Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

🚨 [security] Update activesupport: 6.1.4.4 → 6.1.7.3 (patch) #69

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

🚨 [security] Update activesupport: 6.1.4.4 → 6.1.7.3 (patch) #69

wants to merge 1 commit into from

Conversation

depfu[bot]
Copy link
Contributor

@depfu depfu bot commented Mar 15, 2023

🚨 Your current dependencies have known security vulnerabilities 🚨

This dependency update fixes known security vulnerabilities. Please see the details below and assess their impact carefully. We recommend to merge and deploy this as soon as possible!

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?

↗️ activesupport (indirect, 6.1.4.4 → 6.1.7.3) · Repo · Changelog

Security Advisories 🚨

🚨 Possible XSS Security Vulnerability in SafeBuffer#bytesplice

There is a vulnerability in ActiveSupport if the new bytesplice method is called on a SafeBuffer with untrusted user input.
This vulnerability has been assigned the CVE identifier CVE-2023-28120.

Versions Affected: All. Not affected: None Fixed Versions: 7.0.4.3, 6.1.7.3

Impact

ActiveSupport uses the SafeBuffer string subclass to tag strings as html_safe after they have been sanitized.
When these strings are mutated, the tag is should be removed to mark them as no longer being html_safe.

Ruby 3.2 introduced a new bytesplice method which ActiveSupport did not yet understand to be a mutation.
Users on older versions of Ruby are likely unaffected.

All users running an affected release and using bytesplice should either upgrade or use one of the workarounds immediately.

Workarounds

Avoid calling bytesplice on a SafeBuffer (html_safe) string with untrusted user input.

🚨 ReDoS based DoS vulnerability in Active Support’s underscore

There is a possible regular expression based DoS vulnerability in Active
Support. This vulnerability has been assigned the CVE identifier
CVE-2023-22796.

Versions Affected: All
Not affected: None
Fixed Versions: 6.1.7.1, 7.0.4.1

Impact

A specially crafted string passed to the underscore method can cause the
regular expression engine to enter a state of catastrophic backtracking.
This can cause the process to use large amounts of CPU and memory, leading
to a possible DoS vulnerability.

This affects String#underscore, ActiveSupport::Inflector.underscore,
String#titleize, and any other methods using these.

All users running an affected release should either upgrade or use one of the
workarounds immediately.

Workarounds

There are no feasible workarounds for this issue.

Users on Ruby 3.2.0 or greater may be able to reduce the impact by
configuring Regexp.timeout.

Release Notes

6.1.7.3 (from changelog)

6.1.7.2 (from changelog)

  • No changes.

6.1.7.1 (from changelog)

6.1.7 (from changelog)

  • No changes.

6.1.6.1 (from changelog)

  • No changes.

6.1.5.1 (from changelog)

  • Fix and add protections for XSS in ActionView::Helpers and ERB::Util.

    Add the method ERB::Util.xml_name_escape to escape dangerous characters in names of tags and names of attributes, following the specification of XML.

    Álvaro Martín Fraguas

6.1.5 (from changelog)

  • Fix ActiveSupport::Duration.build to support negative values.

    The algorithm to collect the parts of the ActiveSupport::Duration ignored the sign of the value and accumulated incorrect part values. This impacted ActiveSupport::Duration#sum (which is dependent on parts) but not ActiveSupport::Duration#eql? (which is dependent on value).

    Caleb Buxton, Braden Staudacher

  • Time#change and methods that call it (eg. Time#advance) will now return a Time with the timezone argument provided, if the caller was initialized with a timezone argument.

    Fixes #42467.

    Alex Ghiculescu

  • Clone to keep extended Logger methods for tagged logger.

    Orhan Toy

  • assert_changes works on including ActiveSupport::Assertions module.

    Pedro Medeiros

6.1.4.7 (from changelog)

  • No changes.

6.1.4.6 (from changelog)

  • Fix Reloader method signature to work with the new Executor signature

6.1.4.5 (from changelog)

  • No changes.

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.

✳️ activerecord (6.1.4.4 → 6.1.7.3) · Repo · Changelog

Security Advisories 🚨

🚨 Denial of Service Vulnerability in ActiveRecord’s PostgreSQL adapter

There is a potential denial of service vulnerability present in
ActiveRecord’s PostgreSQL adapter.

This has been assigned the CVE identifier CVE-2022-44566.

