Commit 61318cf1 authored by Stan Hu's avatar Stan Hu

Merge branch '299558-integrate-with-anti-spam-service' into 'master'

Integrate Spamcheck gRPC gem [RUN ALL RSPEC][RUN AS-IF-FOSS]

See merge request gitlab-org/gitlab!52385
parents 5d6d4d4c d063fc83
......@@ -477,6 +477,9 @@ group :ed25519 do
gem 'bcrypt_pbkdf', '~> 1.0'
end
# Spamcheck GRPC protocol definitions
gem 'spamcheck', '~> 0.0.5'
# Gitaly GRPC protocol definitions
gem 'gitaly', '~> 13.11.0.pre.rc1'
......
......@@ -1211,6 +1211,8 @@ GEM
thor (~> 1.0)
tilt (~> 2.0)
yard (~> 0.9, >= 0.9.24)
spamcheck (0.0.5)
grpc (~> 1.0)
spring (2.1.1)
spring-commands-rspec (1.0.4)
spring (>= 0.9.1)
......@@ -1606,6 +1608,7 @@ DEPENDENCIES
slack-messenger (~> 2.3.4)
snowplow-tracker (~> 0.6.1)
solargraph (~> 0.40.4)
spamcheck (~> 0.0.5)
spring (~> 2.1.0)
spring-commands-rspec (~> 1.0.4)
sprockets (~> 3.7.0)
......
......@@ -375,7 +375,7 @@ class ApplicationSetting < ApplicationRecord
if: :external_authorization_service_enabled
validates :spam_check_endpoint_url,
addressable_url: true, allow_blank: true
addressable_url: { schemes: %w(grpc) }, allow_blank: true
validates :spam_check_endpoint_url,
presence: true,
......
......@@ -20,7 +20,7 @@ module Spam
created_at: DateTime.current,
author: owner_name,
author_email: owner_email,
referrer: options[:referrer]
referer: options[:referer]
}
begin
......
......@@ -53,7 +53,7 @@ module Spam
if request
options[:ip_address] = request.env['action_dispatch.remote_ip'].to_s
options[:user_agent] = request.env['HTTP_USER_AGENT']
options[:referrer] = request.env['HTTP_REFERRER']
options[:referer] = request.env['HTTP_REFERER']
else
# TODO: This code is never used, because we do not perform a verification if there is not a
# request. Why? Should it be deleted? Or should we check even if there is no request?
......@@ -123,6 +123,11 @@ module Spam
# https://gitlab.com/gitlab-org/gitlab/-/issues/214739
target.spam! unless target.allow_possible_spam?
create_spam_log(api)
when BLOCK_USER
# TODO: improve BLOCK_USER handling, non-existent until now
# https://gitlab.com/gitlab-org/gitlab/-/issues/329666
target.spam! unless target.allow_possible_spam?
create_spam_log(api)
when ALLOW
target.clear_spam_flags!
end
......
......@@ -10,25 +10,44 @@ module Spam
@request = request
@user = user
@options = options
@verdict_params = assemble_verdict_params(context)
@context = context
end
def execute
external_spam_check_result = external_verdict
spamcheck_result = nil
external_spam_check_round_trip_time = Benchmark.realtime do
spamcheck_result = spamcheck_verdict
end
akismet_result = akismet_verdict
# filter out anything we don't recognise, including nils.
valid_results = [external_spam_check_result, akismet_result].compact.select { |r| SUPPORTED_VERDICTS.key?(r) }
valid_results = [spamcheck_result, akismet_result].compact.select { |r| SUPPORTED_VERDICTS.key?(r) }
# Treat nils - such as service unavailable - as ALLOW
return ALLOW unless valid_results.any?
# Favour the most restrictive result.
valid_results.min_by { |v| SUPPORTED_VERDICTS[v][:priority] }
final_verdict = valid_results.min_by { |v| SUPPORTED_VERDICTS[v][:priority] }
logger.info(class: self.class.name,
akismet_verdict: akismet_verdict,
spam_check_verdict: spamcheck_result,
spam_check_rtt: external_spam_check_round_trip_time.real,
final_verdict: final_verdict,
username: user.username,
user_id: user.id,
target_type: target.class.to_s,
project_id: target.project_id
)
final_verdict
end
private
attr_reader :user, :target, :request, :options, :verdict_params
attr_reader :user, :target, :request, :options, :context
def akismet_verdict
if akismet.spam?
