May 2026 - ruby-core - ml.ruby-lang.org

[ruby-core:125247] [Ruby Bug#21995] mkmf generates a non-deterministic log file
by gemmaro (Gemma Kosaka) 13 May '26

13 May '26

Issue #21995 has been reported by gemmaro (Gemma Kosaka). ---------------------------------------- Bug #21995: mkmf generates a non-deterministic log file https://bugs.ruby-lang.org/issues/21995 * Author: gemmaro (Gemma Kosaka) * Status: Open * ruby -v: ruby 4.1.0dev (2026-04-12T14:25:33Z master 6593cc52d6) +PRISM [x86_64-linux] * Backport: 3.3: UNKNOWN, 3.4: UNKNOWN, 4.0: UNKNOWN ---------------------------------------- Hello, When compiling Ruby's C extensions with mkmf, it can generate a `mkmf.log` file with non-deterministic content, especially at the line below: ```txt ld: /tmp/rubytest.h8pay9/cc3fbuGm.o: in function `t': ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ``` This can be problematic on some distributions/package-managers, since the `mkmf.log` file is included in the files generated by `gem install`. {{collapse(example flow) Note about an example flow where randomized filenames occur: 1. `extconf.rb` has calls to `have_func` 2. `gem install` is run by the user 3. It invokes mkmf to process the `extconf.rb` file 4. The checks fail and print intermediate randomized filenames, which are recorded in `mkmf.log` file 5. `gem install` includes `mkmf.log` in the packaged files 6. When the gem is built/packaged multiple times, a reproducibility problem occurs }} {{collapse(background) As a background, the previously reported/related issues are: * ##15304 * This mentions temporary filenames in the generated Makefile, and they are not reproducible. * The necessity/value of including `Makefile`, `gem_make.out`, and `mkmf.log` files are questioned. * So the [pull request](https://github.com/ruby/rubygems/pull/2481) was created to omit these files when installing gems. * However, deleting these files might break compatibility, so the pull request was closed. * Note: There is an [entry](https://github.com/ruby/rubygems/blob/3e3addb8d222d23ac46ca3b1959b29… about `mkmf.log` in the RubyGems' changelog: > ## 2.2.2 / 2014-02-05 > > ### Bug fixes: > > * [...] > * The `mkmf.log` is now placed next to `gem_make.out` when building extensions. * ##19329 * This points out that `gem install` installs `mkmf.log` file. * It was [transferred](https://github.com/ruby/rubygems/issues/6259) at RubyGems GitHub repository. * There are some issues related to this topic on the GNU Guix channel * For example, when I was packaging Ruby 4.0 for Guix, this reproducibility problem [occured](https://codeberg.org/guix/guix/pulls/5166). * There should be other workarounds in the Nix community, apparently. }} To wrap up, I think the followings are desirable: * Conventional file locations should be kept as is on the RubyGems side. * Just deleting these files breaks compatibilities. * These might be helpful when users report issues to gem owners with detailed information. * It is helpful if generated file contents are as reproducible as possible. * In some contexts where reproducibility matters, this issue makes packaging Ruby gems a bit harder. * Luckily, workarounds are possible in distribution package recipes such as deleting these files. * -> the component that generates `mkmf.log` is mkmf, so it might be resolvable by modifying mkmf itself. Thank you, gemmaro. -- https://bugs.ruby-lang.org/

4 7

[ruby-core:125417] [Ruby Bug#22057] test-bundler failed
by tikkss (Tsutomu Katsube) 12 May '26

