Date: 2017/05/19 (Fri)
Time: 14:00- 19:00 (JST)
Place: Speee Inc., Kurosaki Bldg. 4F, 4-1-4 Roppongi, Minato-ku Tokyo, Japan
log edit:
Attendees: Matz (Yukihiro Matsumoto), muraken (Kenta Murata), Nobuyoshi Nakada, ko1 (Koichi Sasada), Akira Matsuda, Yui Naruse, Hiroshi Shibata, akr (Akira Tanaka), Shota Fukumori, Shyouhei Urabe, Martin Dürst
Language: mostly Japanese (sorry for non native Japanese speakers)

Please add your favorite ticket numbers you want to ask to discuss.

  • NOTE
    • 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 log about discussion to a 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 is scheduled according to when/where we can reserve Matz's time.


  • NOTE: Write at least "ticket number/title/link" and your name (see example below). Explain details on the ticket. If you cannot attend the meeting, we appreciate a short summary because we can understand it more easily (long discussion is difficult to read, especially in a non-native language). Your motivation is also welcome.

About 2.5 timeframe

Carry-over from previous meeting(s)

  • Previous bugs that were not assigned
    • [Bug #13196] Improve keyword argument errors when non-keyword arguments given
    • [Bug #13320] rescue blocks get an entry in backtrace locations
    • [Bug #13324] IRB Segmentation Fault from eval infinite loop
    • [Bug #13330] Array.include? is slow for symbols
    • [Bug #13336] Default Parameters don't work
    • [Bug #13337] Eval and Later Defined Local Variables
    • [Bug #13390] MinGW build test-all SEGV, issue in test framework or error recovery?
    • [Bug #13350] :newline option not respected on Linux
    • [Bug #13351] net/http: Net::HTTP.start sets wrong default arg value
    • watson's
      • [Bug #13341] Improve performance of implicit type conversion
      • [Bug #13342] Improve yielding block performance
      • [Bug #13354] Improve Time#<=> performance
      • [Bug #13357] Improve Time#+ & Time#- performance
      • [Bug #13365] Improve performance of rb_equal() with special constants
      • [Bug #13368] Improve performance of Array#sum with float elements
      • [Bug #13374] Fix one of performance regressions in method calling
    • [Bug #13373] FileUtils methods for copy, move and remove directories is not providing a decent error trace for letting know if it was success or fail
    • [Bug #13101] Date#rfc2822 and Time#rfc2822 don't return the same format
    • [Bug #13312] String#casecmp raises TypeError instead of returning nil
    • [Bug #12684] Delegator#eql? missing
    • [Feature #13389] [PATCH] POP3 support timeout for TLS handshake
    • [Bug #13404] Hash#any? yields arguments to lambdas with proc semantics
    • [Bug #13413] --with-static-linked-ext doesn't install extension files on make install
    • [Bug #13429] Net::SMTP has no read timeout when connexion over TLS
  • [Feature #13309] Locale paramter for Integer(), Float(), Rational() (shyouhei)
  • [Feature #12733] Bundle bundler to ruby core (shyouhei) updates?
  • [Feature #11302] Dir.entries and Dir.foreach without [".", ".."] (shyouhei)
  • [Feature #13332] Forwardable#def_instance_delegator nil (shyouhei)
  • [Feature #13333] block to yield (shyouhei)
  • [Feature #13378] Eliminate 4 of 8 syscalls when requiring file by absolute path (shyouhei)
  • [Feature #13381] [PATCH] Expose rb_fstring and its family to C extensions (shyouhei)
  • [Bug #13315] Single "%" at the end of printf format string appears in the result (shyouhei)
  • [Feature #13385] [PATCH] Make Resolv::DNS::Name validation similar to host and dig commands (shyouhei)
  • [Feature #13383] [PATCH] Module#source_location (shyouhei)
  • [Feature #12063] KeyError#receiver and KeyError#name (shyouhei) status?
  • [Bug #13397] #object_id should not be signed (shyouhei)
  • [Feature #13396] Net::HTTP has no write timeout (shyouhei)
  • [Bug #13407] We have recv_nonblock but not send_nonblock... can we add it? (shyouhei)
  • [Feature #13395] Add a method to check for not nil (shyouhei)

