path_regex_spec.rb 13.3 KB
Newer Older
1 2 3
# coding: utf-8
require 'spec_helper'

4
describe Gitlab::PathRegex do
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
  # Pass in a full path to remove the format segment:
  # `/ci/lint(.:format)` -> `/ci/lint`
  def without_format(path)
    path.split('(', 2)[0]
  end

  # Pass in a full path and get the last segment before a wildcard
  # That's not a parameter
  # `/*namespace_id/:project_id/builds/artifacts/*ref_name_and_path`
  #    -> 'builds/artifacts'
  def path_before_wildcard(path)
    path = path.gsub(STARTING_WITH_NAMESPACE, "")
    path_segments = path.split('/').reject(&:empty?)
    wildcard_index = path_segments.index { |segment| parameter?(segment) }

    segments_before_wildcard = path_segments[0..wildcard_index - 1]

    segments_before_wildcard.join('/')
  end

  def parameter?(segment)
    segment =~ /[*:]/
  end

  # If the path is reserved. Then no conflicting paths can# be created for any
  # route using this reserved word.
  #
  # Both `builds/artifacts` & `build` are covered by reserving the word
  # `build`
  def wildcards_include?(path)
    described_class::PROJECT_WILDCARD_ROUTES.include?(path) ||
      described_class::PROJECT_WILDCARD_ROUTES.include?(path.split('/').first)
  end

39
  def failure_message(constant_name, migration_helper, missing_words: [], additional_words: [])
40
    missing_words = Array(missing_words)
41
    additional_words = Array(additional_words)
42 43 44
    message = ""
    if missing_words.any?
      message += <<-MISSING
45 46 47
      Found new routes that could cause conflicts with existing namespaced routes
      for groups or projects.

48 49 50
      Nest <#{missing_words.join(', ')}> in a route containing `-`, that way
      we know there will be no conflicts with groups or projects created with those
      paths.
51

52 53
      MISSING
    end
54 55

    if additional_words.any?
56
      message += <<-ADDITIONAL
57
      Is <#{additional_words.join(', ')}> in `#{constant_name}` required?
58 59 60 61 62 63
      If they are really required, update these specs to reflect that.

      ADDITIONAL
    end

    message
64 65
  end

66
  let(:all_non_legacy_routes) do
67 68 69
    route_set = Rails.application.routes
    routes_collection = route_set.routes
    routes_array = routes_collection.routes
70 71 72 73 74 75 76 77 78 79 80 81 82 83

    non_legacy_routes = routes_array.reject do |route|
      route.name.to_s =~ /legacy_(\w*)_redirect/
    end

    non_deprecated_redirect_routes = non_legacy_routes.reject do |route|
      app = route.app
      # `app.app` is either another app, or `self`. We want to find the final app.
      app = app.app while app.try(:app) && app.app != app

      app.is_a?(ActionDispatch::Routing::PathRedirect) && app.block.include?('/-/')
    end

    non_deprecated_redirect_routes.map { |route| route.path.spec.to_s }
84 85
  end

86
  let(:routes_without_format) { all_non_legacy_routes.map { |path| without_format(path) } }
87 88 89 90 91 92

  # Routes not starting with `/:` or `/*`
  # all routes not starting with a param
  let(:routes_not_starting_in_wildcard) { routes_without_format.select { |p| p !~ %r{^/[:*]} } }

  let(:top_level_words) do
Nick Thomas's avatar
Nick Thomas committed
93 94 95 96 97 98 99
    routes_not_starting_in_wildcard
      .map { |route| route.split('/')[1] }
      .concat(ee_top_level_words)
      .concat(files_in_public)
      .concat(Array(API::API.prefix.to_s))
      .compact
      .uniq
100 101
  end

102 103
  let(:ee_top_level_words) do
    ['unsubscribes']
104 105 106 107
  end

  let(:files_in_public) do
    git = Gitlab.config.git.bin_path
108
    tracked = `cd #{Rails.root} && #{git} ls-files public`
109 110 111
      .split("\n")
      .map { |entry| entry.gsub('public/', '') }
      .uniq
112
    tracked + %w(assets uploads)
113 114 115 116 117 118 119 120 121 122 123 124 125
  end

  # All routes that start with a namespaced path, that have 1 or more
  # path-segments before having another wildcard parameter.
  # - Starting with paths:
  #   - `/*namespace_id/:project_id/`
  #   - `/*namespace_id/:id/`
  # - Followed by one or more path-parts not starting with `:` or `*`
  # - Followed by a path-part that includes a wildcard parameter `*`
  # At the time of writing these routes match: http://rubular.com/r/Rv2pDE5Dvw
  STARTING_WITH_NAMESPACE = %r{^/\*namespace_id/:(project_)?id}
  NON_PARAM_PARTS = %r{[^:*][a-z\-_/]*}
  ANY_OTHER_PATH_PART = %r{[a-z\-_/:]*}
126
  WILDCARD_SEGMENT = /\*/
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
  let(:namespaced_wildcard_routes) do
    routes_without_format.select do |p|
      p =~ %r{#{STARTING_WITH_NAMESPACE}/#{NON_PARAM_PARTS}/#{ANY_OTHER_PATH_PART}#{WILDCARD_SEGMENT}}
    end
  end