12 May '26

Issue #22057 has been reported by tikkss (Tsutomu Katsube). ---------------------------------------- Bug #22057: test-bundler failed https://bugs.ruby-lang.org/issues/22057 * Author: tikkss (Tsutomu Katsube) * Status: Open * ruby -v: ruby 4.1.0dev (2026-05-03T21:56:31Z master 4b6040ca3e) +PRISM [x86_64-darwin24] * Backport: 3.3: UNKNOWN, 3.4: UNKNOWN, 4.0: UNKNOWN ---------------------------------------- test-bundler failed just one example (./spec/bundler/bundler/cli_common_spec.rb:7). However, test-bundler-parallel is successful. I think this is happening because while test-bundler includes the `--disable-gems` option, it is not specified in the command that launches rspec when running via turbo_tests (Please see comparison of test commands section). Steps to reproduce --- ```console $ git clone git@github.com:ruby/ruby.git $ cd ruby $ ./autogen.sh $ mkdir build && cd build $ mkdir ~/.rubies $ ../configure --prefix="${HOME}/.rubies/ruby-master" $ make $ make test-bundler BUNDLER_SPECS=bundler/cli_common_spec.rb ``` Expected result --- A spec is successful. Actual result --- Failed a spec as follows: ```console $ make test-bundler BUNDLER_SPECS=bundler/cli_common_spec.rb BASERUBY = /Users/zzz/.rbenv/shims/ruby --disable=gems CC = clang -std=gnu23 LD = ld LDSHARED = clang -std=gnu23 -dynamiclib CFLAGS = -fstack-protector-strong -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -fdeclspec -O3 -fno-fast-math -ggdb3 -Wall -Wextra -Wextra-tokens -Wdeprecated-declarations -Wdivision-by-zero -Wdiv-by-zero -Wimplicit-function-declaration -Wimplicit-int -Wpointer-arith -Wwrite-strings -Wold-style-definition -Wmissing-noreturn -Wno-cast-function-type -Wno-constant-logical-operand -Wno-long-long -Wno-missing-field-initializers -Wno-overlength-strings -Wno-parentheses-equality -Wno-self-assign -Wno-tautological-compare -Wno-unused-parameter -Wno-unused-value -Wunused-variable -Wmisleading-indentation -Wshorten-64-to-32 -Wundef -pipe XCFLAGS = -fno-strict-overflow -fvisibility=hidden -fexcess-precision=standard -DRUBY_EXPORT -fPIE -DPRISM_XALLOCATOR -I. -I.ext/include/x86_64-darwin24 -I.ext/include -I../include -I.. -I../prism -I../enc/unicode/17.0.0 -Dmodular_gc_dir= CPPFLAGS = -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -D_DARWIN_UNLIMITED_SELECT -D_REENTRANT DLDFLAGS = -Wl,-undefined,dynamic_lookup -fstack-protector-strong -Wl,-pie -framework CoreFoundation SOLIBS = -lgmp -ldl -lobjc -lpthread LANG = en_US.UTF-8 LC_ALL = LC_CTYPE = MFLAGS = RUSTC = rustc YJIT_RUSTC_ARGS = --crate-name=yjit --crate-type=staticlib --cfg feature="stats_allocator" -g -C lto=thin -C opt-level=3 -C overflow-checks=on --edition=2021 '--out-dir=/tmp/ruby/build/target/release/' '../yjit/src/lib.rs' ZJIT_RUSTC_ARGS = --crate-name=zjit --crate-type=staticlib --cfg feature="stats_allocator" -g -C lto=thin -C opt-level=3 -C overflow-checks=on --edition=2024 '--out-dir=/tmp/ruby/build/target/release/' '../zjit/src/lib.rs' Apple clang version 17.0.0 (clang-1700.6.4.2) Target: x86_64-apple-darwin24.6.0 Thread model: posix InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin generating enc.mk making srcs under enc make[1]: Nothing to be done for `srcs'. generating transdb.h transdb.h unchanged generating makefiles ext/configure-ext.mk ext/configure-ext.mk updated generating makefile exts.mk exts.mk unchanged make[2]: `ruby' is up to date. make[1]: Nothing to be done for `note'. making enc make[1]: Nothing to be done for `enc'. making trans make[1]: Nothing to be done for `trans'. making encs make[1]: Nothing to be done for `encs'. ./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb --extout=.ext -- --disable-gems -C .. -Ilib -r./tool/lib/bundle_env.rb \ spec/bin/bundle install --quiet --gemfile=tool/bundler/dev_gems.rb ./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb --extout=.ext -- --disable-gems \ -r./x86_64-darwin24-fake \ -r../tool/lib/_tmpdir \ -I../spec/bundler -I../spec/lib \ -e "Dir.chdir(ARGV.shift); load('spec/bin/rspec')" \ -s -- -no-report-tmpdir -- ".." \ -r spec_helper -r formatter_overrides spec/bundler/bundler/cli_common_spec.rb Run options: exclude {mri_only: false, man: false, truffleruby_only: true, jruby_only: true, readline: false, permissions: false, no_color_tty: false, ruby_repo: true, git: "2.54.0", rubygems: "4.1.0.dev", realworld: true} F Failures: 1) Bundler::CLI::Common gem_not_found_message should suggest alternate gem names Failure/Error: expect(message).to match("Did you mean 'rails'?") expected "Could not find gem 'ralis'." to match "Did you mean 'rails'?" # ./spec/bundler/bundler/cli_common_spec.rb:11:in 'block (3 levels) in <top (required)>' # ./spec/bundler/spec_helper.rb:164:in 'block (4 levels) in <top (required)>' # ./spec/bundler/spec_helper.rb:164:in 'block (3 levels) in <top (required)>' # ./spec/bundler/support/helpers.rb:414:in 'block in Spec::Helpers#with_gem_path_as' # ./spec/bundler/support/helpers.rb:428:in 'Spec::Helpers#without_env_side_effects' # ./spec/bundler/support/helpers.rb:409:in 'Spec::Helpers#with_gem_path_as' # ./spec/bundler/spec_helper.rb:163:in 'block (2 levels) in <top (required)>' # ./spec/bundler/support/rubygems_ext.rb:132:in 'Kernel#load' # ./spec/bundler/support/rubygems_ext.rb:132:in 'Spec::Rubygems#gem_load_and_activate' # ./spec/bundler/support/rubygems_ext.rb:16:in 'Spec::Rubygems#gem_load' # -e:1:in 'Kernel#load' # -e:1:in '<main>' Finished in 1 minute 9.5 seconds (files took 0.20881 seconds to load) 1 example, 1 failure Failed examples: rspec ./spec/bundler/bundler/cli_common_spec.rb:7 # Bundler::CLI::Common gem_not_found_message should suggest alternate gem names make: *** [yes-test-bundler] Error 1 ``` Comparison of test commands --- test-bundler: ```console ./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb --extout=.ext -- --disable-gems \ -r./x86_64-darwin24-fake \ -r../tool/lib/_tmpdir \ -I../spec/bundler -I../spec/lib \ -e "Dir.chdir(ARGV.shift); load('spec/bin/rspec')" \ -s -- -no-report-tmpdir -- ".." \ -r spec_helper -r formatter_overrides spec/bundler/bundler/cli_common_spec.rb ``` test-bundler-parallel: ```console ./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb --extout=.ext -- --disable-gems \ -r./x86_64-darwin24-fake \ -r../tool/lib/_tmpdir \ -I../spec/bundler \ -e "ruby = ENV['RUBY']" \ -e "ARGV[-1] = File.expand_path(ARGV[-1])" \ -e "ENV['RSPEC_EXECUTABLE'] = ruby + ARGV.shift" \ -e "load ARGV.shift" \ -s -- -no-report-tmpdir -- \ " -C .. -Ispec/bundler -Ispec/lib .bundle/bin/rspec -r spec_helper" \ ../spec/bin/parallel_rspec -r formatter_overrides \ --runtime-log ../tmp/parallel_runtime_rspec.log ../spec/bundler/bundler/cli_common_spec.rb ``` See also --- * https://docs.ruby-lang.org/en/master/contributing/building_ruby_md.html * https://docs.ruby-lang.org/en/master/contributing/testing_ruby_md.html -- https://bugs.ruby-lang.org/

