get content of local html file with rails - ruby-on-rails

what i need is passing local html file to my form, get content of this file and later parse it.
I have this in view
= form_tag :parser, :html => {:multipart => true} do
= file_field_tag :html_file
= submit_tag
And this in controller
def parser
#file = params[:html_file]
end
It seems like form dont get file, just get the string with name of file. How to fix it?

I wrote this to handle uploads in a Rails 2.3.x app. I cannot remember why I had to split this into an if/elsif/else/end statement, but I must have had an error consolidating it to one if statement.
if %w(File Tempfile ActionController::UploadedTempfile ).include?(params[:html_file].class.to_s)
data = params[:html_file].read
elsif %w(StringIO ActionController::UploadedStringIO).include?(params[:html_file].class.to_s)
data = params[:html_file].read
else
logger.error("File does not appear to be a valid class.")
end

Related

undefined method `original_filename' for String

I'm trying to implement a csv upload functionality WITHOUT using gems such as paperclip. Here's the view:
%h1 Add Users From CSV
= form_tag(:action => "upload",:multipart => true,:method => :post) do
= file_field_tag 'csv'
= submit_tag 'Upload'
And here's the controller:
def upload
csv_io = params[:csv]
File.open(Rails.root.join('public', 'uploads', csv_io.original_filename), 'wb') do |file|
file.write(csv_io.read)
end
redirect_to root_path, :notice => "Successfully uploaded csv!"
end
But I got this error message when I'm uploading a csv called data.csv
undefined method `original_filename' for "data.csv":String
I just followed the official Rails guide, but it's still getting error. Can anyone suggest some solutions?
NOTE: I just need to read data from the csv file and it does not need to be saved persistently on server.
The way you're passing the arguments to form_tag, all your arguments are getting treated as part of the first form_tag parameter url_for_options, instead of going in part to the second parameter options (see http://api.rubyonrails.org/classes/ActionView/Helpers/FormTagHelper.html#method-i-form_tag for the method definition).
Per the guide at http://guides.rubyonrails.org/form_helpers.html#uploading-files, you can use the following syntax to achieve what you want:
form_tag({:action => "upload"},:multipart => true)
You don't need to set :method because it defaults to post.

Rails Forms for custom actions

I'm trying to link the input of a form to a specific action in my rails app.
Currently if I go to www.myapp.com/check/:idNumber, I'll be able to trigger the action just fine (which means routes is setup properly?). This action is basically a function call to a ruby/rails script with the parameter "idNumber" being passed to it. If the function is successful, it would return a newly created "Person" object and nil otherwise. This is different than the standard new operation as it determines the the attributes based on some information that it obtained from a database somewhere else.
Rake routes does give me the following:
check /check/:idNumber(.:format) person#check {:id=>/\d+/}
What I'm having trouble implementing is the form itself.
<%= form_tag("/check", :method => "get") do %>
<%= text_field_tag(:idNumber) %>
<% end %>
Controller action:
def check
regCheck = RegCheck.new
#person = regCheck.check_id(params[:idNumber])
if #person.name == nil
redirect_to root_path
end
end
submitting the form above would bring me to myapp.com/check?utf8=✓&idNumber=1234 instead. Can someone tell me what am I doing wrong?
I believe that using the check_path helper that is generated from the routes file is your best bet.
The form should look like this then.
<%= form_tag(check_path) do %>
<%= text_field_tag(:idNumber) %>
<% end %>
Rails forms can be finicky, especially when trying to build really customized forms.
This line
= form_for [#object]
Determines where the form goes, as well as the object that is being implemented. If you want to route the form to a different place, you can user the :url option. This options determines the path of the form, however you must keep in mind that the method is determined by the #object. If it is a new object, the method will be POST, an existing object will use a PUT method.
Let's suppose you want to update an existing object, but you want to send in data for a new object belonging to the existing object. That would look like
= form_for [#object], :as => #child_object, :url => my_optional_custom_path do |f|
# etc...
This generates a form sending a PUT request to the custom url (or the update path for #object if no custom url is supplied. The PUT request is sent with the parameter params[:child_object].
Hopefully this helps!
Best,
-Brian
I don't think it's possible the way you're trying.. The URL for the form is created before the user inputs any data.. So you need to remove the :idNumber from your routing..
If you do you get the following route:
check /check(.:format) person#check
Because the regex is removed now, you need to do this in you're controller:
def check
# Make sure ID is digits only
idNumber = params[:idNumber].gsub(/[^\d]/, '')
regCheck = RegCheck.new
#person = regCheck.check_id(idNumber)
if #person.name == nil
redirect_to root_path
end
end
You're form is allright, but you may want to use check_path like TheBinaryhood suggests..
If you really want it to be check/:idNumber you may also be able to submit the form to another action and redirect it to the right path from there..