  # This will return all paths that are used in a namespaced route
  # before another wildcard path:
  #
  # /*namespace_id/:project_id/builds/artifacts/*ref_name_and_path
  # /*namespace_id/:project_id/info/lfs/objects/*oid
  # /*namespace_id/:project_id/commits/*id
  # /*namespace_id/:project_id/builds/:build_id/artifacts/file/*path
  # -> ['builds/artifacts', 'info/lfs/objects', 'commits', 'artifacts/file']
  let(:all_wildcard_paths) do
    namespaced_wildcard_routes.map do |route|
      path_before_wildcard(route)
    end.uniq
  end

  STARTING_WITH_GROUP = %r{^/groups/\*(group_)?id/}
  let(:group_routes) do
    routes_without_format.select do |path|
      path =~ STARTING_WITH_GROUP
    end
  end

  let(:paths_after_group_id) do
    group_routes.map do |route|
      route.gsub(STARTING_WITH_GROUP, '').split('/').first
157
    end.uniq
158 159 160 161 162 163
  end

  describe 'TOP_LEVEL_ROUTES' do
    it 'includes all the top level namespaces' do
      failure_block = lambda do
        missing_words = top_level_words - described_class::TOP_LEVEL_ROUTES
164 165 166
        additional_words = described_class::TOP_LEVEL_ROUTES - top_level_words
        failure_message('TOP_LEVEL_ROUTES', 'rename_root_paths',
                        missing_words: missing_words, additional_words: additional_words)
167 168 169
      end

      expect(described_class::TOP_LEVEL_ROUTES)
170
        .to contain_exactly(*top_level_words), failure_block
171 172 173 174 175 176 177
    end
  end

  describe 'GROUP_ROUTES' do
    it "don't contain a second wildcard" do
      failure_block = lambda do
        missing_words = paths_after_group_id - described_class::GROUP_ROUTES
178 179 180
        additional_words = described_class::GROUP_ROUTES - paths_after_group_id
        failure_message('GROUP_ROUTES', 'rename_child_paths',
                        missing_words: missing_words, additional_words: additional_words)
181 182 183
      end

      expect(described_class::GROUP_ROUTES)
184
        .to contain_exactly(*paths_after_group_id), failure_block
185 186 187 188 189 190 191 192
    end
  end

  describe 'PROJECT_WILDCARD_ROUTES' do
    it 'includes all paths that can be used after a namespace/project path' do
      aggregate_failures do
        all_wildcard_paths.each do |path|
          expect(wildcards_include?(path))
193
            .to be(true), failure_message('PROJECT_WILDCARD_ROUTES', 'rename_wildcard_paths', missing_words: path)
194 195 196 197 198
        end
      end
    end
  end

199 200
  describe '.root_namespace_route_regex' do
    subject { %r{\A#{described_class.root_namespace_route_regex}/\z} }
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287

    it 'rejects top level routes' do
      expect(subject).not_to match('admin/')
      expect(subject).not_to match('api/')
      expect(subject).not_to match('.well-known/')
    end

    it 'accepts project wildcard routes' do
      expect(subject).to match('blob/')
      expect(subject).to match('edit/')
      expect(subject).to match('wikis/')
    end

    it 'accepts group routes' do
      expect(subject).to match('activity/')
    end

    it 'is not case sensitive' do
      expect(subject).not_to match('Users/')
    end

    it 'does not allow extra slashes' do
      expect(subject).not_to match('/blob/')
      expect(subject).not_to match('blob//')
    end
  end

  describe '.full_namespace_path_regex' do
    subject { described_class.full_namespace_path_regex }

    context 'at the top level' do
      context 'when the final level' do
        it 'rejects top level routes' do
          expect(subject).not_to match('admin/')
          expect(subject).not_to match('api/')
          expect(subject).not_to match('.well-known/')
        end

        it 'accepts project wildcard routes' do
          expect(subject).to match('blob/')
          expect(subject).to match('edit/')
          expect(subject).to match('wikis/')
        end

        it 'accepts group routes' do
          expect(subject).to match('activity/')
        end
      end

      context 'when more levels follow' do
        it 'rejects top level routes' do
          expect(subject).not_to match('admin/more/')
          expect(subject).not_to match('api/more/')
          expect(subject).not_to match('.well-known/more/')
        end

        it 'accepts project wildcard routes' do
          expect(subject).to match('blob/more/')
          expect(subject).to match('edit/more/')
          expect(subject).to match('wikis/more/')
          expect(subject).to match('environments/folders/')
          expect(subject).to match('info/lfs/objects/')
        end

