[pull] master from ruby:master#1173
Merged
Merged
Conversation
Bundler::DSLError#to_s read the offending line number from
`trace_line.split(":")[1]`. On Windows the backtrace path carries a drive
letter such as `C:`, so that field is the path rather than the number, and
the negative index that follows garbles the source snippet. Match the number
that sits right before `:in` or the end of the line instead.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Cover behavior documented in rdoc but not asserted in test/ruby/test_string.rb: * insert returns self * boundary negative index -(length + 1) prepends * out-of-range index (positive and negative) raises IndexError * non-string argument raises TypeError * frozen string raises FrozenError * multibyte: index counts characters, not bytes
These guards no longer match current mswin behavior. The affected tests pass reliably on an x64-mswin64_140 build, repeatedly and under -j16 for the flaky-history ones. Only the mswin conditions are removed. test_ractor and test_transcode keep their darwin and s390x guards. test_thread_queue keeps the MinGW omit and drops only the retired AppVeyor branch. test_box re-enables box_1 only, since box_2 still crashes at process exit under mswin and stays pended. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Loading the extension libraries in a box exceeds the default 10-second assert_separately timeout under -j16 CI load. Use 60 seconds, matching the sibling box test that already does. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Built only with --enable-libfyaml; without the flag the default libyaml backend is unchanged. The parser and emitter are reimplemented against libfyaml's event API in separate translation units guarded by PSYCH_USE_LIBFYAML, and the backend is not supported on Windows. ruby/psych@51f2493a2e Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Scalar type resolution happens in ScalarScanner, not the C backend, so swapping to libfyaml alone still resolved yes/no/on/off to booleans. Key the boolean set on Psych::BACKEND so the libyaml default keeps the YAML 1.1 set while the experimental libfyaml backend follows 1.2. ruby/psych@b374869c59 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Create a fresh parser per parse instead of reusing one via
fy_parser_reset(), which left default tag handles unset and rejected
bare ("---"-less) tag-led documents. Recover the real message and
position by switching the parser's own diagnostic object to collect
mode; creating a replacement diag crashes libfyaml 0.9.6. Drop the
spurious empty tag directive libfyaml reports.
ruby/psych@d383361c23
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Drop the tag when plain_implicit or quoted_implicit is set, matching how libyaml omits a tag that the value resolves to on reload; otherwise nil emitted as "!<tag:yaml.org,2002:null>" instead of an empty scalar. Honor the plain hint when choosing the scalar style, and restore the Check_Type guards on anchor and tag so non-string arguments raise TypeError. ruby/psych@ff15ddf399 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
libfyaml only consumes UTF-8, so a UTF-16 IO fed through the chunked reader reached it as raw bytes and was rejected as invalid UTF-8. When the IO's external encoding is UTF-16LE/BE, slurp the whole stream and transcode it first; a 2-byte unit could otherwise straddle a read boundary. Other non-UTF-8 encodings stay raw and libfyaml rejects them, matching psych's UTF-8/UTF-16-only IO contract (Shift_JIS still raises). ruby/psych@4bd1343551 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…cument The xcalloc'd tag buffers leaked when StringValue, the tuple-length check, or the emit raised mid-way. Wrap the work in rb_ensure so the buffers are always freed, and keep the exported directive strings in a Ruby array so the GC cannot reclaim them while their C pointers are in use. ruby/psych@6fdede0177 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a libfyaml? helper and guard the tests that intentionally diverge on the experimental YAML 1.2 backend: yes/no/on/off are strings rather than booleans, flow collections and block scalars are formatted differently, tags and marks are reported differently, and non-ASCII tags/aliases are rejected. test_boolean asserts the 1.2 string result directly; the formatting and mark cases are skipped. The default libyaml build is unaffected (every guard keys off Psych::BACKEND). ruby/psych@67b6365f5a Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Mark the reconstructed SyntaxError message UTF-8 instead of US-ASCII, so a diagnostic that embeds a multibyte snippet of the input does not raise Encoding::CompatibilityError when concatenated with UTF-8. Add RB_GC_GUARD for the anchor and tag strings in the emitter (matching the existing guard on the scalar value) so their C pointers cannot dangle if a GC runs inside fy_emit_event_create. ruby/psych@1c73f43df1 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Cover what the libfyaml backend does distinctly, each placed with the concern it belongs to: the Psych::BACKEND and Psych.libfyaml_version checks in test_psych.rb, the YAML 1.2 "Norway problem" boolean case in test_boolean.rb, and the "1.1 booleans are not quoted" emission case in test_string.rb. The 1.2 assertions are skipped on the default libyaml backend so the same suite passes under both. ruby/psych@35d0a437a2 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Reject canonical output with NotImplementedError instead of silently ignoring the request, and honor the implicit flag in start_sequence and start_mapping so an implicit tag is not printed as a redundant verbose tag. Relax the unquoted-boolean dump test to allow an optional document end marker. Also correct the version docstrings and a stale comment about how the parser collects diagnostics. ruby/psych@23f60daff1 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The libfyaml-dev package on Ubuntu is 0.8, which segfaults psych's emitter, while 0.9.6 (used on macOS via Homebrew) passes the whole suite. Build the same 0.9.6 release from source in the Linux CI job, and reject libfyaml older than 0.9 in extconf so users get a clear error instead of a runtime crash. Also drop an unused variable in set_canonical. ruby/psych@88ac4a0cc0 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
… integration connection tests
Create spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb
adding non-PQC and PQC server/client connection integration tests.
