Is there a way to get changes applied on an ActiveRecord from the beginning until the end of the transaction? I tried previous_changes, but it just returns changes applied on the last operation on the ActiveRecord. For example:
Account.transaction do
account = Account.create(name: 'test')
account.update(last_name: 'testing')
end
I would like to know that 'name' and 'last_name' has changed in the transaction.
You can inspect ActiveModel::Dirty and just:
Account.transaction do
account = Account.create(name: 'test')
account.set_attributes(last_name: 'testing')
account.previous_changes[:last_name].present?
account.save!
end
Related
Some records in my application have a DOI assigned to them and in that case they should not be deleted. Instead, they should have their description changed and be flagged when a user triggers their deletion. A way to do this, I thought, would be as follows in the relevant model:
before_destroy :destroy_validation
private
def destroy_validation
if metadata['doi'].blank?
# Delete as normal...
nil
else
# This is a JSON field.
modified_metadata = Marshal.load(Marshal.dump(metadata))
description = "Record does not exist anymore: #{name}. The record with identifier content #{doi} was invalid."
modified_metadata['description'] = description
modified_metadata['tombstone'] = true
update_column :metadata, modified_metadata
raise ActiveRecord::RecordNotDestroyed, 'Records with DOIs cannot be deleted'
end
end
This does indeed prevent deletion, but the record appears unchanged afterwards rather than having a modified description. Here's an example of a test:
test "records with dois are not deleted" do
record = Record.new(metadata: metadata)
record.metadata['doi'] = 'this_is_a_doi'
assert record.save
assert_raises(ActiveRecord::RecordNotDestroyed) { record.destroy! }
assert Record.exists?(record.id)
modified_record = Record.find(record.id)
puts "#{record.description}" # This is correctly modified as per the callback code.
puts "#{modified_record.description}" # This is the same as when the record was created.
end
I can only guess that Rails is rolling back the update_column due to an exception having been raised, though I may be mistaken. Is there anything I can do to prevent this?
save and destroy are automatically wrapped in a transaction
https://api.rubyonrails.org/classes/ActiveRecord/Transactions/ClassMethods.html
So destroy fails, transactions is rolled back and you can't see updated column in tests.
You could try with after_rollback callback https://api.rubyonrails.org/classes/ActiveRecord/Transactions/ClassMethods.html#method-i-after_rollback
or do record.destroy check for record.errors, if found update record with method manually record.update_doi if record.errors.any?.
before_destroy :destroyable?
...
def destroyable?
unless metadata['doi'].blank?
errors.add('Doi is not empty.')
throw :abort
end
end
def update_doi
modified_metadata = Marshal.load(Marshal.dump(metadata))
description = "Record does not exist anymore: #{name}. The record with identifier content #{doi} was invalid."
modified_metadata['description'] = description
modified_metadata['tombstone'] = true
update_column :metadata, modified_metadata
end
Tip: use record.reload instead of Record.find(record.id).
This is my class:
class Plan < ActiveRecord::Base
def testing
self.with_lock do
update_columns(lock: true)
byebug
end
end
def testing2
self.lock!
byebug
end
end
I opened two rails consoles.
In first console:
p = Plan.create
=> (basically success)
p.id
=> 12
p.testing2
(byebug) # simulation of halting the execution,
(BYEBUG) # I just leave the rails console open and wait at here. I expect others won't be able to update p because I still got the lock.
On second console:
p = Plan.find(12)
=> (basically said found)
p.name = 'should not be able to be stored in database'
=> "should not be able to be stored in database"
p.save!
=> true # what????? Why can it update my object? It's lock in the other console!
lock! in testing2 doesn't lock while with_lock in testing does work. Can anybody explain why lock! doesn't work?
#lock! uses SELECT … FOR UPDATE to acquire a lock.
According to PostgreSQL doc.
FOR UPDATE causes the rows retrieved by the SELECT statement to be locked as though for update. This prevents them from being locked, modified or deleted by other transactions until the current transaction ends.
You need a transaction to keep holding a lock of a certain row.
Try
console1:
Plan.transaction{Plan.find(12).lock!; sleep 100.days}
console2:
p = Plan.find(12)
p.name = 'should not be able to be stored in database'
p.save
#with_lock acquire a transaction for you, so you don't need explicit transaction.
(This is PostgreSQL document. But I think other databases implement similar logic. )
I have a custom validator that checks if the user has entered the correct SMS code. When the user enters the wrong code I need to log the failed attempt and limit their retries to 3 per code.
I have created the following validator that works however the field is not being incremented.
def token_match
if token != User.find(user_id).verification_token
User.find(user_id).increment!(:verification_fails)
errors.add(:sms_code, "does not match")
end
end
The problem is as soon as I add the error the previous statement is rolled back. If I comment out the errors.add line then the increment works however there is no higher level validation performed.
Change your custom validator to be:
def token_match
if token != User.find(user_id).verification_token
errors.add(:sms_code, "does not match")
end
end
and add in your model after_validation callback to be like this:
after_validation: increase_fails_count
def increase_fails_count
unless self.errors[:sms_code].empty?
user = User.find_by(:id => user_id)
user.increment!(:verification_fails)
user.save
end
end
You can use #update_columns in your validator. It writes directly to db.
u = User.find(user_id)
u.update_columns(verification_fails: u.verification_fails + 1)
This worked for me. But if for some reason it doesn't work for you, maybe you can try running it in a new thread,which creates a new db connection:
Thread.new do
num = User.find(user_id).verification_fails
ActiveRecord::Base.connection_pool.with_connection { |con| con.exec_query("UPDATE users SET verification_fails = #{num} WHERE id = #{user_id}") }
end.join
I am using an after_commit in my application.
