Update cucumber: 2.4.0 → 5.3.0 (major)

[depfu]

Here is everything you need to know about this update. Please take a good look at what changed and the test results before merging this pull request.

What changed?

✳️ cucumber (2.4.0 → 5.3.0) · Repo · Changelog

Release Notes

5.3.0 (from changelog)

Added

• fileattribute cli argument available to attach file to junit formatter

Fixed

• Circle-CI windows build now silently installs MSYS2 using Chocolatey before setting-up the ruby devkit with ridk (#1503 aurelien-reeves)

• --publish and no formatter now uses the pretty formatter per default (#1468 #1500 aurelien-reeves)

5.2.0 (from changelog)

Changed

• --publish uses the response provided by the server as the banner #1472

5.1.2 (from changelog)

Fixed

• Do not send headers after following redirection #1475

5.1.1 (from changelog)

Security fixes

• Update cucumber-create-meta to 2.0.2

5.1.0 (from changelog)

Added

• -X GET in an --out URL will now issue a GET request without a body. If the response is 202 Accepted and the Location header is present, a new PUT request will be sent with the body.

  The main reason for this added behaviour is to allow request bodies larger than 6Mb to be sent while using --publish. This also improves performance since the request body is only sent once (previously it would be sent twice).

Changed

• Set banner border color to green when publishing reports
• Postpone removal of --format=json, embed and puts to version 6.0.0 in deprecation messages

Fixed

• Display banner on stderr when publishing reports #1462

5.0.0 (from changelog)

Added

• --publish automatically publishes reports to reports.cucumber.io
• --publish-quiet does not print information banner about reports.cucumber.io

Changed

• -q, --quiet will also imply --publish-quiet in addition to --no-snippets --no-source --no-duration

Removed

• Dropped support for Ruby 2.3 and 2.4

Fixed

• Update code to be compatible with diff-lcs versions 1.3 and 1.4
• Defer registration of at_exit hook that flushes and closes formatter streams (#1458)
• Updated gems (see git diff for details)
  • cucumber-expressions
  • cucumber-gherkin
  • cucumber-create-meta
  • cucumber-messages
    • Fix issue with timestamp nanos #1438
  • cucumber-html-formatter
    • Add filtering capabilities #1444
• Fix Interceptor that was raising exception when calling puts on the wrapped stream (#1445)

4.1.0 (from changelog)

Changed

• Use cucumber-create-meta to produce the Meta message before the run.

• Updated gems:

  • cucumber-wire ~> 3.1.0
  • cucumber-core ~> 7.1.0
  • cucumber-gherkin ~> 14.0.1
    • Fix issue with empty feature files #1427
  • cucumber-messages ~> 12.2.0
  • cucumber-html-formatter ~> 7.0.0
    • Fix issue with Hook attachments #1420

Fixed

• AfterStep hook do not cause issue when running message formatter. #1433 - #1434

4.0.1 (from changelog)

Fixed

• force reference to diff-lcs to 1.3 as 1.4 introduced breaking changes.

4.0.0 (from changelog)

Changed

• log method can now be called with non-string objects and will run .to_s on them. #1410

Improved

• Display snippet when using undefined parameter type #1411

Does any of this look wrong? Please let us know.

Commits

See the full diff on Github. The new version differs by more commits than we can show here.

↗️ builder (indirect, 3.2.2 → 3.2.4) · Repo · Changelog

↗️ cucumber-core (indirect, 1.5.0 → 8.0.1) · Repo · Changelog

Release Notes

8.0.1 (from changelog)

Fixed

• Make releases from docker using secrets from keybase

7.1.0 (from changelog)

Changed

• Updated gems:
  • cucumber-gherkin ~> 14.0.1
  • cucumber-messages ~> 12.2.0

7.0.0 (from changelog)

Changed

• Updated monorepo libraries:
  • cucumber-gherkin ~> 13
  • cucumber-messages ~> 12

6.0.0 (from changelog)

Changed

• Update to Gherkin 10

Added

• Add envelope event, which are used when emitting Cucumber::Messages
• Add TestCaseCreated and TestStepCreated events, emitted when compiling a Pickle
• Add Id field to TestCase and TestStep
• Added rubocop (with todo file), and removed backports gems (#186, #182 tas50, luke-hill)

Removed

• Remove location for MultiLine arguments

5.0.2 (from changelog)

Changed

• Update to use Gherkin v8

5.0.1 (from changelog)

Removed

• Remove support for ruby 2.2 and below. 2.3 or higher is required now.

5.0.0 (from changelog)

Changed

• Update to use Gherkin v7

Does any of this look wrong? Please let us know.

Commits

See the full diff on Github. The new version differs by more commits than we can show here.

↗️ cucumber-wire (indirect, 0.0.1 → 4.0.1) · Repo

Commits

See the full diff on Github. The new version differs by more commits than we can show here.