How do I create a temp file and write to it then allow users to download it?

I'm working on my first application and I need some help with allowing my users to download a text file with certain variables that are being displayed on the page.
Take a shopping list for example.
Let's say you allow your users to create a shopping list of products, and then display the shopping list with the items on a shopping list page,
e.g. localhost:3000/list/my-list
Take a look at the example code below (which is probably incorrect):
File.open('shopping_list.txt', 'w') do |file|
file.puts 'Item 1: #{product_1.name}'
file.puts 'Item 2: #{product_2.name}'
file.puts 'Item 3: #{product_3.name}'
end
Which then creates a text file that has the following content:
Item 1: Eggs
Item 2: Butter
Item 3: Bread
Users should then be able to download this file (i don't want this file to be stored on the server) via a download link.
I have no idea how to achieve this, but I'm hoping you guys can guide me. :D
TL;DR
create text files populated with model data (perhaps create a method to achieve this?)
text files should not be stored on the server, but created as users click the download button (not sure if this is the rails way but perhaps someone could show me a better way)
I am assuming there is a resource for List with the attribute name as the name of the list and a list has_many Item which has an attribute description
First off, create a download path change your routes config/routes.rb
resources :lists do
member {get "download"}
end
Now if you run a rake routes in the console you should see a route like
/lists/:id/download
Whats more you should now have the helpers download_list_url & download_list_path to use in your view like
<ul>
<% #lists.each do |list| %>
<li> <%= list.name %> - <%= link_to 'Download List', download_list_path(list) %> </li>
<% end %>
</ul>
In your lists_controller add the action, and as you dont actually want to keep the file on the server disk just stream the data as a string
def download
list = List.find(params[:id])
send_data list.as_file,
:filename => "#{list.name}.txt",
:type => "text/plain"
end
Finally you see I have used a as_file method which you should add to the model (I prefer not to do this stuff in controllers, fat models, skinny controllers). So in the List model
def as_file
output = [self.name]
self.items.each {|item| output << item.description }
output.join("\n")
end
You say you don't want to store the file on the server, but "download" it on request; this sounds like you just want to generate and deliver a text document in response to the download link. There are several approaches, but you want to be sure of setting the mime-type so the browser sees it as a text file instead of an html document.
product_info = [
"Item 1: #{product_1.name}",
"Item 2: #{product_2.name}",
"Item 3: #{product_3.name}",
].join("\n")
render :text => product_info # implies :content_type => Mime::Type["text/plain"]
BTW, your example with open/puts above won't output what you think since single-quoted strings don't interpolate.
so, you wish to :
create text files populated with model data (perhaps create a method
to achieve this?)
text files should not be stored on the server, but
created as users click the download button (not sure if this is the
rails way but perhaps someone could show me a better way)
You have the right idea, here's what to do :
Create a method in your model to generate the text file contents. Let's say this method is called list_data
It seems like you have an existing controller action called my_list. Hence we can call our new method in the controller like so :
.
def my_list
# pre-existing code
respond_to do |format|
format.html # show html page as before
format.text do
send_data #list.list_data, :content_type => 'text/plain', :filename => 'my-shopping-list.txt'
end
end
end
To link to the download, just use link_to :action => my_list, :format => 'text'
See http://api.rubyonrails.org/classes/ActionController/DataStreaming.html#method-i-send_data for full docs on send_data
Caveat & explanations : Using the method above, there isn't really an explicit creation of files, Rails is streaming it for you. Hence this method is not suitable for very large files, or when the generation of the file content will take a while. Use a delayed method to generate the file and store it - the file contents somewhere if that's the case - but we can use send_data once it has been generated
You could try a combination of TempFile and send_file. In your controller action ..
file = Tempfile.new('foo')
file.write("hello world")
file.close
send_file file.path
At Rails 2.3 you can use Template Streaming. Working with Redmine I can remember something like that, you have to adapt for your case. Reference: Streaming and file downloads
require "prawn"
class ClientsController < ApplicationController
# Generate a PDF document with information on the client and return it.
# The user will get the PDF as a file download.
def download_pdf
client = Client.find(params[:id])
send_data(generate_pdf, :filename => "#{client.name}.pdf", :type => "application/pdf")
end
private
def generate_pdf(client)
Prawn::Document.new do
text client.name, :align => :center
text "Address: #{client.address}"
text "Email: #{client.email}"
end.render
end
end
Using the Thong Kuah you must just change the "content_type" param:
def my_list
# pre-existing code
respond_to do |format|
format.html # show html page as before
format.text do
send_data #list.list_data, :content_type => 'text/plain', :filename => 'my-shopping-list.txt'
end
end
end

