January 2024 - ruby-core - ml.ruby-lang.org

[ruby-core:115985] [Ruby master Feature#20108] Introduction of Happy Eyeballs Version 2 (RFC8305) in Socket.tcp
by shioimm (Misaki Shioi) 26 Feb '24

26 Feb '24

Issue #20108 has been reported by shioimm (Misaki Shioi). ---------------------------------------- Feature #20108: Introduction of Happy Eyeballs Version 2 (RFC8305) in Socket.tcp https://bugs.ruby-lang.org/issues/20108 * Author: shioimm (Misaki Shioi) * Status: Open * Priority: Normal ---------------------------------------- This is an implementation of Happy Eyeballs version 2 (RFC 8305) in Socket.tcp. ### Background Currently, `Socket.tcp` synchronously resolves names and makes connection attempts with `Addrinfo::foreach.` This implementation has the following two problems. 1. In hostname resolution, the program stops until the DNS server responds to all DNS queries. 2. In a connection attempt, while an IP address is trying to connect to the destination host and is taking time, the program stops, and other resolved IP addresses cannot try to connect. ### Proposal "Happy Eyeballs" ([RFC 8305](https://datatracker.ietf.org/doc/html/rfc8305)) is an algorithm to solve this kind of problem. It avoids delays to the user whenever possible and also uses IPv6 preferentially. I implemented it into `Socket.tcp` by using `Addrinfo.getaddrinfo` in each thread spawned per address family to resolve the hostname asynchronously, and using `Socket::connect_nonblock` to try to connect with multiple addrinfo in parallel. See https://github.com/ruby/ruby/pull/9374 ### Outcome This change eliminates a fatal defect in the following cases. #### Case 1. One of the A or AAAA DNS queries does not return ```ruby require 'socket' class Addrinfo class << self # Current Socket.tcp depends on foreach def foreach(nodename, service, family=nil, socktype=nil, protocol=nil, flags=nil, timeout: nil, &block) getaddrinfo(nodename, service, Socket::AF_INET6, socktype, protocol, flags, timeout: timeout) .concat(getaddrinfo(nodename, service, Socket::AF_INET, socktype, protocol, flags, timeout: timeout)) .each(&block) end def getaddrinfo(_, _, family, *_) case family when Socket::AF_INET6 then sleep when Socket::AF_INET then [Addrinfo.tcp("127.0.0.1", 4567)] end end end end Socket.tcp("localhost", 4567) ``` Because the current `Socket.tcp` cannot resolve IPv6 names, the program stops in this case. It cannot start to connect with IPv4 address. Though `Socket.tcp` with HEv2 can promptly start a connection attempt with IPv4 address in this case. #### Case 2. Server does not promptly return ack for syn of either IPv4 / IPv6 address family ```ruby require 'socket' fork do socket = Socket.new(Socket::AF_INET6, :STREAM) socket.setsockopt(:SOCKET, :REUSEADDR, true) socket.bind(Socket.pack_sockaddr_in(4567, '::1')) sleep socket.listen(1) connection, _ = socket.accept connection.close socket.close end fork do socket = Socket.new(Socket::AF_INET, :STREAM) socket.setsockopt(:SOCKET, :REUSEADDR, true) socket.bind(Socket.pack_sockaddr_in(4567, '127.0.0.1')) socket.listen(1) connection, _ = socket.accept connection.close socket.close end Socket.tcp("localhost", 4567) ``` The current `Socket.tcp` tries to connect serially, so when its first name resolves an IPv6 address and initiates a connection to an IPv6 server, this server does not return an ACK, and the program stops. Though `Socket.tcp` with HEv2 starts to connect sequentially and in parallel so a connection can be established promptly at the socket that attempted to connect to the IPv4 server. In exchange, the performance of `Socket.tcp` with HEv2 will be degraded. ``` 100.times { Socket.tcp("www.ruby-lang.org", 80) } # Socket.tcp (Before) 0.123809 # Socket.tcp (After) 0.224684 ``` This is due to the addition of the creation of IO objects, Thread objects, etc., and calls to `IO::select` in the implementation. -- https://bugs.ruby-lang.org/

4 13

[ruby-core:115169] [Ruby master Bug#19973] Duplicate keyword argument names don't always warn
by tenderlovemaking (Aaron Patterson) 25 Feb '24

25 Feb '24

Issue #19973 has been reported by tenderlovemaking (Aaron Patterson). ---------------------------------------- Bug #19973: Duplicate keyword argument names don't always warn https://bugs.ruby-lang.org/issues/19973 * Author: tenderlovemaking (Aaron Patterson) * Status: Open * Priority: Normal * ruby -v: ruby 3.3.0dev (2023-10-24T19:38:50Z cleanup 3525a9bd22) [arm64-darwin23] * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- Calling a method with duplicate keyword arguments doesn't warn when it could: ```ruby def bar a:, b: a + b end # Warning bar(a: 1, b: 3, a: 2) z = { b: 123 } # No warning bar(a: 1, **z, a: 2) ``` The first call to `bar` gives a warning about duplicate keyword args, but the second call doesn't. I think both cases should emit a warning. -- https://bugs.ruby-lang.org/