↗️ diff-lcs (indirect, 1.2.5 → 1.4.4) · Repo · Changelog

Release Notes

1.4.4 (from changelog)

• Fixed an issue reported by Jun Aruga in the Diff::LCS::Ldiff binary text detection. #44

• Fixed a theoretical issue reported by Jun Aruga in Diff::LCS::Hunk to raise a more useful exception. #43

• Added documentation that should address custom object issues as reported in #35.

• Fixed more diff errors, in part reported in #65.

  • The use of Numeric#abs is incorrect in Diff::LCS::Block#diff_size. The diff size must be accurate for correct change placement.

  • When selecting @max_diff_size in Diff::LCS::Hunk, choose it based on block.diff_size.abs.

  • Made a number of changes that will, unfortunately, increase allocations at the cost of being safe with frozen strings.

  • Add some knowledge that when Diff::LCS::Hunk#diff is called, that we are processing the last hunk, so some changes will be made to how the output is generated.

    • old, ed, and reverse_ed formats have no differences.
    • unified format will report \ No newline at end of file given the correct conditions, at most once. Unified range reporting also differs for the last hunk such that the length of the range is reduced by one.
    • context format will report \No newline at end of file given the correct conditions, up to once per "file". Context range reporting also differs for the last hunk such that the end part of the range is reduced by one to a minimum of one.

• Added a bunch more tests for the cases above, and fixed hunk_spec.rb so that the phrase being compared isn't nonsense French.

• Updated formatting.

• Added a Rake task to assist with manual testing on Ruby 1.8.

1.4.3 (from changelog)

• Fixed several issues with the 1.4 on Rubies older than 2.0. Some of this was providing useful shim functions to Hoe 3.x (which dropped these older Rubies a while ago). Specifically:

  • Removed Array#lazy from a method in Diff::LCS::Hunk.
  • Changed some unit tests to use old-style Symbol-keyed hashes.
  • Changed some unit test helper functions to no longer use keyword parameters, but only a trailing options hash.
  • Made the use of psych dependent on RUBY_VERSION >= 1.9.

  Resolves #63.

1.4.2 (from changelog)

• Camille Drapier fixed a small issue with RuboCop configuration. #59
• Applied another fix (and unit test) to fix an issue for the Chef team. #60, #61

1.4.1 (from changelog)

• Fix an issue where diff sizes could be negative, and they should be. #57, #58

1.4 (from changelog)

• Ruby versions lower than 2.4 are soft-deprecated and will not be run as part of the CI process any longer.
• Akinora MUSHA (knu) added the ability for Diff::LCS::Change objects to be implicitly treated arrays. Originally provided as pull request #47, but it introduced a number of test failures as documented in #48, and remediation of Diff::LCS itself was introduced in #49.
• Resolved #5 with some tests comparing output from system calls to bin/ldiff with some pre-generated output. Resolved #6 with these tests.
• Resolved a previously undetected bin/ldiff issue with --context output not matching diff --context output.
• Resolved an issue with later versions of Ruby not working with an OptParse specification of Numeric; this has been changed to Integer.
• Brandon Fish added truffleruby in #52.
• Fixed two missing classes as reported in #53.

Does any of this look wrong? Please let us know.

Commits

See the full diff on Github. The new version differs by more commits than we can show here.

🆕 activesupport (added, 6.1.1)

🆕 concurrent-ruby (added, 1.1.8)

🆕 cucumber-create-meta (added, 2.0.4)

🆕 cucumber-cucumber-expressions (added, 10.3.0)

🆕 cucumber-gherkin (added, 15.0.2)

🆕 cucumber-html-formatter (added, 9.0.0)

🆕 cucumber-messages (added, 13.2.1)

🆕 cucumber-tag-expressions (added, 2.0.4)

🆕 ffi (added, 1.14.2)

🆕 i18n (added, 1.8.7)

🆕 middleware (added, 0.1.0)

🆕 minitest (added, 5.14.3)

🆕 protobuf-cucumber (added, 3.10.8)

🆕 sys-uname (added, 1.2.2)

🆕 thread_safe (added, 0.3.6)

🆕 tzinfo (added, 2.0.4)

🆕 zeitwerk (added, 2.4.2)

🗑️ gherkin (removed)

🗑️ multi_json (removed)

---

Depfu will automatically keep this PR conflict-free, as long as you don't add any commits to this branch yourself. You can also trigger a rebase manually by commenting with @depfu rebase.

All Depfu comment commands

@​depfu rebase

Rebases against your default branch and redoes this update

@​depfu recreate

Recreates this PR, overwriting any edits that you've made to it

@​depfu merge

Merges this PR once your tests are passing and conflicts are resolved

@​depfu close

Closes this PR and deletes the branch

@​depfu reopen

Restores the branch and reopens this PR (if it's closed)

@​depfu pause

Ignores all future updates for this dependency and closes this PR

@​depfu pause [minor|major]

Ignores all future minor/major updates for this dependency and closes this PR

@​depfu resume

Future versions of this dependency will create PRs again (leaves this PR as is)