I have code which is updating a model's property then calling save!. A Rails.logger.info call shows that the model thinks it has the new values. But the SQL write performed by the save! call is writing the old value to the database.
At first it wasn't writing anything to the database at all when I called save!. I thought it was that the object wasn't thinking its value had changed for some reason: changed? returned false, so I used a _will_change! notification to force a write. But now it is doing a write, but with the old values.
This doesn't happen from the "rails console" command line: there I'm able to update the property and it will return changed? of true, and let me save successfully.
Excerpt from the server log follows. Note that the object thinks it has log_ids of '1234,5678,1137', but writes to the database '1234,5678'.
current log ids are [1234, 5678]
new log ids are [1234, 5678, 1137]; writing log_ids of '1234,5678,1137' to NewsList 13 with dirty true
SQL (2.0ms) UPDATE "news_lists" SET "log_ids" = '1234,5678', "updated_at" = '2012-01-02 02:12:17.612283' WHERE ("news_lists"."id" = 13)
The object property in question is log_ids, which is a string containing several IDs of another kind of object.
The source code that produced the output above:
def add_log(new_log)
new_ids = get_log_ids
Rails.logger.info("current log ids are #{new_ids}")
if new_ids.length >= NewsList.MAX_LENGTH
new_ids.shift
end
log_ids_will_change!
new_ids.push new_log.id
log_ids = new_ids.join ","
Rails.logger.info("new log ids are #{new_ids}; writing log_ids of '#{log_ids}' to NewsList #{id} with dirty #{changed?}")
save!
end
def get_log_ids
if log_ids
log_ids.split(",").map &:to_i
else
[]
end
end
Can anyone suggest what might be going on here?
Add the self to self.log_ids = new_ids.join "," otherwise you will just be assigning to the local variable (namesake) instead of the db-persisted attribute (column).
Related
I'm trying to get all Stripe::BalanceTransaction except those they are already in my JsonStripeEvent
What I did =>
def perform(*args)
last_recorded_txt = REDIS.get('last_recorded_stripe_txn_last')
txns = Stripe::BalanceTransaction.all(limit: 100, expand: ['data.source', 'data.source.application_fee'], ending_before: last_recorded_txt)
REDIS.set('last_recorded_stripe_txn_last', txns.data[0].id) unless txns.data.empty?
txns.auto_paging_each do |txn|
if txn.type.eql?('charge') || txn.type.eql?('payment')
begin
JsonStripeEvent.create(data: txn.to_json)
rescue StandardError => e
Rails.logger.error "Error while saving data from stripe #{e}"
REDIS.set('last_recorded_stripe_txn_last', txn.id)
break
end
end
end
end
But It doesnt get the new one from the API.
Can anyone could help me for this ? :)
Thanks
I think it's because the way auto_paging_each works is almost opposite to what you expect :)
As you can see from its source, auto_paging_each calls Stripe::ListObject#next_page, which is implemented as follows:
def next_page(params={}, opts={})
return self.class.empty_list(opts) if !has_more
last_id = data.last.id
params = filters.merge({
:starting_after => last_id,
}).merge(params)
list(params, opts)
end
It simply takes the last (already fetched) item and adds its id as the starting_after filter.
So what happens:
You fetch 100 "latest" (let's say) records, ordered by descending date (default order for BalanceTransaction API according to Stripe docs)
When you call auto_paging_each on this dataset then, it takes the last record, adds its id as the
starting_after filter and repeats the query.
The repeated query returns nothing because there are noting newer (starting later) than the set you initially fetched.
As far as there are no more newer items available, the iteration stops after the first step
What you could do here:
First of all, ensure that my hypothesis is correct :) - put the breakpoint(s) inside Stripe::ListObject and check. Then 1) rewrite your code to use starting_after traversing logic instead of ending_before - it should work fine with auto_paging_each then - or 2) rewrite your code to control the fetching order manually.
Personally, I'd vote for (2): for me slightly more verbose (probably), but straightforward and "visible" control flow is better than poorly documented magic.
This code just displays the values inside the array model.request_reports
To get the most recent, I have to loop through and compare the current
report.updated_at with the last saved report.update_at value. One thing to find
out is what class the update_at field is and how to compare them against each other. The class is ActiveSupport::TimeZone
I need to keep track of the array index of the report that has the most recent updated_at as I loop so that I can access it after the loop.
The problem is, I don't know how to do this:
msg = ""
reports_arr = model.request_reports
reports_arr.each do |report|
updated_at = report.updated_at
if updated_at
msg = msg + "#{updated_at} --- "
msg = msg + "#{updated_at.class}---"
end
end
msg
To add to #meagar comment. You should be using the DB to do sorts on tables.
With that said we need to know what DB you are using as the exact command differs for each.
Mongo w/ Mongoid would be Model.order_by(:updated_at => 'desc').first
My loop had to go through the array and check by greatest date value because in the system Im using, it automatically sorts the reports array by the field "due_at" which is not the reports most recent updated record. Code below works for me.
msg = ""
reports_arr = model.request_reports
last_modified_report = model.last_modified_report
recent = nil
recent_report = nil
reports_arr.each_with_index do |report,index|
updated_at = report.updated_at
if index == 0
recent = updated_at
recent_report = report
end
if updated_at > recent
recent = updated_at
recent_report = report
end
last_modified_report = recent_report
end
msg = msg + "#{recent}---"
msg = msg + "#{recent_report}---"
msg = msg + "#{last_modified_report}"
model.last_modified_report = last_modified_report
model.save(validate: false)
msg
The OP's answer is only good if you absolutely cannot query the database for the info you want directly. I assume you only want the index so you can find the most recent one?
