ruby-dev

In article <200802291457.m1TEv6nh008515@ci.ruby-lang.org>,
  matz@ruby-lang.org writes:

> matz	2008-02-29 23:57:04 +0900 (Fri, 29 Feb 2008)
>
>   New Revision: 15644
>
>   Modified files:
>     trunk/ChangeLog
>     trunk/string.c
>     trunk/test/ruby/test_m17n_comb.rb
>
>   Log:
>     * test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_chomp): test
>       updated.
>
>   http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/test/ruby/test_m17n_comb.rb?r1=15644&r2=15643&diff_format=u
>   http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/string.c?r1=15644&r2=15643&diff_format=u
>   http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/ChangeLog?r1=15644&r2=15643&diff_format=u

この変更で、test_m17n_comb.rb の test_str_sub と
test_str_sub! に assert(false, "test broken") というコードが
入れられて、test broken とされましたが、これはどのように壊れ
ているのでしょうか?

この assert を取り除くと
"\x80".force_encoding("ASCII-8BIT").sub(Regexp.new("\x80".force_encoding("ASCII-8BIT")),"\xC2\x80".force_encoding("UTF-8"))
と
t="\x80".force_encoding("ASCII-8BIT").dup;t.sub!(Regexp.new("\x80".force_encoding("ASCII-8BIT")),"\xC2\x80".force_encoding("UTF-8")).
で例外が発生する失敗が表れますが、これらは引数も結果も単一の
エンコーディングで表現できる文字列しか表れないので、成功して
もおかしくないはずです。

実際、sub の第二引数のかわりにブロックを用いて同じことを行う
と、そっちは成功します。

"\x80".force_encoding("ASCII-8BIT").sub(Regexp.new("\x80".force_encoding("ASCII-8BIT"))) { "\xC2\x80".force_encoding("UTF-8") }
t="\x80".force_encoding("ASCII-8BIT").dup;t.sub!(Regexp.new("\x80".force_encoding("ASCII-8BIT"))) { "\xC2\x80".force_encoding("UTF-8") }

また、このテストはまつもとさんが r15640 の変更を行うまでは成
功していました。

http://www.rubyist.net/~akr/chkbuild/debian-sarge/ruby-trunk/log/20080229T201901.diff.txt.gz

失敗するのが正しいとするなら、理由はなんですか?
その理由はブロック形式には適用されませんか?
-- 
[田中 哲][たなか あきら][Tanaka Akira]

Thread

Prev Next

In This Thread

Prev Next

[#33946] Symbol#inspect should use rb_str_inspect — sheepman <sheepman@...>

[#33947] Ruby 1.9.0-1 snapshot released — Yukihiro Matsumoto <matz@...>

[#33948] Schedule for the 1.8.7 release — "Akinori MUSHA" <knu@...>

[#33950] Ruby 1.9.0-1 build failure: error: `r15660' undeclared — "Hiroshi Ichikawa" <gimite@...>

[#33952] Float('1__1') — Tadayoshi Funaba <tadf@...>

[#33954] Mac OS X 10.5.2でruby-1.9.0-1のmake testが失敗 — 高尾 宏治 <kouji@...>

[#33955] --encoding affects script encoding — sheepman <sheepman@...>

[#33962] Ruby1.9.0でのインタプリタ組み込みについての質問 — Masayuki Yamaguchi <Yamaguchi.Masayuki@...>

[#33966] Re: [ruby-cvs:22881] Ruby:r15644 (trunk): * test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_chomp): test — Tanaka Akira <akr@...>

[#33967] $SAFE=1でLoadErrorではなくInsecure operationになる — "Ken Date" <itacchi@...>

[#33972] なぜ transcode.c の transcode_loop で from_utf8 が特別扱い — Martin Duerst <duerst@...>

[#33974] Test::Unit::Collector::Dirがtest_*.rb以外集めてくれない — "Ken Date" <itacchi@...>

[#33983] Re: [ruby-cvs:22913] Re: Ruby:r15674 (trunk): * gc.c (add_heap): sort heaps array in ascending order to use — Yukihiro Matsumoto <matz@...>

[#33989] Hash#compare_by_identity breaks commutativity of Hash#== — Tanaka Akira <akr@...>

[#33991] Maybe IRB bug!! by 09 — Tanaka Akira <akr@...>

[#33992] debug.rb:221:in `debug_command': undefined method `callcc' for #<DEBUGGER__::Context:0xb7a85914> (NoMethodError) — Tanaka Akira <akr@...>

[#33993] Re: [ruby-cvs:22935] Ruby:r15695 (trunk): * string.c (is_utf8_lead_byte, count_utf8_lead_bytes_with_ulong): — Tanaka Akira <akr@...>

[#33994] printf "[%08f]\n", 0.0/0.0 — Tanaka Akira <akr@...>

[#33996] No definition for sym_succ など — Martin Duerst <duerst@...>

[#34000] --program-suffix option of configure.bat(mswin32) — KIMURA Koichi <kimura.koichi@...>