1 1

[ruby-core:125062] [Ruby Misc#21956] DevMeeting-2026-05-13
by mame (Yusuke Endoh) 12 May '26

12 May '26

Issue #21956 has been reported by mame (Yusuke Endoh). ---------------------------------------- Misc #21956: DevMeeting-2026-05-13 https://bugs.ruby-lang.org/issues/21956 * Author: mame (Yusuke Endoh) * Status: Open ---------------------------------------- # The next dev meeting **Date: 2026/05/13 13:00-17:00** (JST) Log: *TBD* - Dev meeting *IS NOT* a decision-making place. All decisions should be done at the bug tracker. - Dev meeting is a place we can ask Matz, nobu, nurse and other developers directly. - Matz is a very busy person. Take this opportunity to ask him. If you can not attend, other attendees can ask instead of you (if attendees can understand your issue). - We will write a record of the discussion in the file or to each ticket in English. - All activities are best-effort (keep in mind that most of us are volunteer developers). - The date, time and place of the meeting are scheduled according to when/where we can reserve Matz's time. - *DO NOT* discuss then on this ticket, please. # Call for agenda items If you have a ticket that you want matz and committers to discuss, please post it into this ticket in the following format: ``` * [Ticket ref] Ticket title (your name) * Comment (A summary of the ticket, why you put this ticket here, what point should be discussed, etc.) ``` Example: ``` * [Feature #14609] `Kernel#p` without args shows the receiver (ko1) * I feel this feature is very useful and some people say :+1: so let discuss this feature. ``` - It is recommended to add a comment by 2026/05/10. We hold a preparatory meeting to create an agenda a few days before the dev-meeting. - The format is strict. We'll use [this script to automatically create an markdown-style agenda](https://gist.github.com/mame/b0390509ce1491b43610b9ebb665eb86). We may ignore a comment that does not follow the format. - Your comment is mandatory. We cannot read all discussion of the ticket in a limited time. We appreciate it if you could write a short summary and update from a previous discussion. -- https://bugs.ruby-lang.org/

