名前¶

encoding::warnings - 暗黙のエンコーディング変換時の警告

VERSION¶

この文書は 2007 年 6 月 5 日にリリースされた encoding::warnings の バージョン 0.11 について記述しています。

概要¶

    use encoding::warnings; # or 'FATAL' to raise fatal exceptions

    utf8::encode($a = chr(20000));  # a byte-string (raw bytes)
    $b = chr(20000);                # a unicode-string (wide characters)

    # "Bytes implicitly upgraded into wide characters as iso-8859-1"
    $c = $a . $b;

説明¶

問題の概観¶

デフォルトでは、Perl の Unicode モデルには本質的な非対称性があります: バイト文字列から Unicode 文字列への暗黙の昇格は ISO 8859-1 (Latin-1) で エンコードされていることを仮定していますが、Unicode 文字列は UTF-8 エンコーディングに降格されます。 これは、Unicode の最初の 256 の符号位置はたまたま Latin-1 と同じで あることにより起こります。

しかし、Unicode 文字列を非 Latin1 データ -- つまり UTF-8 やその他の エンコーディングでエンコードされているバイト文字列 -- と混ぜると、この 暗黙の昇格は簡単に問題を引き起こします。 エラーは結合された文字列が出力に書き込まれるまで明らかにならず、 この時点ではいつ暗黙の昇格が起きたかを知ることは不可能です。

問題の検出¶

このモジュールは、このような問題を診断する手順を単純化します。 単にメインプログラムの先頭に以下の行を書きます:

    use encoding::warnings;

以後、文字位置 128 以上のバイトが暗黙に昇格すると警告が発生します。 例: Bytes implicitly upgraded into wide characters as iso-8859-1 at - line 7.

しかし、純粋に ASCII 符号位置 (0x00..0x7F) からだけからなる文字列は この警告を 引き起こしません。

また、このモジュールを以下のようにしてインポートすることによって警告を 致命的にすることもできます:

    use encoding::warnings 'FATAL';

問題の解決¶

Most of the time, this warning occurs when a byte-string is concatenated with a unicode-string. There are a number of ways to solve it: (TBT)

• Upgrade both sides to unicode-strings

  If your program does not need compatibility for Perl 5.6 and earlier, the recommended approach is to apply appropriate IO disciplines, so all data in your program become unicode-strings. See encoding, open and "binmode" in perlfunc for how.

  If your program does not need compatibility for Perl 5.6 and earlier, the recommended approach is to apply appropriate IO disciplines, so all data in your program become unicode-strings. See encoding, open and "binmode" in perlfunc for how. (TBT)

• Downgrade both sides to byte-strings

  The other way works too, especially if you are sure that all your data are under the same encoding, or if compatibility with older versions of Perl is desired.

  The other way works too, especially if you are sure that all your data are under the same encoding, or if compatibility with older versions of Perl is desired. (TBT)

  You may downgrade strings with Encode::encode and utf8::encode. See Encode and utf8 for details.

  You may downgrade strings with Encode::encode and utf8::encode. See Encode and utf8 for details. (TBT)

• Specify the encoding for implicit byte-string upgrading

  If you are confident that all byte-strings will be in a specific encoding like UTF-8, and need not support older versions of Perl, use the encoding pragma:

  If you are confident that all byte-strings will be in a specific encoding like UTF-8, and need not support older versions of Perl, use the encoding pragma: (TBT)

      use encoding 'utf8';

  Similarly, this will silence warnings from this module, and preserve the default behaviour:

  Similarly, this will silence warnings from this module, and preserve the default behaviour: (TBT)

      use encoding 'iso-8859-1';

  However, note that use encoding actually had three distinct effects:

  However, note that use encoding actually had three distinct effects: (TBT)

  • PerlIO layers for STDIN and STDOUT

    This is similar to what open pragma does.

    This is similar to what open pragma does. (TBT)

  • Literal conversions

    This turns all literal string in your program into unicode-strings (equivalent to a use utf8), by decoding them using the specified encoding.

    This turns all literal string in your program into unicode-strings (equivalent to a use utf8), by decoding them using the specified encoding. (TBT)

  • Implicit upgrading for byte-strings

    This will silence warnings from this module, as shown above.

    This will silence warnings from this module, as shown above. (TBT)

  Because literal conversions also work on empty strings, it may surprise some people:

  Because literal conversions also work on empty strings, it may surprise some people: (TBT)

      use encoding 'big5';
  
      my $byte_string = pack("C*", 0xA4, 0x40);
      print length $a;    # 2 here.
      $a .= "";           # concatenating with a unicode string...
      print length $a;    # 1 here!

  In other words, do not use encoding unless you are certain that the program will not deal with any raw, 8-bit binary data at all.

  In other words, do not use encoding unless you are certain that the program will not deal with any raw, 8-bit binary data at all. (TBT)

  However, the Filter => 1 flavor of use encoding will not affect implicit upgrading for byte-strings, and is thus incapable of silencing warnings from this module. See encoding for more details.

  However, the Filter => 1 flavor of use encoding will not affect implicit upgrading for byte-strings, and is thus incapable of silencing warnings from this module. See encoding for more details. (TBT)

警告¶

Perl 5.9.4 以降では、このモジュールの効果はレキシカルです。

5.9.4 より前のバージョンの Perl では、このモジュールはレキシカルブロックの 内側ではなく、スクリプト全体に影響を与えます。

SEE ALSO¶

perlunicode, perluniintro

open, utf8, encoding, Encode

作者¶

Audrey Tang

コピーライト¶

Copyright 2004, 2005, 2006, 2007 by Audrey Tang <cpan@audreyt.org>.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

See http://www.perl.com/perl/misc/Artistic.html