Versions Affected: All.
Not affected: None.
Fixed Versions: 7.0.4.1, 6.1.7.1

Impact

In ActiveRecord <7.0.4.1 and <6.1.7.1, when a value outside the range for a
64bit signed integer is provided to the PostgreSQL connection adapter, it
will treat the target column type as numeric. Comparing integer values
against numeric values can result in a slow sequential scan resulting in
potential Denial of Service.

Workarounds

Ensure that user supplied input which is provided to ActiveRecord clauses do
not contain integers wider than a signed 64bit representation or floats.

🚨 SQL Injection Vulnerability via ActiveRecord comments

There is a possible vulnerability in ActiveRecord related to the
sanitization of comments. This vulnerability has been assigned the CVE
identifier CVE-2023-22794.

Versions Affected: >= 6.0.0
Not affected: < 6.0.0
Fixed Versions: 6.0.6.1, 6.1.7.1, 7.0.4.1

Impact

Previously the implementation of escaping for comments was insufficient for

If malicious user input is passed to either the annotate query method, the
optimizer_hints query method, or through the QueryLogs interface which
automatically adds annotations, it may be sent to the database with
insufficient sanitization and be able to inject SQL outside of the comment.

In most cases these interfaces won’t be used with user input and users
should avoid doing so.

Example vulnerable code:

Post.where(id: 1).annotate("#{params[:user_input]}")

Post.where(id: 1).optimizer_hints("#{params[:user_input]}")

Example vulnerable QueryLogs configuration (the default configuration is not
vulnerable):

config.active_record.query_log_tags = [
  {
    something: -> { <some value including user input> }
  }
]

All users running an affected release should either upgrade or use one of the
workarounds immediately.

Workarounds

Avoid passing user input to annotate and avoid using QueryLogs configuration
which can include user input.

🚨 Possible RCE escalation bug with Serialized Columns in Active Record

There is a possible escalation to RCE when using YAML serialized columns in
Active Record. This vulnerability has been assigned the CVE identifier
CVE-2022-32224.

Versions Affected: All.
Not affected: None
Fixed Versions: 7.0.3.1, 6.1.6.1, 6.0.5.1, 5.2.8.1

Impact

When serialized columns that use YAML (the default) are deserialized, Rails
uses YAML.unsafe_load to convert the YAML data in to Ruby objects. If an
attacker can manipulate data in the database (via means like SQL injection),
then it may be possible for the attacker to escalate to an RCE.

Impacted Active Record models will look something like this:

class User < ApplicationRecord
  serialize :options       # Vulnerable: Uses YAML for serialization
  serialize :values, Array # Vulnerable: Uses YAML for serialization
  serialize :values, JSON  # Not vulnerable
end

All users running an affected release should either upgrade or use one of the
workarounds immediately.

Releases

The FIXED releases are available at the normal locations.

The released versions change the default YAML deserializer to use
YAML.safe_load, which prevents deserialization of possibly dangerous
objects. This may introduce backwards compatibility issues with existing
data.

In order to cope with that situation, the released version also contains two
new Active Record configuration options. The configuration options are as
follows:

  • config.active_storage.use_yaml_unsafe_load

When set to true, this configuration option tells Rails to use the old
"unsafe" YAML loading strategy, maintaining the existing behavior but leaving
the possible escalation vulnerability in place. Setting this option to true
is not recommended, but can aid in upgrading.

  • config.active_record.yaml_column_permitted_classes

The "safe YAML" loading method does not allow all classes to be deserialized
by default. This option allows you to specify classes deemed "safe" in your
application. For example, if your application uses Symbol and Time in
serialized data, you can add Symbol and Time to the allowed list as follows:

config.active_record.yaml_column_permitted_classes = [Symbol, Date, Time]

Workarounds

There are no feasible workarounds for this issue, but other coders (such as
JSON) are not impacted.

Release Notes

6.1.7.3 (from changelog)

  • No changes.

6.1.7.2 (from changelog)

  • No changes.

6.1.7.1 (from changelog)

  • Make sanitize_as_sql_comment more strict

    Though this method was likely never meant to take user input, it was attempting sanitization. That sanitization could be bypassed with carefully crafted input.

    This commit makes the sanitization more robust by replacing any occurrances of "/" or "/" with "/ " or " /". It also performs a first pass to remove one surrounding comment to avoid compatibility issues for users relying on the existing removal.

    This also clarifies in the documentation of annotate that it should not be provided user input.

