I have a Rails 3.2 application, and I want to use one database for many clients and one application. So for every model I have created a field called account_id, now I want to add a global scope for filtering the row in the base of the account_id of the logging user(account_id is a session param). So in initialize I have created a file and put these code
module ActiveRecord
# = Active Record Named \Scopes \
module Scoping
module Default
module ClassMethods
def unscoped #:nodoc:
return (block_given? ? relation.scoping { yield } : relation).where(account_id: Thread.current['account_id'].id)
end
def default_scope(scope = {})
scope = Proc.new if block_given?
if scope.kind_of?(Array) || scope.is_a?(Hash)
scope.merge!({:conditions=>{account_id:Thread.current['account_id'].id}})
end
self.default_scopes = default_scopes + [scope]
end
end
end
end
end
If I logged with user account_id=2 all is ok, but if in the same moment I logged on another browser or computer with account_id=3 ...I have many errors and on the log, I have seen that the application use account_id=2 but also account_id=3 at the same time.
Any solution? How I can rewrite default_scope(scope = {})? Other other idea?
Thread.current[] data is not unique per request. I used to have the same problem. By that time I had been using this gem https://github.com/steveklabnik/request_store. Hope it will help or at least give an idea.
Related
I have a class called CachedObject that stores generic serialized objects indexed by a key. I want this class to implement a create_or_update method. If an object is found it will update it, otherwise it will create a new one.
Is there a way to do this in Rails or do I have to write my own method?
Rails 6
Rails 6 added an upsert and upsert_all methods that deliver this functionality.
Model.upsert(column_name: value)
[upsert] It does not instantiate any models nor does it trigger Active Record callbacks or validations.
Rails 5, 4, and 3
Not if you are looking for an "upsert" (where the database executes an update or an insert statement in the same operation) type of statement. Out of the box, Rails and ActiveRecord have no such feature. You can use the upsert gem, however.
Otherwise, you can use: find_or_initialize_by or find_or_create_by, which offer similar functionality, albeit at the cost of an additional database hit, which, in most cases, is hardly an issue at all. So unless you have serious performance concerns, I would not use the gem.
For example, if no user is found with the name "Roger", a new user instance is instantiated with its name set to "Roger".
user = User.where(name: "Roger").first_or_initialize
user.email = "email#example.com"
user.save
Alternatively, you can use find_or_initialize_by.
user = User.find_or_initialize_by(name: "Roger")
In Rails 3.
user = User.find_or_initialize_by_name("Roger")
user.email = "email#example.com"
user.save
You can use a block, but the block only runs if the record is new.
User.where(name: "Roger").first_or_initialize do |user|
# this won't run if a user with name "Roger" is found
user.save
end
User.find_or_initialize_by(name: "Roger") do |user|
# this also won't run if a user with name "Roger" is found
user.save
end
If you want to use a block regardless of the record's persistence, use tap on the result:
User.where(name: "Roger").first_or_initialize.tap do |user|
user.email = "email#example.com"
user.save
end
In Rails 4 you can add to a specific model:
def self.update_or_create(attributes)
assign_or_new(attributes).save
end
def self.assign_or_new(attributes)
obj = first || new
obj.assign_attributes(attributes)
obj
end
and use it like
User.where(email: "a#b.com").update_or_create(name: "Mr A Bbb")
Or if you'd prefer to add these methods to all models put in an initializer:
module ActiveRecordExtras
module Relation
extend ActiveSupport::Concern
module ClassMethods
def update_or_create(attributes)
assign_or_new(attributes).save
end
def update_or_create!(attributes)
assign_or_new(attributes).save!
end
def assign_or_new(attributes)
obj = first || new
obj.assign_attributes(attributes)
obj
end
end
end
end
ActiveRecord::Base.send :include, ActiveRecordExtras::Relation
The magic you have been looking for has been added in Rails 6
Now you can upsert (update or insert).
For single record use:
Model.upsert(column_name: value)
For multiple records use upsert_all :
Model.upsert_all(column_name: value, unique_by: :column_name)
Note:
Both methods do not trigger Active Record callbacks or validations
unique_by => PostgreSQL and SQLite only
Add this to your model:
def self.update_or_create_by(args, attributes)
obj = self.find_or_create_by(args)
obj.update(attributes)
return obj
end
With that, you can:
User.update_or_create_by({name: 'Joe'}, attributes)
Old question but throwing my solution into the ring for completeness.
