This instance will be upgraded to Heptapod 0.27.1 on 2021-12-08 at 22:00 UTC+1 (a few minutes of down time)

path_regex_spec.rb 15.3 KB
Newer Older
1
2
# frozen_string_literal: true

3
4
require 'spec_helper'

5
RSpec.describe Gitlab::PathRegex do
6
7
8
9
10
  let(:starting_with_namespace) { %r{^/\*namespace_id/:(project_)?id} }
  let(:non_param_parts) { %r{[^:*][a-z\-_/]*} }
  let(:any_other_path_part) { %r{[a-z\-_/:]*} }
  let(:wildcard_segment) { /\*/ }

11
12
13
14
15
16
17
18
19
20
21
  # 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)
22
    path = path.gsub(starting_with_namespace, "")
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
    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

Bob Van Landuyt's avatar
Bob Van Landuyt committed
45
  def failure_message(constant_name, migration_helper, missing_words: [], additional_words: [])
46
    missing_words = Array(missing_words)
47
    additional_words = Array(additional_words)
Bob Van Landuyt's avatar
Bob Van Landuyt committed
48
49
50
    message = ""
    if missing_words.any?
      message += <<-MISSING
51
52
53
      Found new routes that could cause conflicts with existing namespaced routes
      for groups or projects.

54
55
56
      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.
57

Bob Van Landuyt's avatar
Bob Van Landuyt committed
58
59
      MISSING
    end
60
61

    if additional_words.any?
Bob Van Landuyt's avatar
Bob Van Landuyt committed
62
      message += <<-ADDITIONAL
63
      Is <#{additional_words.join(', ')}> in `#{constant_name}` required?
64
65
66
67
68
69
      If they are really required, update these specs to reflect that.

      ADDITIONAL
    end

    message
70
71
  end

72
  let(:all_non_legacy_routes) do
73
74
75
    route_set = Rails.application.routes
    routes_collection = route_set.routes
    routes_array = routes_collection.routes
76
77
78
79
80
81
82
83
84
85
86
87
88
89

    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 }
90
91
  end

92
  let(:routes_without_format) { all_non_legacy_routes.map { |path| without_format(path) } }
93
94
95
96
97
98

  # 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
99
100
101
102
103
104
105
    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
106
107
  end

Bob Van Landuyt's avatar
Bob Van Landuyt committed
108
  let(:ee_top_level_words) do
109
    %w(unsubscribes v2)
110
111
112
  end

  let(:files_in_public) do
113
114
115
116
    hg = Gitlab.config.mercurial.bin_path
    # If it becomes a problem not to have some extensions (evolve etc),
    # we can do this hg call more properly with HGRCPATH etc.
    tracked = `cd #{Rails.root} && #{hg} files public`
117
      .split("\n")
118
      .map { |entry| entry.start_with?('public/-/') ? '-' : entry.gsub('public/', '') }
119
      .uniq
120
    tracked + %w(assets uploads)
121
122
123
124
125
126
127
128
129
130
131
132
  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
  let(:namespaced_wildcard_routes) do
    routes_without_format.select do |p|
133
      p =~ %r{#{starting_with_namespace}/#{non_param_parts}/#{any_other_path_part}#{wildcard_segment}}
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
    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

151
  let(:starting_with_group) { %r{^/groups/\*(group_)?id/} }
152
  let(:group_routes) do
153
    routes_without_format.grep(starting_with_group)
154
155
156
157
  end

  let(:paths_after_group_id) do
    group_routes.map do |route|
158
      route.gsub(starting_with_group, '').split('/').first
159
    end.uniq
160
161
162
163
164
165
  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
166
167
168
        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)
169
170
171
      end

      expect(described_class::TOP_LEVEL_ROUTES)
172
        .to contain_exactly(*top_level_words), failure_block
173
    end
174
175
176
177
178

    # We ban new items in this list, see https://gitlab.com/gitlab-org/gitlab/-/issues/215362
    it 'does not allow expansion' do
      expect(described_class::TOP_LEVEL_ROUTES.size).to eq(41)
    end
179
180
181
182
183
184
  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
Bob Van Landuyt's avatar
Bob Van Landuyt committed
185
186
187
        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)
188
189
190
      end

      expect(described_class::GROUP_ROUTES)
Bob Van Landuyt's avatar
Bob Van Landuyt committed
191
        .to contain_exactly(*paths_after_group_id), failure_block
192
    end
193
194
195
196
197

    # We ban new items in this list, see https://gitlab.com/gitlab-org/gitlab/-/issues/215362
    it 'does not allow expansion' do
      expect(described_class::GROUP_ROUTES.size).to eq(1)
    end
198
199
200
201
202
203
204
  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))
205
            .to be(true), failure_message('PROJECT_WILDCARD_ROUTES', 'rename_wildcard_paths', missing_words: path)
206
207
208
        end
      end
    end
209
210
211
212
213

    # We ban new items in this list, see https://gitlab.com/gitlab-org/gitlab/-/issues/215362
    it 'does not allow expansion' do
      expect(described_class::PROJECT_WILDCARD_ROUTES.size).to eq(21)
    end
214
215
  end

216
217
  describe '.root_namespace_route_regex' do
    subject { %r{\A#{described_class.root_namespace_route_regex}/\z} }
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
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304

    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
305
          expect(subject).not_to match('root/-/')
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
        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
325
          expect(subject).not_to match('root/-/')
326
327
328
329
330
331
332
333
334
335
336
337
338
339
        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

340
341
  describe '.project_route_regex' do
    subject { %r{\A#{described_class.project_route_regex}/\z} }
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357

    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
358
      expect(subject).to match('analytics/')
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
    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
389
      expect(subject).to match('root/analytics/')
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
    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
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462

  shared_examples 'invalid snippet routes' do
    it { is_expected.not_to match('gitlab-org/gitlab/snippets/1.git') }
    it { is_expected.not_to match('snippets/1.git') }
    it { is_expected.not_to match('gitlab-org/gitlab/snippets/') }
    it { is_expected.not_to match('/gitlab-org/gitlab/snippets/1') }
    it { is_expected.not_to match('gitlab-org/gitlab/snippets/foo') }
    it { is_expected.not_to match('root/snippets/1') }
    it { is_expected.not_to match('/snippets/1') }
    it { is_expected.not_to match('snippets/') }
    it { is_expected.not_to match('snippets/foo') }
  end

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

    it { is_expected.to match('gitlab-org/gitlab/snippets/1') }
    it { is_expected.to match('snippets/1') }

    it_behaves_like 'invalid snippet routes'
  end

  describe '.personal_and_project_snippets_path_regex' do
    subject { %r{\A#{described_class.personal_and_project_snippets_path_regex}\z} }

    it { is_expected.to match('gitlab-org/gitlab/snippets') }
    it { is_expected.to match('snippets') }

    it { is_expected.not_to match('gitlab-org/gitlab/snippets/1') }
    it { is_expected.not_to match('snippets/1') }

    it_behaves_like 'invalid snippet routes'
  end
463
end