From attendees

  • [Feature #13483] TracePoint#enable with block for thread-local trace (ko1)
  • [Bug #13249] Access modifiers don't have an effect inside class methods in Ruby >= 2.3 (ko1, ask the approach)
  • watson continues to request pulling his ideas
    • [Bug #13436] Improve performance of Array#<=> with Fixnum/Float/String elements
    • [Bug #13437] Improve performance of Enumerable#{sort_by,min_by,max_by,minmax_by}
    • [Bug #13443] Improve performance of Range#{min,max}
    • [Bug #13482] Improve performance of "set instance variable"
    • [Bug #13447] Improve performance of rb_eql()
    • [Bug #13503] Improve performance of some Time & Rational methods
    • [Bug #13506] Improve performance of Complex#{+,-,*,/,**,abs2}
    • [Bug #13519] Improve performance of some Time methods
    • [Bug #13507] Improve performance of some Complex methods where call Numeric#real? internally
    • [Bug #13553] Improve performance in where push the element into non shared Array object
  • assign who? corner (shyouhei)
    • [Bug #13432] set_trace_funcにproc->is_from_method = TRUEのオブジェクトを渡し、SystemStackErrorを発生させるとRubyVMが停止する
    • [Bug #13446] refinements with prepend for module has strange behavior
    • [Bug #13492] Integer#prime? and Prime.each might produce false positives
    • [Bug #13498] Weakref, Weakmap and define_finalizer don't work on frozen objects
    • [Bug #13513] Resolv::DNS::Message.decode hangs after detecting truncation in UDP messages
    • [Bug #13501] Process.kill behaviour for negative pid is not documented and may be wrong
    • [Bug #13515] Pathname#join doesn't add separator on UNC paths
    • [Bug #13521] [PATCH] Add fallback for DNS resolver registry key on Wine
    • [Bug #13537] ruby crash in rb_gc_mark
    • [Bug #13545] Ruby built by clang 4.0.0 parses Float literal wrongly
    • [Bug #13548] miniruby SEGV while building with non-default CFLAGS (caused by __builtin_setjmp)
    • [Bug #13555] Disable TestTrace#test_trace_stackoverflow
    • [Bug #13524] miniruby: [BUG] Segmentation fault at 0x0055e487e00230 ruby 2.4.1p111 (2017-03-22 revision 58053) [x86_64-li
    • [Bug #13557] there's no way to pass backtrace locations as a massaged backtrace
    • [Bug #10290] segfault when calling a lambda recursively after rescuing SystemStackError
      • Fix confirmed?
    • [Feature #10674] Net::HTTP retries idempotent requests once after a timeout, but its not configurable
    • [Bug #13542] MinGW trunk Builds - Summary of Issues
      • [Bug #13549] MinGW / Windows encoding - Two issues
      • [Bug #13556] MinGW readline Alt / Meta keys
      • [Bug #13569] Windows - TestRubyOptions#test_search - append to paths instead of replacing
    • [Bug #13564] Exception message management
  • [Feature #13383] [PATCH] Module#source_location (shyouhei)
  • [Feature #13434] better method definition in C API (shyouhei)
  • [Feature #13512] System Threads (shyouhei)
  • [Feature #13518] Indented multiline comments (shyouhei)
  • [Feature #13516] Improve the text of the circular require warning (shyouhei)
  • [Feature #13532] Enable :encoding key or open-uri (open()) similar as to how and File.readlines() already allow for (shyouhei)
  • [Bug #13535] Installing Ruby2.4.1 on Solaris 10 (shyouhei) response?
  • [Feature #13552] [PATCH 0/2] reimplement ConditionVariable, Queue, SizedQueue using ccan/list (shyouhei)
  • [Feature #13568] File#path for O_TMPFILE fds are unmeaning (sorah)
  • [Feature #13563] Implement Hash#choice method. (shyouhei)
  • [Feature #13562] Use a sized enumerator with #yield_self (shyouhei) needs?
  • [Feature #13056] base option to Dir.glob (nobu)

From non-attendees

Write your name and your interest (what do you want to ask and to whom?) please.

  • [Feature #8661] Backstrace in reverse order. The OP asked for an option but this is currently always the case for the toplevel exception handler. Is this OK for compatibility? It is surprising for the least (See comment 6). Can we have matz's opinion? (eregon)