Gabriel Sobrinho comments

Results 179 comments of


                                            Gabriel Sobrinho

Respect prereqs for each database on rake db:migrate

Not stale

Avoid second query on in_batches.pluck

Nope, same behavior.

Avoid second query on in_batches.pluck

But that will instantiate the records into AR objects wasting memory, no?

Avoid second query on in_batches.pluck

Why not `yield yielded_relation, batch_relation` so I can get the ids or whatever? :)

Avoid second query on in_batches.pluck

Oh, I see, the pluck would still be executed. Maybe we can work with the BatchEnumerator to add a special case just for pluck?

Avoid second query on in_batches.pluck

We might use a strategy where we decide what to yield like (pseudo): ```ruby if load == :relation yielded_relation = batch_relation elsif load records = batch_relation.records ids = records.map(&:id) yielded_relation...

Avoid second query on in_batches.pluck

Yeah, too much effort for other columns and in that case probably worths to make an extra SQL. Since in my case I'm looking for the primary key anyways, your...

Avoid second query on in_batches.pluck

I'm guessing we could do `ids = batch_relation.pluck(primary_key, *extra_pluck_columns)` and yield along with the ids but too hacky.

Avoid second query on in_batches.pluck

If it's about the primary key, currently can be worked around with: ```ruby Model.in_batches do |batch| ids = batch.where_values_hash[batch.klass.primary_key] end ``` I'm using this with success so far.

'(not showing binary data)' when the content-type is text/html

Found it: ``` [2] pry(#)> puts response_headers HTTP/2 404 date: Mon, 22 Jan 2024 17:29:28 GMT content-type: text/html ``` ``` [6] pry(#)> response_headers.scan(/Content-Type: (\S+(; charset=\S+)?)/).flatten.first => nil [7] pry(#)> response_headers.scan(/Content-Type:...