I'm writing a migration script which has to, well, actually migrate
all data from table A to table B, but I need to access the application
logic in model A for each row migrated. I could do this:
MyModel.all.each do |instance|
# do stuff...
end
but as far as I know that would pre-instantiate all records in memory
beforehand, which could be *really bad* with the amount of records I'm
dealing with, and I only really need to work with a record at a time.
An alternative I'm currently using is:
execute("SELECT * FROM table_a").each_hash do |row|
instance = MyModel.new(row)
end
but I believe this is also sub-optimal (especially since some
attributes are attr_protected, therefore I'm having to take care of
those manually, and other reasons).
Is there any obvious/elegant way of doing this I'm missing?
I'm writing a migration script which has to, well, actually migrate
all data from table A to table B, but I need to access the application
logic in model A for each row migrated. I could do this:
MyModel.all.each do |instance|
# do stuff...
end
but as far as I know that would pre-instantiate all records in memory
beforehand, which could be *really bad* with the amount of records I'm
dealing with, and I only really need to work with a record at a time.
Hi,
Have you thought of using limit and a offset with a regular find?
(below is the code but I didn't check if it's valid)
current_offset = 0
limit = 100
begin
result = MyModel.find(:all, :limit=>limit, :offset=>current_offset)
#some operations
current_offset += limit
end while result.size == 0