3 3

[ruby-core:116516] [Ruby master Bug#20229] Empty keyword splat in array not removed in ARGSPUSH case
by jeremyevans0 (Jeremy Evans) 20 Feb '24

20 Feb '24

Issue #20229 has been reported by jeremyevans0 (Jeremy Evans). ---------------------------------------- Bug #20229: Empty keyword splat in array not removed in ARGSPUSH case https://bugs.ruby-lang.org/issues/20229 * Author: jeremyevans0 (Jeremy Evans) * Status: Open * Priority: Normal * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN, 3.3: UNKNOWN ---------------------------------------- For the following code: ```ruby a = [] kw = {} [*a, **kw] ``` Ruby 2.0-3.3 return `[{}]`, when the desired result (since 2.7) is `[]`. I discovered this while working on an optimization for the ARGSCAT case, where `[*a, 1, **kw]` allocates two arrays instead of a single array. My optimization fixes that by adding a `pushtoarraykwsplat` instruction as a replacement for `newarraykwsplat`. I've used `pushtoarraykwsplat` in the ARGSPUSH case to fix this bug. My pull request for the bug fix and optimization: https://github.com/ruby/ruby/pull/9766 -- https://bugs.ruby-lang.org/

1 2

[ruby-core:115678] [Ruby master Bug#20055] Ripper seems to skip some checks like `void value expression` and `duplicated variable name`
by tompng (tomoya ishida) 20 Feb '24

20 Feb '24

Issue #20055 has been reported by tompng (tomoya ishida). ---------------------------------------- Bug #20055: Ripper seems to skip some checks like `void value expression` and `duplicated variable name` https://bugs.ruby-lang.org/issues/20055 * Author: tompng (tomoya ishida) * Status: Open * Priority: Normal * ruby -v: ruby 3.3.0dev (2023-12-08T07:01:01Z master f2c84c4884) [x86_64-linux] * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- These are syntax error ~~~ruby return + return # void value expression 1 in [a, a] # duplicated variable name 1 + (1 => [a, a]) # both ~~~ But Ripper.sexp returns S-expression, not nil ~~~ruby Ripper.sexp 'return + return' # => [:program, [[:return, [:args_add_block, [[:unary, :+@, [:return0]]], false]]]] Ripper.sexp '1 in [a, a]' # => [:program, [[:case, [:@int, "1", [1, 0]], [:in, [:aryptn, nil, [[:var_field, [:@ident, "a", [1, 6]]], [:var_field, [:@ident, "a", [1, 9]]]], nil, nil], nil, nil]]]] Ripper.sexp '1 + (1 => [a, a]) # both' # => [:program, [[:binary, [:@int, "1", [1, 0]], :+, [:paren, [[:case, [:@int, "1", [1, 5]], [:in, [:aryptn, nil, [[:var_field, [:@ident, "a", [1, 11]]], [:var_field, [:@ident, "a", [1, 14]]]], nil, nil], nil, nil]]]]]]] ~~~ -- https://bugs.ruby-lang.org/

2 1

[ruby-core:111965] [Ruby master Bug#19365] Ractors can access non-shareable values through enumerators
by luke-gru (Luke Gruber) 16 Feb '24

16 Feb '24

Issue #19365 has been reported by luke-gru (Luke Gruber). ---------------------------------------- Bug #19365: Ractors can access non-shareable values through enumerators https://bugs.ruby-lang.org/issues/19365 * Author: luke-gru (Luke Gruber) * Status: Open * Priority: Normal * Backport: 2.7: UNKNOWN, 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- I don't think enumerators should be able to be passed to `Ractor.new` ```ruby obj = Object.new # unshareable value p obj Ractor.new([obj].each) {|f| p f.first }.take ``` -- https://bugs.ruby-lang.org/

2 3

[ruby-core:112155] [Ruby master Bug#19395] Process forking within non-main Ractor creates child stuck in busy loop
by luke-gru (Luke Gruber) 16 Feb '24

16 Feb '24

Issue #19395 has been reported by luke-gru (Luke Gruber). ---------------------------------------- Bug #19395: Process forking within non-main Ractor creates child stuck in busy loop https://bugs.ruby-lang.org/issues/19395 * Author: luke-gru (Luke Gruber) * Status: Open * Priority: Normal * Backport: 2.7: UNKNOWN, 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN ---------------------------------------- ```ruby def test_fork_in_ractor r2 = Ractor.new do pid = fork do exit Ractor.count end pid end pid = r2.take puts "Process #{Process.pid} waiting for #{pid}" _pid, status = Process.waitpid2(pid) # stuck forever if status.exitstatus != 1 raise "status is #{status.exitstatus}" end end test_fork_in_ractor() ``` $ top # shows CPU usage is high for child process -- https://bugs.ruby-lang.org/

