I ran into something really puzzling today. I don''t know exactly why
it''s happening, but I thought I''d share my experiences.
I started a dummy app just to test an idea outside of the context of
my real work today. So, I quickly get things started:
* rails junk
* cd junk
* ruby script/plugin install svn://rubyforge.org/var/svn/rspec/tags/
CURRENT/rspec
* ruby script/plugin install svn://rubyforge.org/var/svn/rspec/tags/
CURRENT/rspec_on_rails
* script/generate rspec
* script/generate rspec_scaffold Junk name:string
* script/generate rspec
* cp ../mrm/config/database.yml config/ (just a good database script
that tells me where I keep my SQLite databases)
* rake db:migrate
* autotest
That gives me an error on @junk = Junk.new in the setup of my
scaffolded Junk model. I wrestled with it for a while, and couldn''t
get around it. I could get the test to pass with rake spec, which
uses the other runner, but not with autotest or script/spec, which I
believe use the same runner.
So I scraped the whole app and started from scratch, this time
generating rspec before I generated the scaffold. Turns out this was
the trick. Now, why the funny problem, I don''t know. I guess the
generator needs to see something different while it generates, but I
can''t imagine what it is. The spec_helper.rb and the spec.opts are
the same, I believe. (I checked that between my working app and non-
working app and didn''t see a difference.) There are the previous-
failures.txt, which I can''t imagine would affect anything. Then
there''s script/spec and script/spec_server. I''m sure it has
something to do with the scaffold generator talking to them somehow,
but since the generated spec is identical, I don''t know.
Anyway, I''m stumped. If anyone understands this, great. It''s
one
less mystery at the end of the day. If not, I''ll just remember to do
my incantations in the correct order and post this in case someone
else runs into the same problem.
Cheers,
David