    [CVE-2023-22794]

  • Added integer width check to PostgreSQL::Quoting

    Given a value outside the range for a 64bit signed integer type PostgreSQL will treat the column type as numeric. Comparing integer values against numeric values can result in a slow sequential scan.

    This behavior is configurable via ActiveRecord::Base.raise_int_wider_than_64bit which defaults to true.

    [CVE-2022-44566]

6.1.7 (from changelog)

  • Symbol is allowed by default for YAML columns

    Étienne Barrié

  • Fix ActiveRecord::Store to serialize as a regular Hash

    Previously it would serialize as an ActiveSupport::HashWithIndifferentAccess which is wasteful and cause problem with YAML safe_load.

    Jean Boussier

  • Fix PG.connect keyword arguments deprecation warning on ruby 2.7

    Fixes #44307.

    Nikita Vasilevsky

6.1.6.1 (from changelog)

  • Change ActiveRecord::Coders::YAMLColumn default to safe_load

    This adds two new configuration options The configuration options are as follows:

    • config.active_storage.use_yaml_unsafe_load

    When set to true, this configuration option tells Rails to use the old "unsafe" YAML loading strategy, maintaining the existing behavior but leaving the possible escalation vulnerability in place. Setting this option to true is not recommended, but can aid in upgrading.

    • config.active_record.yaml_column_permitted_classes

    The "safe YAML" loading method does not allow all classes to be deserialized by default. This option allows you to specify classes deemed "safe" in your application. For example, if your application uses Symbol and Time in serialized data, you can add Symbol and Time to the allowed list as follows:

    config.active_record.yaml_column_permitted_classes = [Symbol, Date, Time]

    [CVE-2022-32224]

6.1.5.1 (from changelog)

  • No changes.

6.1.5 (from changelog)

  • Fix ActiveRecord::ConnectionAdapters::SchemaCache#deep_deduplicate for Ruby 2.6.

    Ruby 2.6 and 2.7 have slightly different implementations of the String#@- method. In Ruby 2.6, the receiver of the String#@- method is modified under certain circumstances. This was later identified as a bug (https://bugs.ruby-lang.org/issues/15926) and only fixed in Ruby 2.7.

    Before the changes in this commit, the ActiveRecord::ConnectionAdapters::SchemaCache#deep_deduplicate method, which internally calls the String#@- method, could also modify an input string argument in Ruby 2.6 -- changing a tainted, unfrozen string into a tainted, frozen string.

    Fixes #43056

    Eric O'Hanlon

  • Fix migration compatibility to create SQLite references/belongs_to column as integer when migration version is 6.0.

    reference/belongs_to in migrations with version 6.0 were creating columns as bigint instead of integer for the SQLite Adapter.

    Marcelo Lauxen

  • Fix dbconsole for 3-tier config.

    Eileen M. Uchitelle

  • Better handle SQL queries with invalid encoding.

    Post.create(name: "broken \xC8 UTF-8")

    Would cause all adapters to fail in a non controlled way in the code responsible to detect write queries.

    The query is now properly passed to the database connection, which might or might not be able to handle it, but will either succeed or failed in a more correct way.

    Jean Boussier

  • Ignore persisted in-memory records when merging target lists.

    Kevin Sjöberg

  • Fix regression bug that caused ignoring additional conditions for preloading has_many through relations.

    Fixes #43132

    Alexander Pauly

  • Fix ActiveRecord::InternalMetadata to not be broken by config.active_record.record_timestamps = false

    Since the model always create the timestamp columns, it has to set them, otherwise it breaks various DB management tasks.

    Fixes #42983

    Jean Boussier

  • Fix duplicate active record objects on inverse_of.

    Justin Carvalho

  • Fix duplicate objects stored in has many association after save.

    Fixes #42549.

    Alex Ghiculescu

  • Fix performance regression in CollectionAssocation#build.

    Alex Ghiculescu

  • Fix retrieving default value for text column for MariaDB.

    fatkodima

6.1.4.7 (from changelog)

  • No changes.

6.1.4.6 (from changelog)

  • No changes.

6.1.4.5 (from changelog)

  • No changes.

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.

✳️ minitest (5.15.0 → 5.18.0) · Repo · Changelog

Release Notes

5.18.0 (from changelog)

  • 2 major enhancements:

    • Added assert_pattern & refute_pattern for pattern matching. (flavorjones)

    • Added matching must_pattern_match & wont_pattern_match to minitest/spec.