I needed this when I needed a specific find but a different create if it doesn't exist.
def self.find_by_or_create_with(args, attributes) # READ CAREFULLY! args for finding, attributes for creating!
obj = self.find_or_initialize_by(args)
return obj if obj.persisted?
return obj if obj.update_attributes(attributes)
end
By chaining find_or_initialize_by and update, this can be achieved in a simple way which avoids the (in my experience, often) unwanted caveats of upsert, and also minimises database calls.
For example:
Class.find_or_initialize_by(
key: "foo",
...
).update(
new_attribute: "bar",
...
)
will return you newly created or updated object.
It is worth noting that if your find_or_initialize_by attributes match multiple Class instances, only the 'first' one will be selected and updated.
You can do it in one statement like this:
CachedObject.where(key: "the given key").first_or_create! do |cached|
cached.attribute1 = 'attribute value'
cached.attribute2 = 'attribute value'
end
The sequel gem adds an update_or_create method which seems to do what you're looking for.
I'm trying to monkey patch ActiveRecord::FinderMethods in order to use hashed ids for my models. So for example User.find(1) becomes User.find("FEW"). Sadly my overwritten method doesn't get called. Any ideas how to overwrite the find_one method?
module ActiveRecord
module FinderMethods
alias_method :orig_find_one, :find_one
def find_one(id)
if id.is_a?(String)
orig_find_one decrypt_id(id)
else
orig_find_one(id)
end
end
end
end
Here's an article that discusses how to actually do what you want by overriding the User.primary_key method like:
class User
self.primary_key = 'hashed_id'
end
Which would allow you to call User.find and pass it the "hashed_id":
http://ruby-journal.com/how-to-override-default-primary-key-id-in-rails/
So, it's possible.
That said, I would recommend against doing that, and instead using something like User.find_by_hashed_id. The only difference is that this method will return nil when a result is not found instead of throwing an ActiveRecord::RecordNotFound exception. You could throw this manually in your controller:
def show
#user = User.find_by_hashed_id(hashed_id)
raise ActiveRecord::RecordNotFound.new if #user.nil?
... continue processing ...
end
Finally, one other note to make this easier on you -- Rails also has a method you can override in your model, to_param, to tell it what property to use when generating routes. By default, of course, it users the id, but you would probably want to use the hashed_id.
class User
def to_param
self.hashed_id
end
end
Now, in your controller, params[:id] will contain the hashed_id instead of the id.
def show
#user = User.find_by_hashed_id(params[:id])
raise ActiveRecord::RecordNotFound.new if #user.nil?
... continue processing ...
end
I agree that you should be careful when doing this, but it is possible.
If you have a method decode_id that converts a hashed ID back to the original id, then the following will work:
In User.rb
# Extend AR find method to allow finding records by an encoded string id:
def self.find(*ids)
return super if ids.length > 1
# Note the short-circuiting || to fall-back to default behavior
find_by(id: decode_id(ids[0])) || super
end
Just make sure that decode_id returns nil if it's passed an invalid hash. This way you can find by Hashed ID and standard ID, so if you had a user with id 12345, then the following:
User.find(12345)
User.find("12345")
User.find(encode_id(12345))
Should all return the same user.
I have a rails 4.2 multi-tenant app using the Apartment gem which has been awesome.
Each company has their own subdomain. I'm using a custom "elevator" which looks at the full request host to determine which "Tenant" should be loaded.
When I create a new company I have an after_create hook to create the new tenant with the proper request host.
This always seems to require a restart of the server both in development and production otherwise I get a Tenant Not Found error.
It's using sqlite in development and postgres in production.
Do I really have to restart the server each time I create a new tenant? Is there an automated way to do this? Maybe just reloading the initializer will work, but I'm not sure how to do that/if that's possible?
