Broken urls - Unsafe characters in public_id are not properly escaped by Cloudinary::Utils.cloudinary_url

[Qapybara28]

Bug report for Cloudinary Ruby SDK

Before proceeding, please update to latest version and test if the issue persists

Describe the bug in a sentence or two.

Unsafe characters in public_id are not double-escaped by Cloudinary::Utils.cloudinary_url so it generates invalid urls.

The problem was discovered after using the public_id returned in the upload JSON.

Issue Type (Can be multiple)

• [ ] Build - Cannot install or import the SDK
• [ ] Performance - Performance issues
• [X] Behaviour - Functions are not working as expected (such as generate URL)
• [ ] Documentation - Inconsistency between the docs and behaviour
• [ ] Other (Specify)

Steps to reproduce

Current result (not working):

assert_match  /%C3%A4%28/, Cloudinary::Utils.cloudinary_url('ä(')

Correct result:

assert_match /%25C3%25A4%2528/, Cloudinary::Utils.cloudinary_url('ä(')

Dirty workaround

module FixCloudinaryUtilsEscaping
  def smart_escape(string, unsafe = %r{([^a-zA-Z\d_.\-/:! '*~]+)})
    super(string, unsafe)
  end

  def finalize_source(source, format, url_suffix)
    source, source_to_sign = super
    [smart_escape(source).gsub(' ', '%20'), source_to_sign]
  end
end

Edit: Updated workaround with escaping edge-cases

[dannyv-cloudinary]

Hi Alexandre.

Thanks for raising this with us. I have created an internal ticket for the development team to address this. We'll update this thread with progress once we have some updates for you.

Thanks, -Danny

[dannyv-cloudinary]

Hi Alexandre.

I've just heard back from the dev team and after some thorough testing, it seems there is no need to double escape as cloudinary_url generates the same url as we get from the upload API response.

If you're still experiencing issues, would you mind providing a little more context and some examples?

Many thanks, -Danny