......@@ -38,54 +57,34 @@ module Spam
end
end
def external_verdict
def spamcheck_verdict
return unless Gitlab::CurrentSettings.spam_check_endpoint_enabled
return if endpoint_url.blank?
begin
result = Gitlab::HTTP.post(endpoint_url, body: verdict_params.to_json, headers: { 'Content-Type' => 'application/json' })
result, _error = spamcheck_client.issue_spam?(spam_issue: target, user: user, context: context)
return unless result
json_result = Gitlab::Json.parse(result).with_indifferent_access
# @TODO metrics/logging
# Expecting:
# error: (string or nil)
# verdict: (string or nil)
# @TODO log if json_result[:error]
# @TODO log if error is not nil https://gitlab.com/gitlab-org/gitlab/-/issues/329545
json_result[:verdict]
rescue *Gitlab::HTTP::HTTP_ERRORS => e
# @TODO: log error via try_post https://gitlab.com/gitlab-org/gitlab/-/issues/219223
# Duplicate logic with Akismet logic in #akismet_verdict
if Gitlab::Recaptcha.enabled? && result != ALLOW
CONDITIONAL_ALLOW
else
result
end
rescue StandardError => e
Gitlab::ErrorTracking.log_exception(e)
nil
rescue StandardError
# @TODO log
# Default to ALLOW if any errors occur
ALLOW
end
end
def assemble_verdict_params(context)
return {} unless endpoint_url.present?
project = target.try(:project)
context.merge({
target: {
title: target.spam_title,
description: target.spam_description,
type: target.class.to_s
},
user: {
created_at: user.created_at,
email: user.email,
username: user.username
},
user_in_project: user.authorized_project?(project)
})
def spamcheck_client
@spamcheck_client ||= Gitlab::Spamcheck::Client.new
end
def endpoint_url
@endpoint_url ||= Gitlab::CurrentSettings.current_application_settings.spam_check_endpoint_url
def logger
@logger ||= Gitlab::AppJsonLogger.build
end
end
end
---
title: Integrate with the Spamcheck anti-spam engine
merge_request: 52385
author:
type: added
# frozen_string_literal: true
require 'spamcheck'
module Gitlab
module Spamcheck
class Client
include ::Spam::SpamConstants
DEFAULT_TIMEOUT_SECS = 2
VERDICT_MAPPING = {
::Spamcheck::SpamVerdict::Verdict::ALLOW => ALLOW,
::Spamcheck::SpamVerdict::Verdict::CONDITIONAL_ALLOW => CONDITIONAL_ALLOW,
::Spamcheck::SpamVerdict::Verdict::DISALLOW => DISALLOW,
::Spamcheck::SpamVerdict::Verdict::BLOCK => BLOCK_USER
}.freeze
ACTION_MAPPING = {
create: ::Spamcheck::Action::CREATE,
update: ::Spamcheck::Action::UPDATE
}.freeze
def initialize
@endpoint_url = Gitlab::CurrentSettings.current_application_settings.spam_check_endpoint_url
# remove the `grpc://` as it's only useful to ensure we're expecting to
# connect with Spamcheck
@endpoint_url = @endpoint_url.gsub(%r(^grpc:\/\/), '')
creds =
if Rails.env.development? || Rails.env.test?