I would like it to trigger only when a particular field is updated in my model. Anyone know how to do that?
Old question, but this is one method that I've found that might work with the after_commit callback (working off paukul's answer). At least, the values both persist post-commit in IRB.
after_commit :callback,
if: proc { |record|
record.previous_changes.key?(:attribute) &&
record.previous_changes[:attribute].first != record.previous_changes[:attribute].last
}
Answering this old question because it still pops up in search results
you can use the previous_changes method which returnes a hash of the format:
{ "changed_attribute" => ["old value", "new value"] }
it's what changes was until the record gets actually saved (from active_record/attribute_methods/dirty.rb):
def save(*) #:nodoc:
if status = super
#previously_changed = changes
#changed_attributes.clear
# .... whatever goes here
so in your case you can check for previous_changes.key? "your_attribute" or something like that
Old question but still pops up in search results.
As of Rails 5 attribute_changed? was deprecated. Using saved_change_to_attribute? instead of attribute_changed? is recommended.
I don't think you can do it in after_commit
The after_commit is called after the transaction is commited Rails Transactions
For example in my rails console
> record = MyModel.find(1)
=> #<MyModel id: 1, label: "test", created_at: "2011-08-19 22:57:54", updated_at: "2011-08-19 22:57:54">
> record.label = "Changing text"
=> "Changing text"
> record.label_changed?
=> true
> record.save
=> true
> record.label_changed?
=> false
Therefore you won't be able to use the :if condition on after_commit because the attribute will not be marked as changed anymore as it has been saved. You may need to track whether the field you are after is changed? in another callback before the record is saved?
This is a very old problem, but the accepted previous_changes solution just isn't robust enough. In an ActiveRecord transaction, there are many reasons why you might save a Model twice. previous_changes only reflects the result of the final save. Consider this example
class Test < ActiveRecord::Base
after_commit: :after_commit_test
def :after_commit_test
puts previous_changes.inspect
end
end
test = Test.create(number: 1, title: "1")
test = Test.find(test.id) # to initialize a fresh object
test.transaction do
test.update(number: 2)
test.update(title: "2")
end
which outputs:
{"title"=>["1", "2"], "updated_at"=>[...]}
but, what you need is:
{"title"=>["1", "2"], "number"=>[1, 2], "updated_at"=>[...]}
So, my solution is this:
module TrackSavedChanges
extend ActiveSupport::Concern
included do
# expose the details if consumer wants to do more
attr_reader :saved_changes_history, :saved_changes_unfiltered
after_initialize :reset_saved_changes
after_save :track_saved_changes
end
# on initalize, but useful for fine grain control
def reset_saved_changes
#saved_changes_unfiltered = {}
#saved_changes_history = []
end
# filter out any changes that result in the original value
def saved_changes
#saved_changes_unfiltered.reject { |k,v| v[0] == v[1] }
end
private
# on save
def track_saved_changes
# maintain an array of ActiveModel::Dirty.changes
#saved_changes_history << changes.dup
# accumulate the most recent changes
#saved_changes_history.last.each_pair { |k, v| track_saved_change k, v }
end
# v is an an array of [prev, current]
def track_saved_change(k, v)
if #saved_changes_unfiltered.key? k
#saved_changes_unfiltered[k][1] = track_saved_value v[1]
else
#saved_changes_unfiltered[k] = v.dup
end
end
# type safe dup inspred by http://stackoverflow.com/a/20955038
def track_saved_value(v)
begin
v.dup
rescue TypeError
v
end
end
end
which you can try out here: https://github.com/ccmcbeck/after-commit
It sounds like you want something like a conditional callback. If you had posted some code I could have pointed you in the right direction however I think you would want to use something like this:
after_commit :callback,
:if => Proc.new { |record| record.field_modified? }
Use gem ArTransactionChanges. previous_changes is not working for me in Rails 4.0.x
Usage:
class User < ActiveRecord::Base
include ArTransactionChanges
after_commit :print_transaction_changes
def print_transaction_changes
transaction_changed_attributes.each do |name, old_value|
puts "attribute #{name}: #{old_value.inspect} -> #{send(name).inspect}"
end
end
end
The ActiveRecord find_or_create_by dynamic finder method allows me to specify a block. The documentation isn't clear on this, but it seems that the block only runs in the create case, and not in the find case. In other words, if the record is found, the block doesn't run. I tested it with this console code:
User.find_or_create_by_name("An Existing Name") do |u|
puts "I'M IN THE BLOCK"
end
(nothing was printed). Is there any way to have the block run in both cases?
As far as I understand block will be executed if nothing found. Usecase of it looks like this:
User.find_or_create_by_name("Pedro") do |u|
u.money = 0
u.country = "Mexico"
puts "User is created"
end
If user is not found the it will initialized new User with name "Pedro" and all this stuff inside block and will return new created user. If user exists it will just return this user without executing the block.
Also you can use "block style" other methods like:
User.create do |u|
u.name = "Pedro"
u.money = 1000
end
It will do the same as User.create( :name => "Pedro", :money => 1000 ) but looks little nicer
and
User.find(19) do |u|
..
end
etc
It doesn't seem to me that this question is actually answered so I will. This is the simplest way, I think, you can achieve that:
User.find_or_create_by_name("An Existing Name or Non Existing Name").tap do |u|
puts "I'M IN THE BLOCK REGARDLESS OF THE NAME'S EXISTENCE"
end
Cheers!