  • 1 bug fix:

    • Support the new message format of NameError in Ruby 3.3 (mame)

5.17.0 (from changelog)

  • 1 minor enhancement:

    • Refactor setup hooks into a SETUP_METHODS constant. (MSP-Greg)

  • 3 bug fixes:

    • Fix kwargs for Mock calls to delegator. (blowmage)

    • Fix kwargs for expectations. (bobmazanec, blowmage)

    • Remove check for .b method. (tenderlove)

5.16.3 (from changelog)

  • 2 bug fixes:

    • Fixed exception sanitization by removing TypeError restriction on rescue.

    • Use A instead of deprecated TESTOPTS in rake test:slow. (davidstosik)

5.16.2 (from changelog)

  • 4 bug fixes:

    • Added MT_KWARGS_HACK kludge for stub to deal with ruby 2.7 kwargs nastiness. (tsugimoto)

    • In #expect, pop Hash class from args if $MT_KWARGS_HACK. (casperisfine)

    • In above scenario, set expected kwargs (as Objects) based on actual kwargs.

    • Nuke ivars if exception fails to marshal twice (eg better_errors). (irphilli)

5.16.1 (from changelog)

  • 2 bug fixes:

    • Apparently adding real kwarg support to mocks/stubs broke some code. Fixed.

      • Use `MT_KWARGS_HACK=1` to activate the kludgy kwargs support w/ caveats.

    • Clarified some doco wrt the block on #stub.

5.16.0 (from changelog)

  • 2 major enhancements:

    • Added Minitest::TestTask.

    • Dropping ruby 2.2 - 2.5. 2.6 is DTM soon too.

  • 11 minor enhancements:

    • Added –show-skips option to show skips at end of run but not require –verbose. (MSP-Greg)

    • Added Minitest.seed, the random seed used by the run.

    • Calling `srand Minitest.seed` before all shuffles to ensure determinism.

    • Extended #stub to handle kwargs for both block and call args. (SampsonCrowley)

    • Extended Mock#__call to display kwargs.

    • Extended Mock#expect to record kwargs.

    • Extended Mock#method_missing to take kwargs & compare them against expected.

    • Mock#method_missing displays better errors on arity mismatch.

    • Removed minor optimization removing empty suites before run.

    • Simplified test randomization (test order will change even with fixed seed).

    • assert_match now returns the MatchData on success. (Nakilon)

  • 3 bug fixes:

    • (Re)Fixed marshalling of exceptions, neutering them in 2 passes.

    • Fixed more problems with rdoc.

    • Had to patch up mock and stub to deal with <=2.7 kwargs oddities

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.

↗️ activemodel (indirect, 6.1.4.4 → 6.1.7.3) · Repo · Changelog

Release Notes

6.1.7.3 (from changelog)

  • No changes.

6.1.7.2 (from changelog)

  • No changes.

6.1.7.1 (from changelog)

  • No changes.

6.1.7 (from changelog)

  • No changes.

6.1.6.1 (from changelog)

  • No changes.

6.1.5.1 (from changelog)

  • No changes.

6.1.5 (from changelog)

  • Clear secure password cache if password is set to nil

    Before:

    user.password = 'something' user.password = nil

    user.password # => 'something'

    Now:

    user.password = 'something' user.password = nil

    user.password # => nil

    Markus Doits

  • Fix delegation in ActiveModel::Type::Registry#lookup and ActiveModel::Type.lookup

    Passing a last positional argument {} would be incorrectly considered as keyword argument.

    Benoit Daloze

  • Fix to_json after changes_applied for ActiveModel::Dirty object.

    Ryuta Kamizono

6.1.4.7 (from changelog)

  • No changes.

6.1.4.6 (from changelog)

  • No changes.

6.1.4.5 (from changelog)

  • No changes.

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.

↗️ concurrent-ruby (indirect, 1.1.9 → 1.2.2) · Repo · Changelog

Release Notes

1.2.2

concurrent-ruby 1.2.2:

  • (#993) Fix arguments passed to Concurrent::Map's default_proc.

1.2.1

concurrent-ruby 1.2.1:

  • (#990) Add missing require 'fiber' for FiberLocalVar.
  • (#989) Optimize Concurrent::Map#[] on CRuby by letting the backing Hash handle the default_proc.