:this_channel_is_insecure
else
GRPC::Core::ChannelCredentials.new
end
@stub = ::Spamcheck::SpamcheckService::Stub.new(@endpoint_url, creds,
timeout: DEFAULT_TIMEOUT_SECS)
end
def issue_spam?(spam_issue:, user:, context: {})
issue = build_issue_protobuf(issue: spam_issue, user: user, context: context)
response = @stub.check_for_spam_issue(issue,
metadata: { 'authorization' =>
Gitlab::CurrentSettings.spam_check_api_key })
verdict = convert_verdict_to_gitlab_constant(response.verdict)
[verdict, response.error]
end
private
def convert_verdict_to_gitlab_constant(verdict)
VERDICT_MAPPING.fetch(::Spamcheck::SpamVerdict::Verdict.resolve(verdict), verdict)
end
def build_issue_protobuf(issue:, user:, context:)
issue_pb = ::Spamcheck::Issue.new
issue_pb.title = issue.spam_title || ''
issue_pb.description = issue.spam_description || ''
issue_pb.created_at = convert_to_pb_timestamp(issue.created_at) if issue.created_at
issue_pb.updated_at = convert_to_pb_timestamp(issue.updated_at) if issue.updated_at
issue_pb.user_in_project = user.authorized_project?(issue.project)
issue_pb.action = ACTION_MAPPING.fetch(context.fetch(:action)) if context.has_key?(:action)
issue_pb.user = build_user_protobuf(user)
issue_pb
end
def build_user_protobuf(user)
user_pb = ::Spamcheck::User.new
user_pb.username = user.username
user_pb.org = user.organization || ''
user_pb.created_at = convert_to_pb_timestamp(user.created_at)
user_pb.emails << build_email(user.email, user.confirmed?)
user.emails.each do |email|
user_pb.emails << build_email(email.email, email.confirmed?)
end
user_pb
end
def build_email(email, verified)
email_pb = ::Spamcheck::User::Email.new
email_pb.email = email
email_pb.verified = verified
email_pb
end
def convert_to_pb_timestamp(ar_timestamp)
Google::Protobuf::Timestamp.new(seconds: ar_timestamp.to_time.to_i,
nanos: ar_timestamp.to_time.nsec)
end
end
end
end
......@@ -425,7 +425,7 @@ RSpec.describe 'Admin updates settings' do
check 'Enable reCAPTCHA for login'
fill_in 'IPs per user', with: 15
check 'Enable Spam Check via external API endpoint'
fill_in 'URL of the external Spam Check endpoint', with: 'https://www.example.com/spamcheck'
fill_in 'URL of the external Spam Check endpoint', with: 'grpc://www.example.com/spamcheck'
fill_in 'Spam Check API Key', with: 'SPAM_CHECK_API_KEY'
click_button 'Save changes'
end
......@@ -435,7 +435,7 @@ RSpec.describe 'Admin updates settings' do
expect(current_settings.login_recaptcha_protection_enabled).to be true
expect(current_settings.unique_ips_limit_per_user).to eq(15)
expect(current_settings.spam_check_endpoint_enabled).to be true
expect(current_settings.spam_check_endpoint_url).to eq 'https://www.example.com/spamcheck'
expect(current_settings.spam_check_endpoint_url).to eq 'grpc://www.example.com/spamcheck'
end
end
......
# frozen_string_literal: true
require 'spec_helper'
RSpec.describe Gitlab::Spamcheck::Client do
include_context 'includes Spam constants'
let(:endpoint) { 'grpc://grpc.test.url' }
let_it_be(:user) { create(:user, organization: 'GitLab') }
let(:verdict_value) { nil }
let(:error_value) { "" }
let_it_be(:issue) { create(:issue, description: 'Test issue description') }
let(:response) do
verdict = ::Spamcheck::SpamVerdict.new
verdict.verdict = verdict_value
verdict.error = error_value
verdict
end
subject { described_class.new.issue_spam?(spam_issue: issue, user: user) }
before do
stub_application_setting(spam_check_endpoint_url: endpoint)
end
describe '#issue_spam?' do
before do
allow_next_instance_of(::Spamcheck::SpamcheckService::Stub) do |instance|
allow(instance).to receive(:check_for_spam_issue).and_return(response)
end
end
using RSpec::Parameterized::TableSyntax
where(:verdict, :expected) do
::Spamcheck::SpamVerdict::Verdict::ALLOW | Spam::SpamConstants::ALLOW
::Spamcheck::SpamVerdict::Verdict::CONDITIONAL_ALLOW | Spam::SpamConstants::CONDITIONAL_ALLOW
::Spamcheck::SpamVerdict::Verdict::DISALLOW | Spam::SpamConstants::DISALLOW
::Spamcheck::SpamVerdict::Verdict::BLOCK | Spam::SpamConstants::BLOCK_USER
end
with_them do
let(:verdict_value) { verdict }
it "returns expected spam constant" do
expect(subject).to eq([expected, ""])
end
end
end
describe "#build_issue_protobuf", :aggregate_failures do
it 'builds the expected protobuf object' do
cxt = { action: :create }
issue_pb = described_class.new.send(:build_issue_protobuf,
issue: issue, user: user,
context: cxt)
expect(issue_pb.title).to eq issue.title
expect(issue_pb.description).to eq issue.description
expect(issue_pb.user_in_project). to be false
expect(issue_pb.created_at).to eq timestamp_to_protobuf_timestamp(issue.created_at)
expect(issue_pb.updated_at).to eq timestamp_to_protobuf_timestamp(issue.updated_at)
expect(issue_pb.action).to be ::Spamcheck::Action.lookup(::Spamcheck::Action::CREATE)
expect(issue_pb.user.username).to eq user.username
end
end
describe '#build_user_proto_buf', :aggregate_failures do
it 'builds the expected protobuf object' do
user_pb = described_class.new.send(:build_user_protobuf, user)
expect(user_pb.username).to eq user.username
expect(user_pb.org).to eq user.organization
expect(user_pb.created_at).to eq timestamp_to_protobuf_timestamp(user.created_at)
expect(user_pb.emails.count).to be 1
expect(user_pb.emails.first.email).to eq user.email
expect(user_pb.emails.first.verified).to eq user.confirmed?