6 6

[ruby-core:125470] [Ruby Feature#22066] MemoryView feedbacks: error reporting and lack of formats
by KitaitiMakoto 12 May '26

12 May '26

Issue #22066 has been reported by KitaitiMakoto (真北市). ---------------------------------------- Feature #22066: MemoryView feedbacks: error reporting and lack of formats https://bugs.ruby-lang.org/issues/22066 * Author: KitaitiMakoto (真北市) * Status: Open ---------------------------------------- Hello. I have been doing small-scale development involving MemoryView for the past two or three years, and I would like to report two use cases I encountered. ## No way to communicate the reason for failure `rb_memory_view_get_func_t` returns `false` when there is a problem, but it would be helpful if the type of problem could also be communicated to the caller - for example, whether the requested flags are not supported, or whether there is an issue with the object's state. In Ruby, exceptions are normally used for this purpose. However, as far as I understand, `rb_memory_view_get_func_t` is currently designed as a boolean-returning function, and there is no documented way to return a structured failure reason while still returning `false`. Currently I output the problem with `rb_warn` before returning `false`. Would it be possible to add an output parameter for an exception, for example: ```c typedef bool (* rb_memory_view_get_func_t)(VALUE obj, rb_memory_view_t *view, int flags); | v typedef bool (* rb_memory_view_get_func_t)(VALUE obj, rb_memory_view_t *view, int flags, VALUE *error); ``` or, add a member function to retrieve the error? ## Insufficient format support When I tried to export GStreamer gem data as a MemoryView, I found that some formats were missing. [GStreamer supports formats such as 24-bit little-endian integers](https://gstreamer.freedesktop.org/documentation/audio/audio-forma… (S24LE), but these could not be expressed using MemoryView's format strings. The size alone can be represented by combining existing format characters like `ccc`, but the semantics are different. Furthermore, there are formats such as `S24_32LE`, where the storage width is 32 bits but the valid bit depth is 24 bits. This also cannot be expressed in the current MemoryView specification. I don't use these formats myself, and since GStreamer appears to be defined quite broadly, I suspect others don't use them much either - but I'm reporting the limitation as a fact. Would it be feasible to either add new format strings, or to express item characteristics using a combination of sign, type, size, and endianness (such as S24LE)? --- As I have only been working in this area in a limited and intermittent way, these proposals may be narrow in perspective, but I hope they can be of some reference for future development. Please feel free to ask if anything needs more detail. -- https://bugs.ruby-lang.org/

1 0

[ruby-core:124843] [Ruby Misc#21916] DevMeeting before RubyKaigi 2026
by ko1 (Koichi Sasada) 12 May '26

12 May '26

Issue #21916 has been reported by ko1 (Koichi Sasada). ---------------------------------------- Misc #21916: DevMeeting before RubyKaigi 2026 https://bugs.ruby-lang.org/issues/21916 * Author: ko1 (Koichi Sasada) * Status: Open ---------------------------------------- RubyKaigi 2026 will be at Hakodate, Japan, Apr 22th - 24th. We would like to try to hold a face-to-face dev meeting at Hakodate. (@matz will also participate!) * Date: 2026/04/21 (Tue.) 15:00-17:00 (The day before RubyKaigi) * Location: RubyKaigi venue (detailed will be announced before the meeting) ### How to participate Open to any RubyKaigi attendees who have a commit bit or who have a topic they particularly want to discuss. Please sign up on the doorkeeper page: https://rubyassociation.doorkeeper.jp/events/195298 The meeting will be held in a hackathon event. Please enjoy Ruby hacking with friends if you want before the meeting. Note: Do not forget to register RubyKaigi ticket too (see "[ruby-core:124591] Invitation to RubyKaigi 2026" from Akira). ### Program * 11:00am door open and start the Hackathon event * 3:00pm opening and self introduction * 3:30pm discuss topics * 5:00pm closing and free time (closing time is depends on topics) * 6:00pm door close ### Call for agenda items If you have a ticket that you want matz and committers to discuss, please post it into this ticket in the following format: ``` * [Ticket ref] Ticket title (your name) * Comment (A summary of the ticket, why you put this ticket here, what point should be discussed, etc.) ``` Example: ``` * [Feature #14609] `Kernel#p` without args shows the receiver (ko1) * I feel this feature is very useful and some people say :+1: so let discuss this feature. ``` - It is recommended to add a comment by 2025/04/20. I'll ask Matz which should be discussed on this meeting and reorder them. - The format is strict. We'll use [this script to automatically create an markdown-style agenda](https://gist.github.com/mame/b0390509ce1491b43610b9ebb665eb86). We may ignore a comment that does not follow the format. - Your comment is mandatory. We cannot read all discussion of the ticket in a limited time. We appreciate it if you could write a short summary and update from a previous discussion. -- https://bugs.ruby-lang.org/

