. Make sense?
>
> > > Third, before you can run specs with ruby or spec, you must
create the
> > > test database. This can be done with rake spec or rake
db:test:prepare
> >
> > The database was already in existence, so I left that part omitted in
the video
> >
> > If I run specs from RAILS_ROOT then everything works, but not if
I''m
> > not in RAILS_ROOT. This is less of an issue I originally thought, but
> > the directory thing is a minor irritation, although I can make sure to
> > run specs from the RAILS_ROOT.
>
> We''d have to change how rspec figures out what behaviour_type to
use
> to reduce the irritation. Any suggestions?
Use expand_path on the current file and see if it is in RAILS_ROOT/specs/views ?

Zach