2 5

[ruby-core:115864] [Ruby master Feature#20080] Implement #begin_and_end method on Range
by stuyam (Stuart Yamartino) 15 Feb '24

15 Feb '24

Issue #20080 has been reported by stuyam (Stuart Yamartino). ---------------------------------------- Feature #20080: Implement #begin_and_end method on Range https://bugs.ruby-lang.org/issues/20080 * Author: stuyam (Stuart Yamartino) * Status: Open * Priority: Normal ---------------------------------------- Followup Reference: #20027 This feature request is to implement a method called `#begin_and_end` on `Range` that returns an array of the first and last value stored in a range: ```ruby (1..300).begin_and_end #=> [1, 300] first, last = (300..1).begin_and_end first #=> 300 last #=> 1 ``` I believe this would be a great addition to Ranges as they are often used to pass around a single object used to hold endpoints, and this allows easier retrieval of those endpoints. This would allow easier deconstruction into start and end values using array deconstruction as well as a simpler way to serialize to a more primitive object such as an array for database storage. This implementation was suggested by @mame in my initial feature suggestion regarding range deconstruction: https://bugs.ruby-lang.org/issues/20027 This implementation would work similar to how `#minmax` works where it returns an array of two numbers, however the difference is that `#minmax` doesn't work with reverse ranges as @Dan0042 pointed out in the link above: ```ruby (1..42).minmax #=> [1, 42] (42..1).minmax #=> [nil, nil] ``` -- https://bugs.ruby-lang.org/

9 16

[ruby-core:116220] [Ruby master Bug#20187] Bundled gems at Ruby 3.3
by hsbt (Hiroshi SHIBATA) 15 Feb '24

15 Feb '24

Issue #20187 has been reported by hsbt (Hiroshi SHIBATA). ---------------------------------------- Bug #20187: Bundled gems at Ruby 3.3 https://bugs.ruby-lang.org/issues/20187 * Author: hsbt (Hiroshi SHIBATA) * Status: Assigned * Priority: Normal * Assignee: hsbt (Hiroshi SHIBATA) * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN, 3.3: UNKNOWN ---------------------------------------- This ticket is for tracking bundled gems status at Ruby 3.4. I added warnings for bundled gems of Ruby 3.3 at https://bugs.ruby-lang.org/issues/19351 and https://bugs.ruby-lang.org/issues/19776. I started to extract(promote) target libraries to bundled gems. These are: * abbrev * base64 * bigdecimal * csv * drb * getoptlong * mutex_m * nkf * observer * resolv-replace * rinda * syslog -- https://bugs.ruby-lang.org/

2 6

[ruby-core:116435] [Ruby master Misc#20210] Invalid source encoding raises ArgumentError, not SyntaxError
by kddnewton (Kevin Newton) 14 Feb '24

14 Feb '24

Issue #20210 has been reported by kddnewton (Kevin Newton). ---------------------------------------- Misc #20210: Invalid source encoding raises ArgumentError, not SyntaxError https://bugs.ruby-lang.org/issues/20210 * Author: kddnewton (Kevin Newton) * Status: Open * Priority: Normal ---------------------------------------- I was hoping we could change the error that is raised when an invalid source encoding is found from an ArgumentError to a SyntaxError. First let me say, if this isn't possible for backward compatibility, I understand. Please do not take this as me not caring about backward compatibility. Right now, if you have the script `# encoding: foo\n"bar"`, it will raise an ArgumentError, not a SyntaxError. If there are other syntax errors in the file, there's no way to concat them together to give feedback to the user. If a user wants to consistently handle the errors coming back from a parse, they currently have to rescue ArgumentError and SyntaxError. Ideally it would all be SyntaxError, so we could handle them consistently and append all errors together. -- https://bugs.ruby-lang.org/

6 7

[ruby-core:115991] [Ruby master Bug#20111] `defined?` returns `expression` for assignment operators combined with fully qualified constants
by andrykonchin (Andrew Konchin) 14 Feb '24

14 Feb '24

Issue #20111 has been reported by andrykonchin (Andrew Konchin). ---------------------------------------- Bug #20111: `defined?` returns `expression` for assignment operators combined with fully qualified constants https://bugs.ruby-lang.org/issues/20111 * Author: andrykonchin (Andrew Konchin) * Status: Open * Priority: Normal * ruby -v: 3.2.2 * Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN, 3.3: UNKNOWN ---------------------------------------- `defined?` doesn't consider &&=, ||= and other ...= operators as `assignment`: ```ruby defined?(A::B += 1) # => "expression" defined?(A::B &&= true) # => "expression" defined?(A::B ||= true) # => "expression" ``` But consider as `assignment` ordinal assignment: ```ruby defined?(A::B = 1) # => "assignment" ``` Wondering if it's expected behaviour. -- https://bugs.ruby-lang.org/

4 4