6 6

[ruby-core:125465] [Ruby Bug#22002] argument stack underflow (-1)
by k0kubun (Takashi Kokubun) 11 May '26

11 May '26

Issue #22002 has been updated by k0kubun (Takashi Kokubun). Backport changed from 3.3: DONTNEED, 3.4: REQUIRED, 4.0: REQUIRED to 3.3: DONTNEED, 3.4: REQUIRED, 4.0: DONE ruby_4_0 commit:a9e4608b5225d665ec9c9353a8350f5b940c20fe merged revision(s) commit:d077df24a2256d760cc534b242b28822d4ef6376. ---------------------------------------- Bug #22002: argument stack underflow (-1) https://bugs.ruby-lang.org/issues/22002#change-117273 * Author: Anonymous * Status: Closed * Assignee: prism * ruby -v: ruby 4.0.1 (2026-01-13 revision e04267a14b) +PRISM [arm64-darwin25] * Backport: 3.3: DONTNEED, 3.4: REQUIRED, 4.0: DONE ---------------------------------------- The following lines of code: ```ruby sleep(0.1) until defined?(@variable) @variable ``` produce this error: ``` home@My-Air testing % bundle exec ruby test.rb -- raw disasm-------- trace: 1 0000 jump <L000> ( 1) <L004> [sp: 0, unremovable: 1, refcnt: 1] 0002 putnil ( 1) <L005> [sp: 1, unremovable: 0, refcnt: 1] 0003 pop ( 1) 0004 jump <L000> ( 1) <L001> [sp: 0, unremovable: 1, refcnt: 5] 0006 putself ( 1) 0007 putobject 0.1 ( 1) 0009 opt_send_without_block <calldata:sleep, 1> ( 1) 0011 pop ( 1) <L000> [sp: 0, unremovable: 0, refcnt: 2] <L012> [sp: 0, unremovable: 0, refcnt: 0] * 0012 branchunless <L001> ( 1) adjust: [label: 4] 0014 putnil ( 1) <L002> [sp: -1, unremovable: 0, refcnt: 4] 0015 pop ( 1) trace: 1 0016 getinstancevariable :@variable, <ivc:0> ( 3) 0019 leave ( 3) --------------------- test.rb:1: argument stack underflow (-1) test.rb: compile error (SyntaxError) ``` -- https://bugs.ruby-lang.org/

1 0

[ruby-core:125276] [Ruby Bug#22003] .bundle extensions not built when doing out-of-source build
by Kulikjak (Jakub Kulik) 11 May '26

11 May '26

Issue #22003 has been reported by Kulikjak (Jakub Kulik). ---------------------------------------- Bug #22003: .bundle extensions not built when doing out-of-source build https://bugs.ruby-lang.org/issues/22003 * Author: Kulikjak (Jakub Kulik) * Status: Open * ruby -v: ruby 4.0.2 (2026-03-17 revision d3da9fec82) +PRISM [amd64-solaris2.11] * Backport: 3.3: UNKNOWN, 3.4: UNKNOWN, 4.0: UNKNOWN ---------------------------------------- Hi, I encountered a pretty edge case extension build bug. When building Ruby out-of-source and executing the configure script with a full path in a workspace that contains "test" in its name, it fails to build all the extensions in the `.bundle/gems` directory. It can be easily reporoduced with the following: ``` > pwd /builds/testing/components/ruby/ruby-40 > mkdir build > cd build > /builds/testing/components/ruby/ruby-40/ruby-4.0.2/configure ... > /usr/gnu/bin/make -j 20 ... <the stuff in the .bundle directory is not built> ``` When doing it this way, the `srcdir` variable in the Makefile will be an absolute path, which is then used in `ext/configure-ext.mk` target and in `configure-ext.mk.tmpl` `Dir.glob` call, where the following line: ``` % next if extconf.any?{|f| f.include?(".bundle/gems") && f.include?("test") } ``` makes it omit all the extensions in the ".bundle/gems" directory. The problem is not there when I use a relative path (`../ruby-4.0.2/configure`) or when building directly in `ruby-4.0.2` (because the `srcdir` is relative in those cases). -- https://bugs.ruby-lang.org/