Generate PDF file using AJAX call

I'm trying to generate a PDF file using AJAX call in Rails3. The following code generates a PDF file which I have created using PRAWN gem.
<%= link_to "Generate pdf", books_path(#book, :format => 'pdf') %>
I do not want user to view the PDF until they order it. So, the goal is to create a PDF file in the server.
Any ideas or thoughts much appreciated.
Use this, make sure your remote action does not return the PDF, but simple generates and stores it on the server.
link_to "Generate PDF", prepare_books_path(#book), :remote => true, :method => :put
This will work in Rails 3. If you're using jQuery, make sure to read this article on how to set things up correctly.
Your controller action may look like this:
def prepare
# Do your thing to generate the PDF
render :text => "PDF Generated", :status => 200
end
I used the PUT-method because you are altering the state of your data (e.g. you are generating something new, you don't want a bot or crawler to automatically call that).
Firstly, it beats me why you would do something on a request like generating a PDF, when the user is not expecting that action. Isn't better to only generate the pdf when the user requests for it?
Thanks Ariejan.
I modified your code as following and it did just what I wanted.
<%= link_to "Generate Story Book", pdfbook_stories_path(:format => 'pdf'), :remote => true %>
And for the controller,
def pdfbook
#stories = current_account.stories
respond_to do |format|
format.pdf {}
end
end

Ruby On Rails: using arrays with link_to

I was curious on how to use arrays in the link_to method in ruby on rails for example:
Controller:
def index
#test = [1,2,3]
end
View:
<%= link_to "test", {:action => 'index'}, :test => #test %>
When looking at the source then, I end up with something to the effect of:
test
My guess is that the array's to_string or something similar is getting called to set the value of test in the html.
My goal is to be able to have a form in which people can submit data on the page, and then once they've submitted the data and return to the page, if they click on the link the data will persist through clicking on the link.
*Ideally I would like to do this without having to pass the parameters in the url.
Thank you.
If you want to keep data you should probably use cookies. They are very easy to use, just assign a value with the following in the action:
cookies[:some_key] = "some value"
and retrieve it with this:
cookies[:some_key] # returns "some value"
However, just to clarify what link_to is doing in your example:
<%= link_to "test", {:action => 'index'}, :test => #test %>
When looking at the source then, I end up with something to the effect of:
test
The reason is that you are passing #test to the third argument in link_to, which is a hash of html attributes, hence why it's turned into one. To have it become an parameter on the link, you need to pass it with the second, eg, {:action => 'index', :text => #test}. As noted above, however, this is not necessarily the best way to tackle this and, in addition, it's usually best to also pass the controller name to link_to or, better yet, use a named route.
If I understand well, you want to keep the datas submitted by the user after they validate the form ?
Well Rails is able to do that without any of your code line needed.
Based on the supposition that you have a route resource "objects"
In your controller :
def edit
#object = Object.find_by_id params[:id]
end
def update
#object = Object.find_by_id params[:id]
if #object.update_attributes params[:object]
# The datas have been successfully saved. You redirect wherever you want to.
else
render :action => 'edit'
end
end
and in your view :
<% form_for #object do |f| %>
<%= text_field :name %>
<% end %>
When the form fails to validate, the "name" text field automatically gets the previous entered data.
If after that you still need to reload your datas, you don't need to add them as a parameter in a link tag.
You get the object in your controller and passes it's datas to the view where you display it.
I would just write a view helper that formats it into a string with good separators, like commas.
That isn't a good way to be passing along information though. Try session variables, cookies, or url-encoded variables instead.
The best match to what you are doing would be url-encoded variables, which will show up in a form similar to this:
test
My guess is that it is using Array#join.
You could try something like
:test => #test.join( ',' )
and then parse the string in your controller. But it is somewhat error prone if the user enters the same character you chose as delimiter.
But, assuming the linked page is also served by Rails, I think the best solution would be to use the flash area to store the results on the server
flash[ :submitted_params ] = params;
and in the controller for the linked page
old_params = flash[ :submitted_params ] || {}

Resources