I have been messing around with this for a month and haven't been able to find a solution that works. Please help!
initializers/apartment.rb
require 'apartment/elevators/host_hash'
config.tenant_names = lambda { Company.pluck :request_host }
Rails.application.config.middleware.use 'Apartment::Elevators::HostHash', Company.full_hosts_hash
initializers/host_hash.rb
require 'apartment/elevators/generic'
module Apartment
module Elevators
class HostHash < Generic
def initialize(app, hash = {}, processor = nil)
super app, processor
#hash = hash
end
def parse_tenant_name(request)
if request.host.split('.').first == "www"
nil
else
raise TenantNotFound,
"Cannot find tenant for host #{request.host}" unless #hash.has_key?(request.host)
#hash[request.host]
end
end
end
end
end
Company Model
after_create :create_tenant
def self.full_hosts_hash
Company.all.inject(Hash.new) do |hash, company|
hash[company.request_host] = company.request_host
hash
end
end
private
def create_tenant
Apartment::Tenant.create(request_host)
end
What ended up working
I changed the elevator configuration to get away from the HostHash one that's in the apartment gem and used a completely custom one. Mostly based off of an issue on the apartment gem github: https://github.com/influitive/apartment/issues/280
initializers/apartment.rb
Rails.application.config.middleware.use 'BaseSite::BaseElevator'
app/middleware/base_site.rb
require 'apartment/elevators/generic'
module BaseSite
class BaseElevator < Apartment::Elevators::Generic
def parse_tenant_name(request)
company = Company.find_by_request_host(request.host)
return company.request_host unless company.nil?
fail StandardError, "No website found at #{request.host} not found"
end
end
end
I think the problem could be that your host_hash.rb lives in the initializers directory. Shouldn't it be in a folder called "middleware"?, as per the Apartment gem ReadME you referenced in your comment. In that example they used app/middleware/my_custom_elevator.rb. Perhaps yours might look like app/middleware/host_hash.rb?
Right now the file is in initializers, so it's loading from there. But your apartment.rb references it by Rails.application.config.middleware.use. Just a hunch but in addition to loading it initially, it may be looking for it in a nonexistent middleware folder. I'd go ahead and create app/middleware, put the file in there instead, and see what happens. Not sure but you might need to alter require paths too.
Let us know if that helps.
I have a class called CachedObject that stores generic serialized objects indexed by a key. I want this class to implement a create_or_update method. If an object is found it will update it, otherwise it will create a new one.
Is there a way to do this in Rails or do I have to write my own method?
Rails 6
Rails 6 added an upsert and upsert_all methods that deliver this functionality.
Model.upsert(column_name: value)
[upsert] It does not instantiate any models nor does it trigger Active Record callbacks or validations.
Rails 5, 4, and 3
Not if you are looking for an "upsert" (where the database executes an update or an insert statement in the same operation) type of statement. Out of the box, Rails and ActiveRecord have no such feature. You can use the upsert gem, however.
Otherwise, you can use: find_or_initialize_by or find_or_create_by, which offer similar functionality, albeit at the cost of an additional database hit, which, in most cases, is hardly an issue at all. So unless you have serious performance concerns, I would not use the gem.
For example, if no user is found with the name "Roger", a new user instance is instantiated with its name set to "Roger".
user = User.where(name: "Roger").first_or_initialize
user.email = "email#example.com"
user.save
Alternatively, you can use find_or_initialize_by.
user = User.find_or_initialize_by(name: "Roger")
In Rails 3.
user = User.find_or_initialize_by_name("Roger")
user.email = "email#example.com"
user.save
You can use a block, but the block only runs if the record is new.
User.where(name: "Roger").first_or_initialize do |user|
# this won't run if a user with name "Roger" is found
user.save
end
User.find_or_initialize_by(name: "Roger") do |user|
# this also won't run if a user with name "Roger" is found
user.save
end
If you want to use a block regardless of the record's persistence, use tap on the result:
User.where(name: "Roger").first_or_initialize.tap do |user|
user.email = "email#example.com"
user.save
end
In Rails 4 you can add to a specific model:
def self.update_or_create(attributes)
assign_or_new(attributes).save
end
def self.assign_or_new(attributes)
obj = first || new
obj.assign_attributes(attributes)
obj
end
and use it like
User.where(email: "a#b.com").update_or_create(name: "Mr A Bbb")
Or if you'd prefer to add these methods to all models put in an initializer:
module ActiveRecordExtras
module Relation
extend ActiveSupport::Concern
module ClassMethods
def update_or_create(attributes)
assign_or_new(attributes).save
end
def update_or_create!(attributes)
assign_or_new(attributes).save!
end
def assign_or_new(attributes)
obj = first || new
obj.assign_attributes(attributes)
obj
end
end
end
end
ActiveRecord::Base.send :include, ActiveRecordExtras::Relation
The magic you have been looking for has been added in Rails 6
Now you can upsert (update or insert).
