Mar 2017 Meeting

calendar_today March 13, 2017 confirmation_number 56 tickets discussed Source English

DevMeeting-2017-03-13

Date: 2017/03/13 (Wed) Time: 14:00- 19:00 (JST) Place: Money Forward inc. Headquarter Sign-up: https://ruby.connpass.com/event/51783/ Attendees: Yukihiro Matsumoto (matz), Koichi Sasada, Nobuyoshi Nakada, Yui Naruse, Shyouhei Urabe, Akira Tanaka, Hiroshi Shibata, Masaya Tarui, Martin Dürst (duerst) 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.

Agenda

  • 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

  • https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/ReleaseEngineering25

Carry-over from previous meeting(s)

  • [Feature #13110] Byte-based operations for String (shugo)
  • [Bug #13105] String#to_f and String#to_r don’t stop successive underscores (nobu)
  • [Feature #13109] using in refinements is required to be physically placed before the refined method call (shyouhei)
  • [Bug #12705] yielding args to a lambda uses block/proc rather than lambda/method semantics (shyouhei) status?
  • [Feature #13095] [PATCH] io.c (rb_f_syscall): remove deprecation notice (shyouhei)
  • [Feature #13122] Special syntax for Hash#default_proc (shyouhei)
  • [Feature #12650] Use UTF-8 encoding for ENV on Windows (duerst)
  • [Feature #6284] Add composition for procs (shyouhei)
  • [Feature #13137] Hash Shorthand (shyouhei)
  • [Feature #13166] Feature Request: Byte Arrays for Ruby 3 (shyouhei)
  • [Feature #9116] String#rsplit missing (shyouhei)
  • [Feature #4532] [PATCH] add IO#pread and IO#pwrite methods (shyouhei)
  • [Bug #13146] Float::NANs in Hashes are confusing (more than usual). (mrkn)
  • [Bug #13134] Rational() inconsistency (nobu)

From attendees

  • [Bug #13257] Symbol#respond_to?(:singleton_class) should be false (naruse)
  • [Misc #13163] Uncaught exceptions may not be reported when Thread#report_on_exception=true and Thread#abort_on_exception=true (naruse)
  • [Bug #13181] Unexpected line in rescue backtrace (shyouhei) what’s this?
  • this week’s “assign who?” section (shyouhei)
    • [Bug #13271] Clarifications on refinement spec
    • [Bug #13269] test/readline/test_readline.rb and mingw
    • [Bug #13298] mingw SEGV TestEnumerable#test_callcc
    • [Bug #13280] net/ftp: Putbinaryfile (on Windows) requires blocksize equal to file size
    • [Bug #13276] Dir.glob returns empty array when OS has no more file handles (expected exception)
    • [Bug #13284] IA64 ruby 2.4 miniruby segfault
    • [Bug #13286] Segfault when using rb_debug_inspector_open / rb_debug_inspector_frame_binding_get with Fiddle, but not when directly from C extension
    • [Bug #13305] Occasional segfaults after defining methods while running coverage
    • [Bug #13307] Changing scheme from http to https for the URI does not change the port number
  • [Bug #13282] opt_str_freeze does not always dedupe (shyouhei) seems OK to me
  • [Feature #13295] [PATCH] compile.c: apply opt_str_freeze to String#-@ (uminus) (shyohei) seems OK to me as well
  • [Feature #13179] Deep Hash Update Method (shyouhei)
  • [Feature #8263] Support discovering yield state of individual Fibers (shyouhei)
  • [Feature #8576] Add optimized method type for constant value methods (shyouhei)
  • [Bug #13188] Reinitialize Ruby VM. (shyouhei) New C API …?
  • [Feature #2740] Extend const_missing to pass in the nesting (shyouhei)
  • [Feature #13211] Hash#delete taking a splat (shyouhei)
  • [Bug #11567] Segmentation fault CFUNC :gets (shyouhei)
  • [Feature #13172] Method that yields object to block and returns result (shyouhei) new name candicate
  • [Feature #13224] Add FrozenError as a subclass of RuntimeError (shyouhei)
  • [Misc #13230] Better Do … while structure (shyouhei)
  • [Feature #13245] [PATCH] reject inter-thread TLS modification (shyouhei)
  • [Feature #13252] C API for creating strings without copying (shyouhei)
  • [Bug #13249] Access modifiers don’t have an effect inside class methods in Ruby >= 2.3 (shyouhei) design issue or bug?
  • [Misc #13283] Disable `&’ interpreted as argument prefix warning when passing symbol to Enumerable#map (duerst)
  • [Feature #13290] A method to use a hash like in a case construction (duerst)
  • [Feature #13265] TracePoint for basic operation redefinition (shyouhei)
  • [Bug #12834] prepend getting prepended even if it already exists in the ancestors chain (shyouhei)
  • [Bug #7976] TracePoint call is at call point, not call site (shyouhei)
  • [Bug #13257] Symbol#singleton_class should be undef (shyouhei)
  • [Feature #12901] Anonymous functions without scope lookup overhead (shyouhei) status?
  • [Feature #13166] Feature Request: Byte Arrays for Ruby 3 (shyouhei)
  • [Feature #12573] Introduce a straightforward way to discover whether a process is running (shyouhei)
  • [Feature #13263] Add companion integer nth-root method to recent Integer#isqrt (shyouhei)
  • [Feature #12698] Method to delete a substring by regex match (shyouhei)
  • [Feature #13290] A method to use a hash like in a case construction (shyouhei)
  • [Feature #9453] Return symbols of defined methods for attr and friends (shyouhei) look at the last comment
  • [Feature #13302] Provide a (force) –enable-openssl switch for ruby ./configure (or similar) (shyouhei)
  • [Feature #13303] String#any? as !String#empty? (shyouhei)
  • [Feature #13077] [PATCH] introduce String#fstring method (ko1) some discussions on twitter they don’t like -'str' notation.