end
context 'when user has multiple email addresses' do
let(:secondary_email) {create(:email, :confirmed, user: user)}
before do
user.emails << secondary_email
end
it 'adds emsils to the user pb object' do
user_pb = described_class.new.send(:build_user_protobuf, user)
expect(user_pb.emails.count).to eq 2
expect(user_pb.emails.first.email).to eq user.email
expect(user_pb.emails.first.verified).to eq user.confirmed?
expect(user_pb.emails.last.email).to eq secondary_email.email
expect(user_pb.emails.last.verified).to eq secondary_email.confirmed?
end
end
end
private
def timestamp_to_protobuf_timestamp(timestamp)
Google::Protobuf::Timestamp.new(seconds: timestamp.to_time.to_i,
nanos: timestamp.to_time.nsec)
end
end
......@@ -216,7 +216,8 @@ RSpec.describe ApplicationSetting do
setting.spam_check_endpoint_enabled = true
end
it { is_expected.to allow_value('https://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.to allow_value('grpc://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value('https://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value('nonsense').for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value(nil).for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value('').for(:spam_check_endpoint_url) }
......@@ -227,7 +228,8 @@ RSpec.describe ApplicationSetting do
setting.spam_check_endpoint_enabled = false
end
it { is_expected.to allow_value('https://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.to allow_value('grpc://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value('https://example.org/spam_check').for(:spam_check_endpoint_url) }
it { is_expected.not_to allow_value('nonsense').for(:spam_check_endpoint_url) }
it { is_expected.to allow_value(nil).for(:spam_check_endpoint_url) }
it { is_expected.to allow_value('').for(:spam_check_endpoint_url) }
......
......@@ -123,7 +123,7 @@ RSpec.describe API::Settings, 'Settings', :do_not_mock_admin_mode_setting do
issues_create_limit: 300,
raw_blob_request_limit: 300,
spam_check_endpoint_enabled: true,
spam_check_endpoint_url: 'https://example.com/spam_check',
spam_check_endpoint_url: 'grpc://example.com/spam_check',
spam_check_api_key: 'SPAM_CHECK_API_KEY',
disabled_oauth_sign_in_sources: 'unknown',
import_sources: 'github,bitbucket',
......@@ -169,7 +169,7 @@ RSpec.describe API::Settings, 'Settings', :do_not_mock_admin_mode_setting do
expect(json_response['issues_create_limit']).to eq(300)
expect(json_response['raw_blob_request_limit']).to eq(300)
expect(json_response['spam_check_endpoint_enabled']).to be_truthy
expect(json_response['spam_check_endpoint_url']).to eq('https://example.com/spam_check')
expect(json_response['spam_check_endpoint_url']).to eq('grpc://example.com/spam_check')
expect(json_response['spam_check_api_key']).to eq('SPAM_CHECK_API_KEY')
expect(json_response['disabled_oauth_sign_in_sources']).to eq([])
expect(json_response['import_sources']).to match_array(%w(github bitbucket))
......