[#34002] sprintf("% e", Inf) — Tanaka Akira <akr@...>

[#34005] Array#slice! may be too slow and allocate memory too much — "Yusuke ENDOH" <mame@...>

[#34006] 11.divmod(3.5) — Tanaka Akira <akr@...>

[#34008] local_variables contains strings — Tanaka Akira <akr@...>

[#34010] RUBY_*定数のencoding — Kazuhiro NISHIYAMA <zn@...>

[#34011] Should --verbose be equal to -v ? — Yugui <yugui@...>

[#34019] DelegateClassのバックトレース — "GO Noguchi" <gonoguti@...>

[#34020] MurmurHash problem — Nobuyoshi Nakada <nobu@...>

[#34021] singleton class extended by a module with module_function — Tanaka Akira <akr@...>

[#34022] patch of lazy sweep gc — authorNari <authornari@...>

[#34030] uint32_t — KIMURA Koichi <kimura.koichi@...>

[#34033] ruby coding guideline — "NARUSE, Yui" <naruse@...>

[#34037] Ruby performance gains on SPARC — Yukihiro Matsumoto <matz@...>

[#34042] IOはこれからもEnumerableなのか — rubikitch@...

[#34048] CGI::Cookie.new を高速化するバッチ — "Makoto Kuwata" <kwa@...>

[#34052] IO should provide #each_char and #chars — Yugui <yugui@...>

[#34059] putsでSEGV — rubikitch@...

[#34066] Bignum#div — Tadayoshi Funaba <tadf@...>

[#34067] Array#take,take_while,drop,drop_whlie — "Yusuke ENDOH" <mame@...>

[#34068] lgamma_r requires _REENTRANT on Solaris — "Yusuke ENDOH" <mame@...>

[#34071] procを組み合わせた再帰呼び出しで例外が投げられない — wanabe <s.wanabe@...>

[#34072] rb_memsearch optimization — "NARUSE, Yui" <naruse@...>

[#34077] 異なるエンコーディングだと同じバイト列でも==にならない件 — rubikitch@...

[#34086] extend spawn to change attributes of child process. — Tanaka Akira <akr@...>

[#34093] 拡張ライブラリ初期化中でのmodule_eval — Kouhei Sutou <kou@...>

[#34094] tempfile.unlink 後に tempfile.close — SATOH Fumiyasu <fumiyas@...>

[#34095] (再送) Cygwin で Resolv.getaddress が失敗する — Kouhei Yanagita <yanagi@...>

[#34100] delegate Kernel#{gets,readline,readlines} to ARGF — Nobuyoshi Nakada <nobu@...>

[#34105] rational.rb, complex.rb and mathn.rb — Tadayoshi Funaba <tadf@...>

[#34109] LP64: date.rb:321:in `convert': integer 86400000000000 too big to convert to `int' (RangeError) — Tanaka Akira <akr@...>

[#34112] Module#define_methodがprivateなわけ — rubikitch@...

[#34114] Is 'Class.new(Class)' valid? — rucila <rucila@...>

[#34116] Improving Fixnum#gcd(Fixnum) — "Akinori MUSHA" <knu@...>

[#34117] backporting features from 1.9 to 1.8 — "Akinori MUSHA" <knu@...>

[#34119] Process.daemon kills other threads — Tanaka Akira <akr@...>

[#34122] --disable=all --enable=gems — Kazuhiro NISHIYAMA <zn@...>

[#34128] enumerator with certain built-in methods dumps core — "Yusuke ENDOH" <mame@...>

[#34136] thread.c (fastthread)の状況 — Daigo Moriwaki <techml@...>

[#34144] [質問２点] C からの定数参照 & thread switching コストの低減 — Hidetoshi NAGAI <nagai@...>

[#34150] sigsetjmp significantly slower on OSX. — Yukihiro Matsumoto <matz@...>

[#34158] Complex組み込み — Masahiro TANAKA <masa16.tanaka@...>

[#34159] ruby-trunk Marshal.dump bug — nagachika <rucila@...>

[#34162] GC heap size less patch — authorNari <authornari@...>

[#34163] Array#shift/unshift の高速化 — wanabe <s.wanabe@...>

[#34180] メソッドにクラスを定義させるとprivateになる？ — rubikitch@...

[#34184] Re: [ruby-cvs:23097] Ruby:r15858 (trunk): * io.c: IO.copy_stream implemented. — "U.Nakamura" <usa@...>

[#34189] Re: [ruby-cvs:23106] Re: Ruby:r15866 (trunk): * numeric.c (num_quo): should convert its operand to Rational. — Tadayoshi Funaba <tadf@...>

[#34191] Object#singleton_class — rubikitch@...

[ruby-dev:33966] Re: [ruby-cvs:22881] Ruby:r15644 (trunk): * test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_chomp): test

Thread

In This Thread

[#33954] Mac OS X 10.5.2でruby-1.9.0-1のmake testが失敗 — 高尾宏治 <kouji@...>

[#34072] rb_memsearch　optimization — "NARUSE, Yui" <naruse@...>