From non-attendees

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

  • [Bug #13223] What should the behavior of File.join be when File::SEPARATOR is set? Can I apply this patch? (tenderlove)

  • example: [Feature #10917] Add GC.stat[:total_time] when GC profiling enabled (ko1)

Log

Date: 2017/03/13 (Mon)

Time: 14:00- 19:00 (JST)

Place: Money Forward inc. Headquarter

Sign-up: https://ruby.connpass.com/event/47745/

log edit: https://docs.google.com/document/d/1fZPbRMn2zjVAflvLz1IFWs3Kdijnm2Um4uNLammqURE/edit

log: TBD

Next meeting

  • 4/17

  • at CookPad HQ, yiebisu

About 2.5 timeframe

https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/ReleaseEngineering25

  • 合宿

  • hsbt: I asked shugo-san and he got budget for this.

  • next action: we have to send him a propsal.

  • Jul - Spet?

  • Kanto or not?

  • hsbt: I’ll handle.

Current status of Ruby 2.4

  • 2.4.1?

  • naruse: working on it.

  • 安定版保守事業

  • transition progress report

  • Apr. 1st thtough May 31th

  • nobody here is on contract.
  • ko1: should there be a ceremony for transition?

Carry-over from previous meeting(s)

  • [Feature #13110] Byte-based operations for String (shugo)

  • matz: OK for byteoffset.
  • akr: destructive modification using bytesplice may cause string rescan because it can break validity of the string.
  • duerst: is byteindex needed? is it enough to add a method for MatchData?
  • naruse: that is byteoffset here.

  • akr: byteindex may have problem for inter-codepoint index.

  • [Bug #13105] String#to_f and String#to_r don’t stop successive underscores (nobu)

  • nobu: is this intentional?
  • ko1: ignore all underscores?
  • shyouhei: all? at the beginning?
  • ko1: _ generates zero for to_i.
  • akr: to_i once allowed multiple underscores but prohibits now.
  • tal: ruby’s parser does not allow multiple undersocres as literals.

  • matz: these days other languages follow us like java, python.

  • python: PEP-515, no multiple undersocres

  • java: no pre-post undersocres, but multiple inter-underscores allowed.

  • matz: OK, lets make it as we do as literals.

  • [Feature #13109] using in refinements is required to be physically placed before the refined method call (shyouhei)

  • matz: I don’t think it’s a bug. It’s intentional.  But we may allow this as an extension.

  • [Bug #12705] yielding args to a lambda uses block/proc rather than lambda/method semantics (shyouhei) status?

  • ko1: we should fix the interpreter
  • ko1: we introduced ad-hoc change to lambda creation but that turned out to be a mistake, and observed here.

  • ko1: to be fixed in 2.5.

  • [Feature #13095] [PATCH] io.c (rb_f_syscall): remove deprecation notice (shyouhei)

  • shyouhei: do we want to remove system calls?
  • akr: yes.
  • ko1: it is about foreign-function interface
  • akr: syscall could be easier than libffi because syscall has only one entry point.
  • matz: I think it’s safer to avoid using this feature so kind of reject in mood but …
  • akr: current situation is too difficult to use properly, too easy to get used.

  • akr: my suggestion is to move it to ext/syscall.

  • [Feature #13122] Special syntax for Hash#default_proc (shyouhei)

  • tal: not only default proc, but also there are default values.

  • matz: reject

  • [Feature #12650] Use UTF-8 encoding for ENV on Windows (duerst)

  • nobu: ENV is about inter-process, we cannot change alone.
  • shyouhei: seems they need something like “UTF-8 mode”
  • nobu: that should be “-U”
  • shyouhei: does that option change the encoding of ENV?
  • naruse: it might be possible to use default internal.
  • akr: is ENV in windows handled by kernel?
  • nobu: yes
  • akr: then what encoding?
  • nobu: so-called wide character.
  • duerst: when a PC is used by only one person ENV tends to remain single encoding but when people share a PC, ENV is shared across users. (cp437?)

  • nobu: system-wide.