As "Bundler::Fetcher local SSL server #connection PQC connects with client cert
auth" failed with the following error due to hardcoded `OpenSSL::PKey::RSA.new` in
`Bundler::Fetcher#connection`, fixed it to support ML-DSA ssl_client_cert.
```
$ bin/rspec spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb
...
Failures:
1) Bundler::Fetcher local SSL server #connection PQC connects with client cert auth
Failure/Error: fetcher = Bundler::Fetcher.new(remote)
OpenSSL::PKey::PKeyError:
incorrect pkey type: UNDEF
# /home/jaruga/.local/ruby-4.1.0-debug-3ef48ef9c8-openssl-4.1.0-7194354488/lib/ruby/4.1.0+1/openssl/pkey.rb:394:in 'OpenSSL::PKey::RSA#initialize'
# /home/jaruga/.local/ruby-4.1.0-debug-3ef48ef9c8-openssl-4.1.0-7194354488/lib/ruby/4.1.0+1/openssl/pkey.rb:394:in 'Class#new'
# /home/jaruga/.local/ruby-4.1.0-debug-3ef48ef9c8-openssl-4.1.0-7194354488/lib/ruby/4.1.0+1/openssl/pkey.rb:394:in 'OpenSSL::PKey::RSA.new'
# ./bundler/lib/bundler/fetcher.rb:321:in 'Bundler::Fetcher#connection'
# ./bundler/lib/bundler/fetcher.rb:140:in 'Bundler::Fetcher#initialize'
# ./spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb:69:in 'RSpec::ExampleGroups::BundlerFetcherLocalSSLServer#fetch_path'
# ./spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb:60:in 'block (4 levels) in <top (required)>'
...
```
Create test/rubygems/local_ssl_server_utilities.rb to manage utility methods
called by RubyGems test-unit and Bundler rspec tests.
Assisted-by: Claude:claude-opus-4-6[1m]
ruby/rubygems@6aea5fcc17
…:HTTP
The hardcoded Gem::Net::HTTP in Artifice.deactivate, is actually
the replaced Artifice::Net::HTTP. This doesn't restore the original
Gem::Net::HTTP.
Restore the saved original Gem::Net::HTTP in Artifice.deactivate
This issue caused the tests sometimes failed in
spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb
using Artifice::Net::HTTP unintentionally. Because localhost is not included in
spec/support/artifice/helpers/endpoint.rb - permitted_hosts caused HTTP status 403
(response.body: "Host not permitted"). But Gem::Net::HTTP should be used in the
tests instead of Artifice::Net::HTTP. Possibly this issue happens when
Artifice.deactivate is called in other tests such as spec/commands/ssl_spec.rb.
That's why this issue sometimes happened rather than always.
```
1) Bundler::Fetcher local SSL server #connection PQC connects
Failure/Error: expect(response.code).to eq("200")
expected: #<Encoding:UTF-8> "200"
got: #<Encoding:ASCII-8BIT> "403"
(compared using ==)
# ./spec/bundler/fetcher/gem_remote_fetcher_local_ssl_server_spec.rb:49:in 'block (4 levels) in <top (required)>'
# ./spec/spec_helper.rb:164:in 'block (4 levels) in <top (required)>'
# ./spec/spec_helper.rb:164:in 'block (3 levels) in <top (required)>'
# ./spec/support/helpers.rb:414:in 'block in Spec::Helpers#with_gem_path_as'
# ./spec/support/helpers.rb:428:in 'Spec::Helpers#without_env_side_effects'
# ./spec/support/helpers.rb:409:in 'Spec::Helpers#with_gem_path_as'
# ./spec/spec_helper.rb:163:in 'block (2 levels) in <top (required)>'
# ./lib/rubygems.rb:306:in 'Kernel#load'
# ./lib/rubygems.rb:306:in 'Gem.activate_and_load_bin_path'
```
spec/commands/ssl_spec.rb
```
...
after(:each) do
...