2 1

[ruby-core:125279] [Ruby Bug#22004] parse.y doesn't executes loop body with `while true || true` condition
by Earlopain (Earlopain _) 11 May '26

11 May '26

Issue #22004 has been reported by Earlopain (Earlopain _). ---------------------------------------- Bug #22004: parse.y doesn't executes loop body with `while true || true` condition https://bugs.ruby-lang.org/issues/22004 * Author: Earlopain (Earlopain _) * Status: Open * ruby -v: ruby 4.1.0dev (2026-04-15T13:47:25Z master d077df24a2) +PRISM [x86_64-linux] * Backport: 3.3: WONTFIX, 3.4: REQUIRED, 4.0: REQUIRED ---------------------------------------- ```rb while true || true puts 1 end ``` This should never exit but with parse.y, the body is never executed. It seems to happen when one of the conditions is a truthy literal. Something similar happens with `unitl`: ```rb until false && true puts 1 end ``` By moving the literals out to methods that just return them, it behaves like I would expect. Seems to have started happening in Ruby 3.3. -- https://bugs.ruby-lang.org/

3 2

[ruby-core:124918] [Ruby Bug#21940] Ruby::Box: `$_` returns stale value due to gvar_tbl caching
by dak2 (Daichi Kamiyama) 11 May '26

11 May '26

Issue #21940 has been reported by dak2 (Daichi Kamiyama). ---------------------------------------- Bug #21940: Ruby::Box: `$_` returns stale value due to gvar_tbl caching https://bugs.ruby-lang.org/issues/21940 * Author: dak2 (Daichi Kamiyama) * Status: Open * ruby -v: 4.0.1 * Backport: 3.2: UNKNOWN, 3.3: UNKNOWN, 3.4: UNKNOWN, 4.0: UNKNOWN ---------------------------------------- ## Environment - Ruby Version v4.0.1 - OS: macOS 15.6.1 (arm64) - `RUBY_BOX=1` enabled ## Reproduction ```shell echo -e "a\nb" | RUBY_BOX=1 ruby -e 'gets; $_; gets; p $_' ``` ## Expected ``` "b\n" ``` The second gets reads `"b"` and sets it to `$_`. `p $_` should return the latest value. ## Actual ``` "a\n" ``` `$_` returns the stale value from the first gets. ## Root Cause `$_` is a special variable stored in `svar`. Its getter (`rb_lastline_get`) reads the current value from svar each time it is called. However, when Box is enabled, I believe `rb_gvar_get()` caches the value in the Box's `gvar_tbl` on first access. On subsequent reads, it returns the cached value directly without calling the getter. Since gets updates `$_` via `rb_lastline_set()` which bypasses `rb_gvar_set()`, the cache becomes stale. -- https://bugs.ruby-lang.org/

3 4

[ruby-core:125215] [Ruby Bug#21986] RubyVM::AST incorrect location for literals followed by modifier `if`
by kddnewton (Kevin Newton) 11 May '26

11 May '26

Issue #21986 has been reported by kddnewton (Kevin Newton). ---------------------------------------- Bug #21986: RubyVM::AST incorrect location for literals followed by modifier `if` https://bugs.ruby-lang.org/issues/21986 * Author: kddnewton (Kevin Newton) * Status: Open * Backport: 3.2: UNKNOWN, 3.3: UNKNOWN, 3.4: UNKNOWN, 4.0: UNKNOWN ---------------------------------------- ``` irb(main):001> RubyVM::AbstractSyntaxTree.parse("1.0if true") => (SCOPE@1:0-1:10 tbl: [] args: nil body: (IF@1:0-1:10 (TRUE@1:6-1:10) (FLOAT@1:3-1:3 1.0) nil)) irb(main):002> RubyVM::AbstractSyntaxTree.parse("1if true") => (SCOPE@1:0-1:8 tbl: [] args: nil body: (IF@1:0-1:8 (TRUE@1:4-1:8) (INTEGER@1:1-1:1 1) nil)) irb(main):003> ``` -- https://bugs.ruby-lang.org/

2 1