#carrierwave

[tree-sitter]

Capturing Webcam Image For Uploading with Carrierwave and Cropbox

The goal of this tutorial is to capture the image from the web cam and be able to upload it as if you are uploading an image in your Ruby on Rails project.

[denguk]

Carrierwave: Update image versions

Product.all.each { |x| x.product_images.each { |img| img.picture.recreate_versions! } }

[mrblackmountain]

[coding] : passing parameters to carrierwave uploads

TL:DR; If you want to use jCrop, carrierwave and fog to crop images BEFORE they are uploaded to the cloud, do the following:

1) Add instance variables to your model:

2) permit those params in the controller (be careful about the order, the file needs to be the LAST thing you pass):

3) add the cropping method to your uploader and add process :crop to the version you want to have cropped:

That’s it :) Interested about the long story?

Here we go: This one has bothered me for two full days now and since there were no suitable stackoverflow answers and carrierwave tutorials didn’t work I thought I’ll post this here. In fact, it’s a pretty trivial problem:

We’re currently rewriting parts of Smaply. We use jCrop in frontend to give users the ability to choose which part of their uploaded picture they want to make the profile image of personas. It looks something like this:

So, when the request comes in from the frontend, it’s basically a multipart-form with four additional values (crop_x, crop_y, crop_w and crop_h) which define the area to be cropped.

Carrierwave offers the possibility to create versions and give them sizes. What it doesn’t suggest however, is a standard way to pass my request params over to those instance methods.

There are several suggested solutions:

 http://stackoverflow.com/questions/21706445/rails-jcrop-rmagick-bootstrap-no-method-error suggests that you should recreate the version after creating the image which really sounds like unnecessary overhead to me and brings problems if you use fog.

http://railscasts.com/episodes/182-cropping-images-revised same suggestion as above. I don’t want to upload, then crop and upload again.

https://medium.com/philosophie-is-thinking/carrierwave-process-parameters-or-makeit800-com-casualty-of-the-zeitgeist-12e80a80d754 suggests to use instance variables in the uploader. TL:DR; IT DOESN’T WORK since a new instance is created as soon as you pass the file param.

http://stackoverflow.com/questions/10631060/how-can-i-pass-arbitrary-user-defined-values-into-a-carrierwave-image-processor works perfectly fine in Rails 3.

At first I fell in love with suggestion number 3, since it felt better to have those instance variables in the Uploader class itself. However this approach just doesn’t work because Carrierwave triggers the upload as soon as you pass in the file param and therefore creates a new instance of the Uploader class where those attributes are obviously nil.

I converted to “puts driven development” mode and added some statements and a byebug marker in order to be able to debug a little better and understand the Carrierwave internals. Here is the log output that catched my attention (I pixelated some swearing of mine):

So, option Number 3 didn’t work either. Back to the basic idea of just adding those instance variables to the model itself as suggested in the last Stackoverflow Link above. But somehow it still didn’t work until I scrolled further down:

So, several hours of bug hunt just to crop an image according to user input. That’s it. Since I didn’t find anything comprehensive on the topic, I decided to wrap up my findings and hope that it turns out to be useful for some of you out there.

[banker-hacker]

use #carrierwave to save image from #mechanize in #ruby

you need to convert the object from a Mechanize::Image object to one that carrierwave will recognize. You do this by using the old Ruby standby class File.

In development, you can just save the image anywhere in your file directory. But in production, specifically in @heroku, you are limited from saving any files to the directory. So using our existing carrierwave route that we created for allowing users to upload avatars, I did the following:

captcha_el = login.search('img')[4].attributes['src'] captcha_file = @agent.get(captcha_el).save! 'captcha1.jpg' file = File.open(captcha_file) user = User.find(1) user.update! avatar: file

in this example, to make it easy on myself, I’m saving the image as the avatar for a dummy user account created specifically for that purpose. That way, I don’t need to create any new models. This is a hack but it saves time. the User model looks like the following:

class User < ActiveRecord::Base mount_uploader :avatar, AvatarUploader end

and the AvatarUploader model:

class AvatarUploader < CarrierWave::Uploader::Base include CarrierWave::MiniMagick end

[ayn]

[edwardkenfox]

Rails: 添付ファイルの表示

CarrierWaveなどでアップロードしたファイルをviewで表示する、というのがよくある。けどデフォルトの挙動では、添付ファイルのリンクを踏む（ファイルのパスにアクセスする）と、ファイルのダウンロードが始まってしまい、ちょっと憂鬱だった。

PDFや画像などはそのままブラウザで見たいので、ダウンロードではなくブラウザ内で表示させるには、

send_file(filepath, filename: @memo_file[:filename], disposition: 'inline')

とすればOK。別タブで表示するには、別途viewファイルでのリンクにtarget: '_blank'の指定もすればよろし。

[rubydailystream]

After being asked about how to use Fixtures in CarrierWave (the Rails file uploader), I wrote an article about how I’d do it.

[imshanks-blog]

carrierwave rename exist filename

二种方法，推荐第二种

https://gist.github.com/38232602c169c508c2c4.githttps://gist.github.com/38232602c169c508c2c4.git