  • [Feature #6284] Add composition for procs (shyouhei)

  • ko1: is it a matter of speed?
  • shyouhei: it seems composition is faster now
  • nobu: notation seems also be the problem.
  • duerst: notation is a point. (f ∘ g)(x) can be either f(g(x)) or g(f(x)).
  • matz: I have no strong opinion on it.

  • matz: I’ll respond.

  • [Feature #13137] Hash Shorthand (shyouhei)

  • matz: reject

  • [Feature #13166] Feature Request: Byte Arrays for Ruby 3 (shyouhei)

  • akr: I’m for this but I don’t think I can persuade matz.
  • matz: exactly.

  • akr: so I proposed setbit/getbit, in order for the OP to create a dedicated class on top of it.

  • [Feature #9116] String#rsplit missing (shyouhei)

  • akr: what’s this?
  • shyouhei: they say python’s
  • naruse: use case?

  • tal: rsplit(…,2) can be generally written using regexp. ohter cases when we cant do this using regexp?

  • [Feature #4532] [PATCH] add IO#pread and IO#pwrite methods (shyouhei)

  • akr: portability concern? This is POSIX but added recently.
  • shyouhei: POSIX issue 5
  • akr: at least 2001 version has this.

  • matz: OK.

  • [Bug #13146] Float::NANs in Hashes are confusing (more than usual). (mrkn)

  • nobu: prohibit inserting NaNs into hashs?

  • naruse: that’s nonsense.

  • [Bug #13134] Rational() inconsistency (nobu)

  • matz: OK.

From attendees

  • [Bug #13304] public function rb_thread_fd_close is removed at r57422 (naruse)

  • naruse: there are usages for this function.
  • akr: why do you want to remove this?
  • nobu: this is not a function to close a fd (bad naming). rb_io_close shall be used instead.
  • ko1: then why not just let it an alias of rb_io_close?
  • nobu: shuld we deprecate?
  • naruse: no. you can’t deprecate without transition path.

  • conclusion: delete DEPRECATE

  • [Feature #12926] -l flag for line end processing should use chomp! instead of chop! (naruse)

  • naruse: should we backport this?

  • matz: I don’t think so.

  • [Bug #13181] Unexpected line in rescue backtrace (shyouhei) what’s this?

  • this week’s “assign whom?” section (shyouhei)

  • [Bug #13271] Clarifications on refinement spec
  • [Bug #13269] test/readline/test_readline.rb and mingw
  • [Bug #13298] mingw SEGV TestEnumerable#test_callcc
  • [Bug #13280] net/ftp: Putbinaryfile (on Windows) requires blocksize equal to file size

  • [Bug #13276] Dir.glob returns empty array when OS has no more file handles (expected exception)

  • nobu: this one predates 1.0. should we backport?

  • matz: I don’t think so.

  • [Bug #13284] IA64 ruby 2.4 miniruby segfault
  • [Bug #13286] Segfault when using rb_debug_inspector_open / rb_debug_inspector_frame_binding_get with Fiddle, but not when directly from C extension
  • [Bug #13305] Occasional segfaults after defining methods while running coverage

  • [Bug #13307] Changing scheme from http to https for the URI does not change the port number

  • [Bug #13282] opt_str_freeze does not always dedupe (shyouhei)

  • shyouhei: this is about power_assert
  • akr: what’s wrong with it?
  • shyouhei: seems tracepoint?

  • ko1: don’t know the exact reason.

  • [Feature #13295] [PATCH] compile.c: apply opt_str_freeze to String#-@ (uminus) (shyohei) seems OK to me as well

  • ko1: this is buggy.  Can’t redefine -@ in this patch.

  • [Feature #13179] Deep Hash Update Method (shyouhei)

  • nobu: #bury has already been rejected, no update since then.

  • [Misc #13283] Disable `&’ interpreted as argument prefix warning when passing symbol to Enumerable#map (duerst)

  • duerst: the warning is older than Symbol#to_proc
  • shyouhei: why not suppress the warning for symbol literals, not for variables
  • nobu: lexer / parser gets complicated that case.
  • nobu: is this a warning shown everywhere? -> only when -w
  • matz: mruby won’t warn this case?

  • nobu: it does.

  • [Feature #13290] A method to use a hash like in a case construction (duerst)

  • matz: I don’t understand the usage.  Let me ask the OP.

  • [Feature #13077] [PATCH] introduce String#fstring method (ko1) some discussions on twitter they don’t like -‘str’ notation.

  • shyouhei: we can add other stylish names later. -@ is “for the time being”.
  • matz: if you have opinions, have a separate issue.

From non-attendees

  • [Bug #13223] What should the behavior of File.join be when File::SEPARATOR is set? Can I apply this patch? (tenderlove)

  • ko1: this is a bug.
  • nobu: this is either we fix the document to allow other separators or not.
  • akr: I’d like to refrain from adding new feature.
  • nobu: File.split does not honor the constant.
  • shyouhei: It seems jruby honors this constant.
  • matz: OK, lets not use File::SEPARATOR, and explicitly state in document that we might ignore reassignments to that constant.