1.2.0

concurrent-ruby 1.2.0:

  • (#975) Set the Ruby compatibility version at 2.3
  • (#962) Fix ReentrantReadWriteLock to use the same granularity for locals as for Mutex it uses.
  • (#983) Add FiberLocalVar
  • (#934) concurrent-ruby now supports requiring individual classes (public classes listed in the docs), e.g., require 'concurrent/map'
  • (#976) Let Promises.any_fulfilled_future take an Event
  • Improve documentation of various classes
  • (#972) Remove Rubinius-related code

concurrent-ruby-edge 0.7.0:

  • (#975) Set the Ruby compatibility version at 2.3
  • (#934) concurrent-ruby now supports requiring individual classes (public classes listed in the docs), e.g., require 'concurrent/map'
  • (#972) Remove Rubinius-related code

1.1.10

concurrent-ruby:

  • (#951) Set the Ruby compatibility version at 2.2
  • (#939, #933) The caller_runs fallback policy no longer blocks reads from the job queue by worker threads
  • (#938, #761, #652) You can now explicitly prune_pool a thread pool (Sylvain Joyeux)
  • (#937, #757, #670) We switched the Yahoo stock API for demos to Alpha Vantage (Gustavo Caso)
  • (#932, #931) We changed how SafeTaskExecutor handles local jump errors (Aaron Jensen)
  • (#927) You can use keyword arguments in your initialize when using Async (Matt Larraz)
  • (#926, #639) We removed timeout from TimerTask because it wasn't sound, and now it's a no-op with a warning (Jacob Atzen)
  • (#919) If you double-lock a re-entrant read-write lock, we promote to locked for writing (zp yuan)
  • (#915) monotonic_time now accepts an optional unit parameter, as Ruby's clock_gettime (Jean Boussier)

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.

↗️ i18n (indirect, 1.8.11 → 1.12.0) · Repo · Changelog

Release Notes

1.12.0

What's Changed

  • Revert "Add support for CLDR data in I18n::Backend::Pluralization" by @radar in #633 -- this was causing breaking changes unintentionally.

Full Changelog: v1.11.0...v1.12.0

1.11.0

What's Changed

New Contributors

Full Changelog: v1.10.0...v1.11.0

1.10.0

What's Changed

New Features

Bug fixes

Other changes

  • Remove pry from Gemfile as it is not used by @dvzrv in #608

New Contributors

Full Changelog: v1.9.1...v1.10.0

1.9.1

What's Changed

Full Changelog: v1.9.0...v1.9.1

1.9.0

Minor version bump: The number of changes in this release are more than I would feel comfortable including in a point release. Therefore, I have bumped the minor version number here. -- @radar

What's Changed

New Contributors

Full Changelog: v1.8.11...v1.9.0

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.

↗️ tzinfo (indirect, 2.0.4 → 2.0.6) · Repo · Changelog

Release Notes

2.0.6

  • Eliminate Object#untaint deprecation warnings on JRuby 9.4.0.0. #145.

TZInfo v2.0.6 on RubyGems.org

2.0.5

  • Changed DateTime results to always use the proleptic Gregorian calendar. This affects DateTime results prior to 1582-10-15 and any arithmetic performed on the results that would produce a secondary result prior to 1582-10-15.
  • Added support for eager loading all the time zone and country data by calling either TZInfo::DataSource#eager_load! or TZInfo.eager_load!. Compatible with Ruby On Rails' eager_load_namespaces. #129.
  • Ignore the SECURITY file from Arch Linux's tzdata package. #134.

TZInfo v2.0.5 on RubyGems.org

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.

↗️ zeitwerk (indirect, 2.5.1 → 2.6.7) · Repo · Changelog

Release Notes

2.6.7 (from changelog)

  • Reset module state on Zeitwerk::NameError.

    If an autoload is triggered, the file is loaded successfully, but the expected constant does not get defined, Ruby resets the state of the module. In particular, autoload? returns nil for that constant name, and constants does not include the constant name (starting with Ruby 3.1).

    Zeitwerk is more strict, not defining the expected constant is an error condition and the loader raises Zeitwerk::NameError. But this happens during the require call and the exception prevents Ruby from doing that cleanup.

    With this change, the parent module is left in a state that makes more sense and is consistent with what Ruby does.

  • A message is logged if an autoload did not define the expected constant.

    When that happens, Zeitwerk::NameError is raised and you normally see the exception. But if the error is shallowed, and you are inspecting the logs to investigate something, this new message may be helpful.

  • By default, Zeitwerk::Loader#dirs filters ignored root directories out. Please, pass ignored: true if you want them included.

    It is very strange to configure a root directory and also ignore it, the edge case is supported only for completeness. However, in that case, client code listing root directories rarely needs the ignored ones.

