Forum https://discourse.hanamirb.org – Code of Conduct http://hanamirb.org/community/#code-of-conduct
Hello Good People!
We would like to incorporate real-time server to browser communication into our Web application which is served by a Hanami REST API. Is that possible to implement web sockets between our Web Application and our Hanami server? Can we use something similar to what Rails has, i.e. ActionCable?
Thank you!
Hello Good People,
I am learing Hanami Validations and trying to build a custom predicate. In fact, I would like to create a rule like this:
rule(check_dates: [:check_in, :check_out]) do |check_in, check_out|
check_in < check_out
endThis fails with:
< predicate arity is invalidwhich makes sense since the documentation says that we cann’t write arbitrary Ruby inside the predicate block.
But, how can I implement such a rule? what is the best practice for this case?
Thank you very much in advance.
|current, other| to |other, current| and get away with it?
Hello Good People,
I am implementing an API on Hanami and when the API is called from a browser, the browser does a pre-flight check and sends an OPTIONS request to the end-point. I am thinking about adding the following entry to my config/routes.rb file:
options '/*', to: ->(_) { [200, { 'Allow' => 'OPTIONS, HEAD, GET, POST, PUT, PATCH, DELETE' }, ['']] }so that all the OPTIONS requests, on any path, to return all the verbs. Is that a good practice? Is there anything else that I should do better?
Thank you very much in advance.
But that would be a lie, wouldn't it? What if some endpoint does not support a http verb? But you claimed it did. I don't think this is a good practice.
Also, why is the browser doing a preflight request? Is this CORS?
Oh, and you should probably implement such things in your reverse proxy. That way is much faster than relying on your ruby app for an answer to OPTIONS requests.
'*' with your domain
How do I set both the primary DB and replica DB (read only) connections in the Hanami.configure?
I want to set up the following site. https://github.com/jeremyevans/sequel/blob/master/doc/sharding.rdoc
Hello Good People,
I am trying to get the Author of a Book within a Book#to_enqueued_item() method like this:
class Book < Hanami::Entity
def to_enqueued_item
bwa = book_with_author
{
title: title,
isbn: isbn,
author_code: bwa.author.code
}
end
private
def book_with_author
return self unless author.nil?
BookRepository.new.find_with_author(id)
end
endWhere the BookRepository defines a Belongs To association to the Author like in the example here
I wanted to ask whether my #to_enqueued_item() implementation looks good in terms of practice, since the instance of the Book that I may have at hand might not have the author.
thank you in advance.
nil can mean both that the author wasn't loaded yet or that the book does not have an author.Personally I'm a fan of null objects, i.e., implement a
NullAuthor and return it if a book does not have an author. That eliminates many nil checks and you also solve the above issue.
Has anybody done a Hanami project that does not make use of a SQL database, but serialises to and from a structured flat file used by other, third-party code. (Think of a Gemfile.lock file, for example.) After skimming the Hanami::Model and Sequel codebases, I’m left with the impression that I should create bespoke Serialiser and Deserialiser services that consume and emit raw Hashes of Entity objects defined using a manual schema.
While I’ve convinced myself lthat it should be possible to make this work, I haven’t convinced myself lthat this is the best/most “Hanami-like” way to do things. (Better) ideas gratefully welcomed.