diff options
| -rw-r--r-- | ChangeLog | 4 | ||||
| -rw-r--r-- | enc/trans/iso2022.trans | 19 | ||||
| -rw-r--r-- | test/ruby/test_transcode.rb | 18 |
3 files changed, 34 insertions, 7 deletions
@@ -1,3 +1,7 @@ +Thu Apr 1 17:17:00 2010 NARUSE, Yui <naruse@ruby-lang.org> + + * enc/trans/iso2022.trans: CP50221 supports 8bit JIS. + Thu Apr 1 16:44:00 2010 Eric Hodel <drbrain@segment7.net> * lib/rdoc, test/rdoc: Imported RDoc 2.5 diff --git a/enc/trans/iso2022.trans b/enc/trans/iso2022.trans index a93f211f62..0acb7b22d3 100644 --- a/enc/trans/iso2022.trans +++ b/enc/trans/iso2022.trans @@ -46,7 +46,7 @@ "1b2442" => :func_so, # designate JIS X 0208 1983 to G0. "ESC $ B" "0e" => :func_so, # designate JIS X 0201 katakana to G0. "SO" "0f" => :func_so, # designate US-ASCII to G0. "SI" - "{00-0d,10-1a,1c-92}" => :func_si, + "{00-0d,10-1a,1c-92,a1-df}" => :func_si, } transcode_generate_node(ActionMap.parse(map), "cp50221_decoder") @@ -269,22 +269,26 @@ static VALUE fun_si_cp50221_decoder(void *statep, const unsigned char *s, size_t l) { unsigned char *sp = statep; + int c; switch (*sp) { case G0_ASCII: + if (0xA1 <= s[0] && s[0] <= 0xDF) + return (VALUE)FUNso; return (VALUE)NOMAP; case G0_JISX0201_KATAKANA: - if (0x21 <= s[0] && s[0] <= 0x5f) + c = s[0] & 0x7F; + if (0x21 <= c && c <= 0x5f) return (VALUE)FUNso; break; case G0_JISX0208_1978: - if (0x21 <= s[0] && s[0] <= 0x28 || 0x30 <= s[0] && s[0] <= 0x74) + if ((0x21 <= s[0] && s[0] <= 0x28) || (0x30 <= s[0] && s[0] <= 0x74)) return (VALUE)iso2022jp_decoder_jisx0208_rest; break; case G0_JISX0208_1983: - if (0x21 <= s[0] && s[0] <= 0x28 || + if ((0x21 <= s[0] && s[0] <= 0x28) || s[0] == 0x2D || - 0x30 <= s[0] && s[0] <= 0x74 || - 0x79 <= s[0] && s[0] <= 0x7C) + (0x30 <= s[0] && s[0] <= 0x74) || + (0x79 <= s[0] && s[0] <= 0x7C)) /* 0x7F <= s[0] && s[0] <= 0x92) */ return (VALUE)iso2022jp_decoder_jisx0208_rest; break; @@ -327,7 +331,8 @@ fun_so_cp50221_decoder(void *statep, const unsigned char *s, size_t l, unsigned *sp = G0_ASCII; return 0; default: - if (*sp == G0_JISX0201_KATAKANA) { + if (*sp == G0_JISX0201_KATAKANA || + (0xA1 <= s[0] && s[0] <= 0xDF && *sp == G0_ASCII)) { o[0] = 0x8E; o[1] = s[0] | 0x80; } diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb index bca51ba251..7f73d31797 100644 --- a/test/ruby/test_transcode.rb +++ b/test/ruby/test_transcode.rb @@ -1357,6 +1357,24 @@ class TestTranscode < Test::Unit::TestCase "\xA1\xA1".encode("ISO-2022-JP", "EUC-JP")) end + def test_cp50221 + assert_equal("!", "\e(B\x21".encode("utf-8", "cp50221")) + assert_equal("!", "\e(J\x21".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\xB1".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\e(B\xB1".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\e(J\xB1".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\e(I\xB1".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\e(I\x31".encode("utf-8", "cp50221")) + assert_equal("\uFF71", "\x0E\xB1".encode("utf-8", "cp50221")) + assert_equal("\u3000", "\e$@\x21\x21".encode("utf-8", "cp50221")) + assert_equal("\u3000", "\e$B\x21\x21".encode("utf-8", "cp50221")) + assert_equal("\u2460", "\e$B\x2D\x21".encode("utf-8", "cp50221")) + assert_equal("\u7e8a", "\e$B\x79\x21".encode("utf-8", "cp50221")) + assert_equal("\u5fde", "\e$B\x7A\x21".encode("utf-8", "cp50221")) + assert_equal("\u72be", "\e$B\x7B\x21".encode("utf-8", "cp50221")) + assert_equal("\u91d7", "\e$B\x7C\x21".encode("utf-8", "cp50221")) + end + def test_iso_2022_jp_1 # check_both_ways("\u9299", "\x1b$(Dd!\x1b(B", "iso-2022-jp-1") # JIS X 0212 区68 点01 銙 end |