......@@ -9,11 +9,11 @@ RSpec.describe Spam::SpamActionService do
let(:issue) { create(:issue, project: project, author: user) }
let(:fake_ip) { '1.2.3.4' }
let(:fake_user_agent) { 'fake-user-agent' }
let(:fake_referrer) { 'fake-http-referrer' }
let(:fake_referer) { 'fake-http-referer' }
let(:env) do
{ 'action_dispatch.remote_ip' => fake_ip,
'HTTP_USER_AGENT' => fake_user_agent,
'HTTP_REFERRER' => fake_referrer }
'HTTP_REFERER' => fake_referer }
end
let_it_be(:project) { create(:project, :public) }
......@@ -80,7 +80,7 @@ RSpec.describe Spam::SpamActionService do
{
ip_address: fake_ip,
user_agent: fake_user_agent,
referrer: fake_referrer
referer: fake_referer
}
end
......@@ -222,6 +222,38 @@ RSpec.describe Spam::SpamActionService do
end
end
context 'spam verdict service advises to block the user' do
before do
allow(fake_verdict_service).to receive(:execute).and_return(BLOCK_USER)
end
context 'when allow_possible_spam feature flag is false' do
before do
stub_feature_flags(allow_possible_spam: false)
end
it_behaves_like 'only checks for spam if a request is provided'
it 'marks as spam' do
response = subject
expect(response.message).to match(expected_service_check_response_message)
expect(issue).to be_spam
end
end
context 'when allow_possible_spam feature flag is true' do
it_behaves_like 'only checks for spam if a request is provided'
it 'does not mark as spam' do
response = subject
expect(response.message).to match(expected_service_check_response_message)
expect(issue).not_to be_spam
end
end
end
context 'when spam verdict service conditionally allows' do
before do
allow(fake_verdict_service).to receive(:execute).and_return(CONDITIONAL_ALLOW)
......
......@@ -7,18 +7,18 @@ RSpec.describe Spam::SpamVerdictService do
let(:fake_ip) { '1.2.3.4' }
let(:fake_user_agent) { 'fake-user-agent' }
let(:fake_referrer) { 'fake-http-referrer' }
let(:fake_referer) { 'fake-http-referer' }
let(:env) do
{ 'action_dispatch.remote_ip' => fake_ip,
'HTTP_USER_AGENT' => fake_user_agent,
'HTTP_REFERRER' => fake_referrer }
'HTTP_REFERER' => fake_referer }
end
let(:request) { double(:request, env: env) }
let(:check_for_spam) { true }
let_it_be(:user) { create(:user) }
let(:issue) { build(:issue, author: user) }
let_it_be(:issue) { create(:issue, author: user) }
let(:service) do
described_class.new(user: user, target: issue, request: request, options: {})
end
......@@ -28,7 +28,7 @@ RSpec.describe Spam::SpamVerdictService do
before do
allow(service).to receive(:akismet_verdict).and_return(nil)
allow(service).to receive(:external_verdict).and_return(nil)
allow(service).to receive(:spamcheck_verdict).and_return(nil)
end
context 'if all services return nil' do
......@@ -63,7 +63,7 @@ RSpec.describe Spam::SpamVerdictService do
context 'and they are supported' do
before do
allow(service).to receive(:akismet_verdict).and_return(DISALLOW)
allow(service).to receive(:external_verdict).and_return(BLOCK_USER)
allow(service).to receive(:spamcheck_verdict).and_return(BLOCK_USER)
end
it 'renders the more restrictive verdict' do
......@@ -74,7 +74,7 @@ RSpec.describe Spam::SpamVerdictService do
context 'and one is supported' do
before do
allow(service).to receive(:akismet_verdict).and_return('nonsense')
allow(service).to receive(:external_verdict).and_return(BLOCK_USER)
allow(service).to receive(:spamcheck_verdict).and_return(BLOCK_USER)
end
it 'renders the more restrictive verdict' do
......@@ -85,7 +85,7 @@ RSpec.describe Spam::SpamVerdictService do
context 'and none are supported' do
before do
allow(service).to receive(:akismet_verdict).and_return('nonsense')
allow(service).to receive(:external_verdict).and_return('rubbish')
allow(service).to receive(:spamcheck_verdict).and_return('rubbish')
end