Even if automatic sorting is on one column, your query for the data can have it sorted on a different column.
model.request_reports.order_by(:updated_at => 'desc').first
If you have a default scope that's messing with your query, you can ask for an unscoped list, although I doubt a default ordering would cause any trouble.
model.unscoped.order_by(:updated_at => 'desc').first
You can string together queries that are already written: that can be useful even if request_reports is a query or scope you have somewhere.
It will be way less expensive than getting everything, and looping through it - you are always better off finding a way to get just the info you need in a db query if you can.
Hi I've this piece of code
class Place < ActiveRecord::Base
def self.find_or_create_by_latlon(lat, lon)
place_id = call_external_webapi
result = Place.where(:place_id => place_id).limit(1)
result = Place.create(:place_id => place_id, ... ) if result.empty? #!
result
end
end
Then I'd like to do in another model or controller
p = Post.new
p.place = Place.find_or_create_by_latlon(XXXXX, YYYYY) # race-condition
p.save
But Place.find_or_create_by_latlon takes too much time to get the data if the action executed is create and sometimes in production p.place is nil.
How can I force to wait for the response before execute p.save ?
thanks for your advices
You're right that this is a race condition and it can often be triggered by people who double click submit buttons on forms. What you might do is loop back if you encounter an error.
result = Place.find_by_place_id(...) ||
Place.create(...) ||
Place.find_by_place_id(...)
There are more elegant ways of doing this, but the basic method is here.
I had to deal with a similar problem. In our backend a user is is created from a token if the user doesn't exist. AFTER a user record is already created, a slow API call gets sent to update the users information.
def self.find_or_create_by_facebook_id(facebook_id)
User.find_by_facebook_id(facebook_id) || User.create(facebook_id: facebook_id)
rescue ActiveRecord::RecordNotUnique => e
User.find_by_facebook_id(facebook_id)
end
def self.find_by_token(token)
facebook_id = get_facebook_id_from_token(token)
user = User.find_or_create_by_facebook_id(facebook_id)
if user.unregistered?
user.update_profile_from_facebook
user.mark_as_registered
user.save
end
return user
end
The step of the strategy is to first remove the slow API call (in my case update_profile_from_facebook) from the create method. Because the operation takes so long, you are significantly increasing the chance of duplicate insert operations when you include the operation as part of the call to create.
The second step is to add a unique constraint to your database column to ensure duplicates aren't created.
The final step is to create a function that will catch the RecordNotUnique exception in the rare case where duplicate insert operations are sent to the database.
This may not be the most elegant solution but it worked for us.
I hit this inside a sidekick job that retries and gets the error repeatedly and eventually clears itself. The best explanation I've found is on a blog post here. The gist is that postgres keeps an internally stored value for incrementing the primary key that gets messed up somehow. This rings true for me because I'm setting the primary key and not just using an incremented value so that's likely how this cropped up. The solution from the comments in the link above appears to be to call ActiveRecord::Base.connection.reset_pk_sequence!(table_name) This cleared up the issue for me.
begin
result = Place.where(:place_id => place_id).limit(1)
result = Place.create(:place_id => place_id, ... ) if result.empty? #!
rescue ActiveRecord::StatementInvalid => error
#save_retry_count = (#save_retry_count || 1)
ActiveRecord::Base.connection.reset_pk_sequence!(:place)
retry if( (#save_retry_count -= 1) >= 0 )
raise error
end
I am trying to write a simple function to clean a filename string and update the object. When I save a test string it works, but when I try to save the string variable I've created, nothing happens. But when I return the string, the output seems to be correct! What am I missing?
def clean_filename
clean_name = filename
clean_name.gsub! /^.*(\\|\/)/, ''
clean_name.gsub! /[^A-Za-z0-9\.\-]/, '_'
clean_name.gsub!(/\_+/, ' ')
#update_attribute(:filename, "test") #<-- correctly sets filename to test
#update_attribute(:filename, clean_name) #<-- no effect????? WTF
#return clean_name <-- seems to returns the correct string
end
Thank you very much.
Is the update only going through if the object ID has changed? I think it is reasonable to update the slot only when the object itself has changed.
Have you ever tried to use gsub instead of gsub!, so that the object ID changes?
I have a stored proc on an existing 3rd party application (SQL 2005) that I wish to interact with.
It is an insert statement followed by a select statement as follows;
Set #CustomerId = Cast(SCOPE_IDENTITY() As [int])
Select #CustomerId
Using VB6 how do I access the value of #CustomerID?
set rs = cmd.Execute
is not returning a resultset as expected...
[Edit]
rs.Fields.Count is 0.
Any attempt to access the resulting recordset, like rs(0).Value simply causes an "Item not found..." error.
I would guess that your stored procedure is returning more than one recordset.
If this is the case, you can use the NextRecordset() method to iterate through them.
MSDN:
If a row-returning command executes successfully but returns no records,
the returned Recordset object will be
open but empty. Test for this case by
verifying that the BOF and EOF
properties are both True.
If a non–row-returning command executes successfully, the returned
Recordset object will be closed, which
you can verify by testing the State
property on the Recordset.
When there are no more results, recordset will be set to Nothing.
This means I would suggest something like this to solve your problem:
Set rs = cmd.Execute
''# fetch the first value of the last recordset
Do Until rs Is Nothing
If rs.State = adStateOpen Then
If Not (rs.BOF And rs.EOF) Then
''# You can do a different sanity check here, or none at all
If rs.Fields(0).Type = adInteger Then
CustomerId = rs.Fields(0).Value
End If
End If
End If
Set rs = rs.NextRecordSet
Loop
MsgBox CustomerId