Artifice.deactivate
...
end
...
```
Assisted-by: Claude:claude-opus-4-6[1m]
ruby/rubygems@0ae3860638
…plicitly.
This commit fixes the following failures.
The `require "support/artifice/compact_index"` is not enough
because the file is not loaded when it is called second time.
```
1) Bundler::ParallelInstaller connect to make jobserver takes all available slots
Failure/Error: raise HTTPError, e.message
Bundler::HTTPError:
Could not reach host gem.repo2. Check your network connection and try again.
# ./spec/bundler/installer/parallel_installer_spec.rb:151:in 'block (4 levels) in <top (required)>'
# ./spec/bundler/installer/parallel_installer_spec.rb:221:in 'RSpec::ExampleGroups::BundlerParallelInstaller::ConnectToMakeJobserver#redefine_build_jobs'
# ./spec/bundler/installer/parallel_installer_spec.rb:150:in 'block (3 levels) in <top (required)>'
# ./spec/spec_helper.rb:164:in 'block (4 levels) in <top (required)>'
# ./spec/spec_helper.rb:164:in 'block (3 levels) in <top (required)>'
# ./spec/support/helpers.rb:414:in 'block in Spec::Helpers#with_gem_path_as'
# ./spec/support/helpers.rb:428:in 'Spec::Helpers#without_env_side_effects'
# ./spec/support/helpers.rb:409:in 'Spec::Helpers#with_gem_path_as'
# ./spec/spec_helper.rb:163:in 'block (2 levels) in <top (required)>'
# ./lib/rubygems.rb:306:in 'Kernel#load'
# ./lib/rubygems.rb:306:in 'Gem.activate_and_load_bin_path'
# ------------------
# --- Caused by: ---
# Bundler::HTTPError:
# Could not reach host gem.repo2. Check your network connection and try again.
# ./spec/bundler/installer/parallel_installer_spec.rb:151:in 'block (4 levels) in <top (required)>'
...
```
Assisted-by: Claude:claude-opus-4-6[1m]
ruby/rubygems@62d4fb06fc
Check for `step` before the length, although it looks impossible to create 0-step `ArithmeticSequence` currently.
Move `Symbol#empty?` from the C `sym_empty` into `symbol.rb`, implemented as `RBOOL(RSTRING_LEN(rb_sym2str(self)) == 0)`, mirroring the existing Symbol#to_s and #name leaf builtins. Also added an `annotate_builtin! ` in `zjit/src/cruby_methods.rs` to tell ZJIT that the method does not need GC now. `Symbol#empty?` as a leaf builtin, via `benchmark-driver` (`GC.disable`, `loop_count: 20_000_000`, Apple M1). Speedup vs master (CFUNC `sym_empty`): | benchmark | interpreter | `--yjit` | `--zjit` | |---|---:|---:|---:| | `:"".empty?` | **1.24x** | **1.30x** | **1.19x** | | `:abcdefgh.empty?` | **1.19x** | **1.19x** | **1.17x** | Windows native and WSL2 also get faster: | benchmark | interpreter | |---|---:| | `:"".empty?` | **1.47x** | | `:abcdefgh.empty?` | **1.36x** | ### WSL2 (Ubuntu) `x86_64-linux` | benchmark | interpreter | `--yjit` | `--zjit` | |---|---:|---:|---:| | `:"".empty?` | **1.39x** | **1.33x** | **1.32x** | | `:abcdefgh.empty?` | **1.37x** | **1.39x** | **1.33x** |
ruby/psych#805 added files psych_emitter_fy.c and psych_parser_fy.c so we need to update the depend file too.
No need for a separate pass. We can handle it with the other specialization.
Co-authored-by: Takashi Kokubun <takashikkbn@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
8 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See Commits and Changes for more details.
Created by
pull[bot] (v2.0.0-alpha.4)
Can you help keep this open source service alive? 💖 Please sponsor : )