0.0
No release in over a year
Fake CCD server for employment tribunals
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Development

~> 2.0
~> 13.0
~> 3.0

Runtime

~> 0.7
~> 3.21
~> 6.2
~> 1.0
~> 2.0, >= 2.0.9
 Project Readme

EtFakeCcd

This gem can either be used as part of a test suite, providing a rack endpoint which can be loaded using webmock.

OR

It can be used as a standalone server

Installation

Add this line to your application's Gemfile:

gem 'et_fake_ccd'

And then execute:

$ bundle

Or install it yourself as:

$ gem install et_fake_ccd

Usage

To run the server, use the command

et_fake_ccd start

or if you want to specify which json schema file you want to use for validation (no default is provided as it changes in CCD regularly using the config file), you can either specify the otion 'create_case_schema' as shown below OR you can specify the environment variable ET_FAKE_CCD_CREATE_CASE_SCHEMA instead.

The 'master' definition of this file is here https://raw.githubusercontent.com/hmcts/et-ccd-export/develop/spec/json_schemas/case_create.json so use wget or curl etc.. or just a browser to download it if you want real life validation.

et_fake_ccd start --create_case_schema=<path_to_file>

Testing Using This Fake Server

In general, this server will try to act like the normal CCD server but only in terms of accepting and storing data. So, it stores the data that the application gives it and it allows that data to be read back so that a test suite or a manual process can validate that data.

It also has some basic validation so that it behaves in a similar way to CCD, but due to the complexity of CCD itself, this app can never reflect exactly how it validates.

So, for this reason, deliberate errors are added to this server.

Testing Using Deliberate Errors

When you think about it, if there was a way in the application to force an error, the developer would fix it ! So, we have to rely on having some deliberate errors where the data will pass through all the validation within the app, but this server then knows to respond as if it was an error.

This will allow a tester or a test suite to provide the magic values that will trigger certain error scenarios and prove that the app behaves as expected under these error conditions.

This allows for basic testing and is not intended to be real life - only real life will give you real life errors.

For example, lets say we chose the name "Fred Bloggs" (sorry, the real fred bloggs - you must get fed up with people using your name) to always respond with a 403 error saying "Forbidden".

Obviously, you could not force the app to get a real forbidden error, so as a tester or a developer writing test code, you would know from this readme that Fred Bloggs always causes a 403 error, therefore you would expect the exported case in the admin to be marked with an error as an example.

So, here is a list of deliberate errors that Im sure will increase in size :-

Force a 403 error every time

  • Use the lead claimant name "Force ErrorForbidden-" (replace with a number unique to your test)

Force a 403 error for just the first time

  • Use the lead claimant name "Force ErrorForbidden-" (replace with a number unique to your test)

Force a 504 error

  • Use the lead claimant name "Force ErrorGatewayTimeout-" (replace with a number unique to your test)
  • Use the lead claimant title of "Mr" to only force the error on the first occurence
  • Use the lead claimant title of "Mrs" to force the error on every occurence

Force a 502 error

  • Use the lead claimant name "Force ErrorBadGateway-" (replace with a number unique to your test)
  • Use the lead claimant title of "Mr" to only force the error on the first occurence
  • Use the lead claimant title of "Mrs" to force the error on every occurence

Force a 422 error

  • Use the lead claimant name "Force ErrorUnprocessableEntity-" (replace with a number unique to your test)
  • Use the lead claimant title of "Mr" to only force the error on the first occurence
  • Use the lead claimant title of "Mrs" to force the error on every occurence

Improved Deliberate Error Control

The above deliberate error mechanism worked, it does not give quite enough control.

A new system has therefore been developed that you can use instead of it.

This uses special configuration settings in the 'External System' in the admin.

There are 2 key entries

  1. 'extra_headers'
  2. 'send_request_id'

extra_headers

'extra_headers' should contain a JSON encoded value of a hash. Each entry in the hash is a header to add to every request to this fake ccd server.

force_failures header

The force_failures header should contain a hash which looks like this

    {
      "idam_stage": { ..spec.. },
      "token_stage": { ..spec.. },
      "data_stage": { ..spec.. }
    } 

The 4 different stages give control of when the error will happen

The 'idam_stage' is the stage of the transaction when an IDAM token is requested. However, this does not happen all of the time because IDAM tokens are cached, so you will not necessarily see one request per transaction.

The 'token_stage' is used in most transactions such as case creation where a 'token' is the starting stage - which then allows the case to be created against this token.

The 'documents' stage is used in transactions that require documents uploading before the case is created.

The 'data_stage' is used in most transactions and means the actual data creation.

The '..spec..' is the same irrespective of which stage and is described below:

    [a, b, c]

a, b, c (you can specify as many as you want here, not just 3) are http response codes for the 'nth' request. i.e. the first argument is for the first request, the 2nd for the 2nd etc.. The special value of 0 means allow the normal response and do not force an error - allowing for patterns such as error on the 1st, 2nd and 5th.

Any non zero value is the http status code to respond with.

send_request_id

'send_request_id' should be set to 'true' to enable a request identifier to be sent with every request to this fake ccd server. This identifier is unique to a particular export from the main system, so it will persist even across retries of the same export.

This is then used to assist in special rules where errors are forced on the 'nth' request for example.

Development

After checking out the repo, run bin/setup to install dependencies. Then, run rake spec to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and tags, and push the .gem file to rubygems.org.

Contributing

Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/et_fake_ccd.

License

The gem is available as open source under the terms of the MIT License.