        it 'accepts group routes' do
          expect(subject).to match('activity/more/')
        end
      end
    end

    context 'at the second level' do
      context 'when the final level' do
        it 'accepts top level routes' do
          expect(subject).to match('root/admin/')
          expect(subject).to match('root/api/')
          expect(subject).to match('root/.well-known/')
        end

        it 'rejects project wildcard routes' do
          expect(subject).not_to match('root/blob/')
          expect(subject).not_to match('root/edit/')
          expect(subject).not_to match('root/wikis/')
          expect(subject).not_to match('root/environments/folders/')
          expect(subject).not_to match('root/info/lfs/objects/')
        end

        it 'rejects group routes' do
288
          expect(subject).not_to match('root/-/')
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
        end
      end

      context 'when more levels follow' do
        it 'accepts top level routes' do
          expect(subject).to match('root/admin/more/')
          expect(subject).to match('root/api/more/')
          expect(subject).to match('root/.well-known/more/')
        end

        it 'rejects project wildcard routes' do
          expect(subject).not_to match('root/blob/more/')
          expect(subject).not_to match('root/edit/more/')
          expect(subject).not_to match('root/wikis/more/')
          expect(subject).not_to match('root/environments/folders/more/')
          expect(subject).not_to match('root/info/lfs/objects/more/')
        end

        it 'rejects group routes' do
308
          expect(subject).not_to match('root/-/')
309 310 311 312 313 314 315 316 317 318 319 320 321 322
        end
      end
    end

    it 'is not case sensitive' do
      expect(subject).not_to match('root/Blob/')
    end

    it 'does not allow extra slashes' do
      expect(subject).not_to match('/root/admin/')
      expect(subject).not_to match('root/admin//')
    end
  end

323 324
  describe '.project_route_regex' do
    subject { %r{\A#{described_class.project_route_regex}/\z} }
325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340

    it 'accepts top level routes' do
      expect(subject).to match('admin/')
      expect(subject).to match('api/')
      expect(subject).to match('.well-known/')
    end

    it 'rejects project wildcard routes' do
      expect(subject).not_to match('blob/')
      expect(subject).not_to match('edit/')
      expect(subject).not_to match('wikis/')
      expect(subject).not_to match('environments/folders/')
      expect(subject).not_to match('info/lfs/objects/')
    end

    it 'accepts group routes' do
341
      expect(subject).to match('analytics/')
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
    end

    it 'is not case sensitive' do
      expect(subject).not_to match('Blob/')
    end

    it 'does not allow extra slashes' do
      expect(subject).not_to match('/admin/')
      expect(subject).not_to match('admin//')
    end
  end

  describe '.full_project_path_regex' do
    subject { described_class.full_project_path_regex }

    it 'accepts top level routes' do
      expect(subject).to match('root/admin/')
      expect(subject).to match('root/api/')
      expect(subject).to match('root/.well-known/')
    end

    it 'rejects project wildcard routes' do
      expect(subject).not_to match('root/blob/')
      expect(subject).not_to match('root/edit/')
      expect(subject).not_to match('root/wikis/')
      expect(subject).not_to match('root/environments/folders/')
      expect(subject).not_to match('root/info/lfs/objects/')
    end

    it 'accepts group routes' do
372
      expect(subject).to match('root/analytics/')
373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413
    end

    it 'is not case sensitive' do
      expect(subject).not_to match('root/Blob/')
    end

    it 'does not allow extra slashes' do
      expect(subject).not_to match('/root/admin/')
      expect(subject).not_to match('root/admin//')
    end
  end

  describe '.namespace_format_regex' do
    subject { described_class.namespace_format_regex }

    it { is_expected.to match('gitlab-ce') }
    it { is_expected.to match('gitlab_git') }
    it { is_expected.to match('_underscore.js') }
    it { is_expected.to match('100px.com') }
    it { is_expected.to match('gitlab.org') }
    it { is_expected.not_to match('?gitlab') }
    it { is_expected.not_to match('git lab') }
    it { is_expected.not_to match('gitlab.git') }
    it { is_expected.not_to match('gitlab.org.') }
    it { is_expected.not_to match('gitlab.org/') }
    it { is_expected.not_to match('/gitlab.org') }
    it { is_expected.not_to match('gitlab git') }
  end

  describe '.project_path_format_regex' do
    subject { described_class.project_path_format_regex }

    it { is_expected.to match('gitlab-ce') }
    it { is_expected.to match('gitlab_git') }
    it { is_expected.to match('_underscore.js') }
    it { is_expected.to match('100px.com') }
    it { is_expected.not_to match('?gitlab') }
    it { is_expected.not_to match('git lab') }
    it { is_expected.not_to match('gitlab.git') }
  end
end