Question

this is my method

def add
    @result
    @log = Log.new(params[:log])
    if @log.save
      result = "success";
    else
      result = "fail";
    end
    render :json => result
end

and this is my routes.rb

match "/logs/add/:value/:category/:note/:own" => "logs#add"

when i try putting this URL:

http://localhost:3000/logs/add/338/testcat/testnote/testown

and it returns the json correctly and new item was added to the database but
all of the fields(value,category,note,own) are null.

please help 🙁

---

I’ve already solved the above problem, but

What if i want to create multiple objects by sending from 1 url to my ruby.
Ex:

localhost:3000/logs/add/338/testcat/testnote/testown

The above request will create only 1 log to my database
If I want to create many logs by using the above solution, I have to send one url for one log.
I want my ruby to be able to create multiple logs from 1 url.
I mean that I can create many logs by sending only one url to the server.
What routing,method,url should be?
(I heard something about “url/add/param1&&param2” ?

Answer 1

You’re calling Log.new(params[:log]) but there isn’t a log parameter in the request.

You can fix this by updating your #new call:

@log = Log.new({
    :value => params[:value],
    :category => params[:category],
    :note => params[:note],
    :own => params[:own]
})

but bear in mind that you will then have a GET request which changes your database state. This is potentially a security flaw, because a potential attacker could get you to update your database just by following a link (from their website, in an email etc). This could even be a problem without a malicious attacker – Google will happily follow GET links – and even if you hide your site behind a login, it’s not impossible to think of a situation where Chrome may try and speed up your browsing session by “intelligently” preloading a link it spots in the page before you click it.

Generally all of your data-changing actions should be limited to POST requests