it 'renders the more restrictive verdict' do
......@@ -150,48 +150,87 @@ RSpec.describe Spam::SpamVerdictService do
end
end
describe '#external_verdict' do
subject { service.send(:external_verdict) }
describe '#spamcheck_verdict' do
subject { service.send(:spamcheck_verdict) }
context 'if a Spam Check endpoint enabled and set to a URL' do
let(:spam_check_body) { {} }
let(:spam_check_http_status) { nil }
let(:endpoint_url) { "grpc://www.spamcheckurl.com/spam_check" }
let(:spam_client) do
Gitlab::Spamcheck::Client.new
end
before do
stub_application_setting(spam_check_endpoint_enabled: true)
stub_application_setting(spam_check_endpoint_url: "http://www.spamcheckurl.com/spam_check")
stub_request(:post, /.*spamcheckurl.com.*/).to_return( body: spam_check_body.to_json, status: spam_check_http_status )
stub_application_setting(spam_check_endpoint_url: endpoint_url)
end
context 'if the endpoint is accessible' do
let(:spam_check_http_status) { 200 }
let(:error) { nil }
let(:error) { '' }
let(:verdict) { nil }
let(:spam_check_body) do
{ verdict: verdict, error: error }
before do
allow(service).to receive(:spamcheck_client).and_return(spam_client)
allow(spam_client).to receive(:issue_spam?).and_return([verdict, error])
end
context 'the result is a valid verdict' do
let(:verdict) { 'allow' }
let(:verdict) { ALLOW }
it 'returns the verdict' do
expect(subject).to eq ALLOW
end
end
context 'the verdict is an unexpected string' do
let(:verdict) { 'this is fine' }
context 'when recaptcha is enabled' do
before do
allow(Gitlab::Recaptcha).to receive(:enabled?).and_return(true)
end
it 'returns the string' do
expect(subject).to eq verdict
using RSpec::Parameterized::TableSyntax
# rubocop: disable Lint/BinaryOperatorWithIdenticalOperands
where(:verdict_value, :expected) do
::Spam::SpamConstants::ALLOW | ::Spam::SpamConstants::ALLOW
::Spam::SpamConstants::CONDITIONAL_ALLOW | ::Spam::SpamConstants::CONDITIONAL_ALLOW
::Spam::SpamConstants::DISALLOW | ::Spam::SpamConstants::CONDITIONAL_ALLOW
::Spam::SpamConstants::BLOCK_USER | ::Spam::SpamConstants::CONDITIONAL_ALLOW
end
# rubocop: enable Lint/BinaryOperatorWithIdenticalOperands
with_them do
let(:verdict) { verdict_value }
it "returns expected spam constant" do
expect(subject).to eq(expected)
end
end
end
context 'the JSON is malformed' do
let(:spam_check_body) { 'this is fine' }
context 'when recaptcha is disabled' do
before do
allow(Gitlab::Recaptcha).to receive(:enabled?).and_return(false)
end
it 'returns allow' do
expect(subject).to eq ALLOW
[::Spam::SpamConstants::ALLOW,
::Spam::SpamConstants::CONDITIONAL_ALLOW,
::Spam::SpamConstants::DISALLOW,
::Spam::SpamConstants::BLOCK_USER].each do |verdict_value|
let(:verdict) { verdict_value }
let(:expected) { verdict_value }
it "returns expected spam constant" do
expect(subject).to eq(expected)
end
end
end
context 'the verdict is an unexpected value' do
let(:verdict) { :this_is_fine }
it 'returns the string' do
expect(subject).to eq verdict
end
end
......@@ -219,11 +258,13 @@ RSpec.describe Spam::SpamVerdictService do
end
end
context 'the HTTP status is not 200' do
let(:spam_check_http_status) { 500 }
context 'the requested is aborted' do
before do
allow(spam_client).to receive(:issue_spam?).and_raise(GRPC::Aborted)
end
it 'returns nil' do
expect(subject).to be_nil
expect(subject).to be(ALLOW)
end
end
......@@ -239,11 +280,11 @@ RSpec.describe Spam::SpamVerdictService do
context 'if the endpoint times out' do
before do
stub_request(:post, /.*spamcheckurl.com.*/).to_timeout
allow(spam_client).to receive(:issue_spam?).and_raise(GRPC::DeadlineExceeded)
end
it 'returns nil' do
expect(subject).to be_nil
expect(subject).to be(ALLOW)
end
end
end
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment