I'm also experiencing a bizarre symptom with `Sequel::Migrator.run(DB,
"db/migrate")` which makes me think that somehow my development environment
is screwed up. But I just can't figure out how to untangle it. When I run
my migration task (which is the same task I've been using for over a year)
on an already-populated database, expecting it to migrate from version 38
to version 39, it doesn't throw any errors, but then the columns that
should have been added with 039_xxx.rb aren't created. However, if I run
the migration task on an empty database, it works just fine. *sigh*
On Wednesday, January 17, 2018 at 10:05:04 PM UTC-7, Barbara Carradini
wrote:
>
> Quick follow-up: I've tried uninstalling/re-installing sequel, sequel_pg,
> pg, guard, and minitest. No luck.
>
>
> On Wednesday, January 17, 2018 at 10:01:45 PM UTC-7, Barbara Carradini
> wrote:
>>
>> I have a large, existing code base for a project that's been on hiatus.
>> Last time I was working on the project, all automated unit tests were
>> running perfectly. Now that I've come back to it, I'm suddenly hitting
>> errors on previously existing dataset declarations that rely on
>> the <table_name>__<column_name> syntax to specify a column in a particular
>> table (to keep things straight when joining tables that share column names).
>>
>>
>> For example, here's one of my models:
>>
>> module DonorSee
>> class Project < Sequel::Model
>> set_dataset dataset.exclude(status:
>> 'banned').order(:projects__created_at).reverse
>> ....
>> end
>> end
>>
>>
>> And here's the error I'm hitting during automation testing (using
>> MiniTest and Guard):
>>
>> Sequel::DatabaseError: Sequel::DatabaseError:
>> PG::UndefinedColumn: ERROR: column "projects__created_at" does not exist
>>
>> LINE 1: ...M "projects" WHERE ("projects"."id" = 4) ORDER BY
>> "projects_...
>>
>>
>> What's even weirder is that I was able to reproduce the same error using
>> Pry, which interacts with my development database as opposed to my test
>> database. If I attempted to use <table_name>__<column_name> syntax, it
>> would tell me "column ... does not exist." But after trying a few things to
>> sort out my dev environment, I no longer see the error when I use Pry, but
>> still see it when running MiniTest.
>>
>> Here's some relevant environment info:
>>
>> Barbaras-MacBook-Pro:donorsee-api barbara$ rbenv version
>>
>> 2.3.1 (set by /Users/barbara/.rbenv/version)
>>
>> Barbaras-MacBook-Pro:donorsee-api barbara$ gem list sequel
>>
>>
>> *** LOCAL GEMS ***
>>
>>
>> sequel (5.4.0, 4.40.0, 4.36.0, 4.35.0, 4.34.0)
>>
>> sequel_pg (1.8.1, 1.6.17)
>>
>> Barbaras-MacBook-Pro:donorsee-api barbara$ gem list pg
>>
>>
>> *** LOCAL GEMS ***
>>
>>
>> pg (1.0.0, 0.21.0, 0.18.4)
>>
>> sequel_pg (1.8.1, 1.6.17)
>>
>>
>>
--
You received this message because you are subscribed to the Google Groups
"sequel-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/sequel-talk.
For more options, visit https://groups.google.com/d/optout.
