Faraday: Feature request: support `Faraday::FlatParamsEncoder` in `Faraday::Adapter::Test`

Created on 19 Aug 2021  路  4Comments  路  Source: lostisland/faraday

1291 introduced strict_mode in Faraday::Adapter::Test::Stubs, and I want more features for the mode: supporting Faraday::FlatParamsEncoder. For example, if I want to call an HTTP request with the same query parameter a multiple times, I expect the test stub to check whether the same parameters with different values are all correctly passed.

class Client
  def initialize(conn)
    @conn = conn
  end

  def sushi(jname, params: {})
    res = @conn.get("/#{jname}", params)
    data = JSON.parse(res.body)
    data['name']
  end
end

RSpec.describe Client do
  let(:stubs) { Faraday::Adapter::Test::Stubs.new(strict_mode: true) }
  let(:conn)   { Faraday.new(request: { params_encoder: Faraday::FlatParamsEncoder }) { |b| b.adapter(:test, stubs) } }
  let(:client) { Client.new(conn) }

  it 'handles the same multiple URL parameters' do
    # The same parameter `a` is expected to appear multiple times with different values.
    stubs.get('/ebi?a=x&a=y&a=z') { [200, { 'Content-Type' => 'application/json' }, '{"name": "shrimp"}'] }

    # Test should pass
    expect(client.sushi('ebi', params: { a: %w[x y z] })).to eq('shrimp')

    # Test should not pass due to the lack of "x" and "y" for `a`
    # expect(client.sushi('ebi', params: { a: %w[z] })).to eq('shrimp')
    stubs.verify_stubbed_calls
  end
end

What do you think? I want to open a pull request if it makes sense.

feature help wanted

Most helpful comment

Also, remember that main branch is now pointing at the upcoming v2.0 release.
If you'd like this in a 1.x release as well, you'll need to branch off the 1.x branch 馃憤

All 4 comments

Hi @yykamei, the above makes sense to me, I'd welcome a PR.
Out of curiosity, what happens currently with the tests above? Is the last one passing?

Also, remember that main branch is now pointing at the upcoming v2.0 release.
If you'd like this in a 1.x release as well, you'll need to branch off the 1.x branch 馃憤

Hi @iMacTia, thanks for your reply.

Out of curiosity, what happens currently with the tests above? Is the last one passing?

Oh, it seems not to pass in the latest 1.x branch. I think this is because Stub#params is constructed with Faraday::Utils.parse_nested_query; it looks like overriding previous query parameters.

Randomized with seed 45661

Client
  handles the same multiple URL parameters (FAILED - 1)

Failures:

  1) Client handles the same multiple URL parameters
     Failure/Error:
       raise Stubs::NotFound, "no stubbed request for #{env[:method]} "\
                              "#{normalized_path} #{env[:body]}"

     Faraday::Adapter::Test::Stubs::NotFound:
       no stubbed request for get /ebi?a=x
     # ./lib/faraday/adapter/test.rb:253:in `call'
     # ./lib/faraday/rack_builder.rb:154:in `build_response'
     # ./lib/faraday/connection.rb:511:in `run_request'
     # ./lib/faraday/connection.rb:200:in `get'
     # ./ok.rb:11:in `sushi'
     # ./ok.rb:30:in `block (2 levels) in <top (required)>'

Finished in 0.00391 seconds (files took 0.78136 seconds to load)
1 example, 1 failure

Failed examples:

But, if I change the code like this, the test will pass. I rewrote the issue description.

diff --git a/a.rb b/a.rb
index c4b7084..91d795d 100644
--- a/a.rb
+++ b/a.rb
@@ -27,7 +27,7 @@ RSpec.describe Client do
     expect(client.sushi('ebi', params: { a: %w[x y z] })).to eq('shrimp')

     # Test should not pass due to the lack of "y" and "z" for `a`
-    expect(client.sushi('ebi', params: { a: %w[x] })).to eq('shrimp')
+    expect(client.sushi('ebi', params: { a: %w[z] })).to eq('shrimp')
     stubs.verify_stubbed_calls
   end
 end

Thanks anyway. I'm going to make some changes 馃挭

This issue was resolved with #1316.

Thank you!

Was this page helpful?
0 / 5 - 0 ratings

Related issues

Lewiscowles1986 picture Lewiscowles1986  路  4Comments

iMacTia picture iMacTia  路  3Comments

JasonBarnabe picture JasonBarnabe  路  4Comments

subvertallchris picture subvertallchris  路  5Comments

aleksb86 picture aleksb86  路  3Comments