For single record use:
Model.upsert(column_name: value)
For multiple records use upsert_all :
Model.upsert_all(column_name: value, unique_by: :column_name)
Note:
Both methods do not trigger Active Record callbacks or validations
unique_by => PostgreSQL and SQLite only
Add this to your model:
def self.update_or_create_by(args, attributes)
obj = self.find_or_create_by(args)
obj.update(attributes)
return obj
end
With that, you can:
User.update_or_create_by({name: 'Joe'}, attributes)
Old question but throwing my solution into the ring for completeness.
I needed this when I needed a specific find but a different create if it doesn't exist.
def self.find_by_or_create_with(args, attributes) # READ CAREFULLY! args for finding, attributes for creating!
obj = self.find_or_initialize_by(args)
return obj if obj.persisted?
return obj if obj.update_attributes(attributes)
end
By chaining find_or_initialize_by and update, this can be achieved in a simple way which avoids the (in my experience, often) unwanted caveats of upsert, and also minimises database calls.
For example:
Class.find_or_initialize_by(
key: "foo",
...
).update(
new_attribute: "bar",
...
)
will return you newly created or updated object.
It is worth noting that if your find_or_initialize_by attributes match multiple Class instances, only the 'first' one will be selected and updated.
You can do it in one statement like this:
CachedObject.where(key: "the given key").first_or_create! do |cached|
cached.attribute1 = 'attribute value'
cached.attribute2 = 'attribute value'
end
The sequel gem adds an update_or_create method which seems to do what you're looking for.
I'm trying to sort out the routing for a multinational web store which only has a presence in certain countries. Each store is different and has a different catalogue of products and there is also a world-wide store for all other countries. I have set up Nginx to prepend the two letter country code from an lookup of the user's IP address so that my Rails app can figure which store to direct the visitor to. I then use Sven Fuch's excellent routing-filter to capture that code, do a lookup against a global SHOPS object and if a match is found then the country code is used, otherwise they get the default (world-wide) store. My routing filter currently looks like this:
module RoutingFilter
class Country < Filter
countries_pattern ||= %r(^/(?i)([a-zA-Z]{2})(?=/|$))
def around_recognize(path, env, &block)
country = "#{extract_segment!(countries_pattern, path)}".upcase
yield(path, env).tap do |params|
params[:shop] = SHOPS.fetch(country.to_sym) || DEFAULT_SHOP
end
end
def around_generate(params, &block)
puts params
shop = params.delete(:shop)
yield.tap do |result|
prepend_segment!(result, shop[:country_code]) if shop
end
end
end
end
Now the curious thing is, the params collection does not contain a :shop param when the around_generate method is executed. My code is directly based on the pagination filter included with the routing-filter gem (I'm not using the locale filter as each of these stores is also multilingual - i18n is handled using accept-language header instead). The original pagination filter by Sven Fuchs looks like this:
module RoutingFilter
class Pagination < Filter
PAGINATION_SEGMENT = %r(/page/([\d]+)/?$)
def around_recognize(path, env, &block)
page = extract_segment!(PAGINATION_SEGMENT, path)
yield(path, env).tap do |params|
params[:page] = page.to_i if page
end
end
def around_generate(params, &block)
page = params.delete(:page)
yield.tap do |result|
append_segment!(result, "page/#{page}") if append_page?(page)
end
end
protected
def append_page?(page)
page && page.to_i != 1
end
end
end
In my filter, shop = params.delete(:shop) results in a Nil object error and I can see from "puts params" that it is indeed not present. Does anyone have any suggestions as to why I'm unable to store and retrieve the :shop param?
Edit: I should mention that I have checked that the parameter gets set correctly in around_recognize - a "puts" of the params collection here does indeed contain the correct :shop object.
If you don't pass a :shop param to url_for (or whatever url generation helper you use here) it won't be passed to around_generate either. around_generate wraps around the url generation part of the routing system.
The Pagination filter assumes the same, e.g. it would be called like blog_posts_path(:page => 2).
But maybe that's not what you want. If you have a look at the Locale filter then this assumes that you sometimes pass a locale to the url helper but sometimes you don't. If :locale is not given it will look it up from I18n.locale which is the current locale set for this request. Maybe you want something similar here?
You could also have a look at the controller's default_url_options. IIRC you can set a default option here, too, so maybe this could work for you. I'm not using this approach anywhere though, so I'm just guessing.
HTH