  • Documentation improvements.

  • Enforcement of private interfaces continues with another gradual patch.

2.6.6 (from changelog)

  • The new eager_load_namespace had a bug when eager loading certain namespaces with collapsed directories. This has been fixed.

2.6.5 (from changelog)

  • Controlled errors in a couple of situations:

    • Attempting to eager load or reload without previously invoking setup now raises Zeitwerk::SetupRequired.

    • The method Zeitwerk::Loader#push_dir raises Zeitwerk::Error if it gets an anonymous custom namespace.

    These should be backwards compatible, because they raise in circumstances that didn't work anyway. The goal here is to provide a meaningful error upfront.

  • Enforcement of private interfaces continues with another gradual patch.

2.6.4 (from changelog)

Ruby does not have gem-level visibility, so sometimes you need things to be public for them to be accessible internally. But they do not belong to the public interface of the gem.

A method that is undocumented and marked as @private in the source code is clearly private API, regardless of its formal Ruby visibility.

This release starts a series of gradual patches in which private interface is enforced with stricter formal visibility.

2.6.3 (from changelog)

  • v2.6.2 introduced a regression in the logic that checks whether two loaders want to manage the same root directories. It has been fixed.

2.6.2 (from changelog)

  • Zeitwerk::Loader#load_file allows you to load an individual Ruby file. Check its documentation for details.

  • Zeitwerk::Loader#eager_load_dir allows you to eager load a directory, recursively. Check its documentation for details.

  • Zeitwerk::Loader#eager_load_namespace allows you to eager a namespace, recursively. Namespaces are global, this method loads only what the receiver manages from that namespace, if anything. Check its documentation for details.

  • Zeitwerk::Loader.eager_load_namespace broadcasts eager_load_namespace to all registered loaders. Check its documentation for details.

  • Documents shadowed files. They always existed, but were not covered by the documentation.

  • Other assorted documentation improvements.

2.6.1 (from changelog)

  • Zeitwerk::Loader#dirs allows you to instrospect the root directories configured in the receiver. Please check its documentation for details.

2.6.0 (from changelog)

  • Directories are processed in lexicographic order.

    Different file systems may list directories in different order, and with this change we ensure that client code eager loads consistently across platforms, for example.

  • Before this release, subdirectories of root directories always represented namespaces (unless ignored or collapsed). From now on, to be considered namespaces they also have to contain at least one non-ignored Ruby file with extension .rb, directly or recursively.

    If you know beforehand a certain directory or directory pattern does not represent a namespace, it is intentional and more efficient to tell Zeitwerk to ignore it.

    However, if you don't do so and have a directory tasks that only contains Rake files, arguably that directory is not meant to represent a Ruby module. Before, Zeitwerk would define a top-level Tasks module after it; now, it does not.

    This feature is also handy for projects that have directories with auxiliary resources mixed in the project tree in a way that is too dynamic for an ignore pattern to be practical. See #216.

    In the unlikely case that an existing project has an empty directory for the sole purpose of defining a totally empty module (no code, and no nested classes or modules), such module has now to be defined in a file.

    Directories are scanned again on reloads.

  • On setup, loaders created with Zeitwerk::Loader.for_gem issue warnings if lib has extra, non-ignored Ruby files or directories.

    This is motivated by existing gems with directories under lib that are not meant to define Ruby modules, like directories for Rails generators, for instance.

    This warning can be silenced in the unlikely case that the extra stuff is actually autoloadable and has to be managed by Zeitwerk.

    Please, check the documentation for further details.

    This method returns an instance of a private subclass of Zeitwerk::Loader now, but you cannot rely on the type, just on the interface.

2.5.4 (from changelog)

  • If a file did not define the expected constant, there was a reload, and there were on_unload callbacks, Zeitwerk still tried to access the constant during reload, which raised. This has been corrected.

2.5.3 (from changelog)

  • The change introduced in 2.5.2 implied a performance regression that was particularly dramatic in Ruby 3.1. We'll address #198 in a different way.

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.

Depfu Status

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)

@depfu depfu bot added the depfu label Mar 15, 2023
@codeclimate
Copy link

codeclimate bot commented Mar 15, 2023

Code Climate has analyzed commit fdcf2f5 and detected 0 issues on this pull request.

View more on Code Climate.

@depfu depfu bot mentioned this pull request Mar 15, 2023
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
depfu
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
0 participants