JIS X 0201's MIME Encode default should be ISO-2022-JP B encoding. [nkf-bug:20079]

[nkf/nkf.git] / nkf.c

diff --git a/nkf.c b/nkf.c
index aab1bcd..979955c 100644 (file)
--- a/nkf.c
+++ b/nkf.c
@@ -20,11 +20,11 @@
  *
  * 3. This notice may not be removed or altered from any source distribution.
  */
-#define NKF_VERSION "2.0.9"
-#define NKF_RELEASE_DATE "2009-06-23"
+#define NKF_VERSION "2.1.1"
+#define NKF_RELEASE_DATE "2010-01-05"
 #define COPY_RIGHT \
     "Copyright (C) 1987, FUJITSU LTD. (I.Ichikawa).\n" \
-    "Copyright (C) 1996-2009, The nkf Project."
+    "Copyright (C) 1996-2010, The nkf Project."
 
 #include "config.h"
 #include "nkf.h"
@@ -210,6 +210,8 @@ struct {
 } encoding_name_to_id_table[] = {
     {"US-ASCII",               ASCII},
     {"ASCII",                  ASCII},
+    {"646",                    ASCII},
+    {"ROMAN8",                 ASCII},
     {"ISO-2022-JP",            ISO_2022_JP},
     {"ISO2022JP-CP932",                CP50220},
     {"CP50220",                        CP50220},
@@ -221,6 +223,7 @@ struct {
     {"ISO-2022-JP-2004",       ISO_2022_JP_2004},
     {"SHIFT_JIS",              SHIFT_JIS},
     {"SJIS",                   SHIFT_JIS},
+    {"PCK",                    SHIFT_JIS},
     {"WINDOWS-31J",            WINDOWS_31J},
     {"CSWINDOWS31J",           WINDOWS_31J},
     {"CP932",                  WINDOWS_31J},
@@ -1028,7 +1031,7 @@ nkf_each_char_to_hex(void (*f)(nkf_char c2,nkf_char c1), nkf_char c)
     int shift = 20;
     c &= VALUE_MASK;
     while(shift >= 0){
-       if(c >= 1<<shift){
+       if(c >= NKF_INT32_C(1)<<shift){
            while(shift >= 0){
                (*f)(0, bin2hex(c>>shift));
                shift -= 4;
@@ -2230,13 +2233,15 @@ nkf_iconv_utf_16(nkf_char c1, nkf_char c2, nkf_char c3, nkf_char c4)
 static nkf_char
 w_iconv16(nkf_char c2, nkf_char c1, nkf_char c0)
 {
-    return 0;
+    (*oconv)(c2, c1);
+    return 16; /* different from w_iconv32 */
 }
 
 static nkf_char
 w_iconv32(nkf_char c2, nkf_char c1, nkf_char c0)
 {
-    return 0;
+    (*oconv)(c2, c1);
+    return 32; /* different from w_iconv16 */
 }
 
 static size_t
@@ -3062,11 +3067,11 @@ push_hold_buf(nkf_char c2)
 }
 
 static int
-h_conv(FILE *f, int c1, int c2)
+h_conv(FILE *f, nkf_char c1, nkf_char c2)
 {
-    int ret, c4, c3;
+    int ret;
     int hold_index;
-
+    nkf_char c3, c4;
 
     /** it must NOT be in the kanji shifte sequence      */
     /** it must NOT be written in JIS7                   */
@@ -3816,6 +3821,7 @@ static const unsigned char *mime_pattern[] = {
     (const unsigned char *)"\075?ISO-8859-1?Q?",
     (const unsigned char *)"\075?ISO-8859-1?B?",
     (const unsigned char *)"\075?ISO-2022-JP?B?",
+    (const unsigned char *)"\075?ISO-2022-JP?B?",
     (const unsigned char *)"\075?ISO-2022-JP?Q?",
 #if defined(UTF8_INPUT_ENABLE)
     (const unsigned char *)"\075?UTF-8?B?",
@@ -3836,7 +3842,7 @@ nkf_char (*mime_priority_func[])(nkf_char c2, nkf_char c1, nkf_char c0) = {
 };
 
 static const nkf_char mime_encode[] = {
-    EUC_JP, SHIFT_JIS, ISO_8859_1, ISO_8859_1, JIS_X_0208, JIS_X_0201_1976_K,
+    EUC_JP, SHIFT_JIS, ISO_8859_1, ISO_8859_1, JIS_X_0208, JIS_X_0201_1976_K, JIS_X_0201_1976_K,
 #if defined(UTF8_INPUT_ENABLE)
     UTF_8, UTF_8,
 #endif
@@ -3845,7 +3851,7 @@ static const nkf_char mime_encode[] = {
 };
 
 static const nkf_char mime_encode_method[] = {
-    'B', 'B','Q', 'B', 'B', 'Q',
+    'B', 'B','Q', 'B', 'B', 'B', 'Q',
 #if defined(UTF8_INPUT_ENABLE)
     'B', 'Q',
 #endif
@@ -4439,7 +4445,7 @@ mime_getc(FILE *f)
        }
        if (c1=='='&&c2<SP) { /* this is soft wrap */
            while((c1 =  (*i_mgetc)(f)) <=SP) {
-               if ((c1 = (*i_mgetc)(f)) == EOF) return (EOF);
+               if (c1 == EOF) return (EOF);
            }
            mime_decode_mode = 'Q'; /* still in MIME */
            goto restart_mime_q;
@@ -4649,7 +4655,8 @@ mime_prechar(nkf_char c2, nkf_char c1)
                base64_count = 1;
            }
        } else {
-           if (base64_count + mimeout_state.count/3*4> 66) {
+           if (!(c2 == 0 && (c1 == CR || c1 == LF)) &&
+                   base64_count + mimeout_state.count/3*4> 66) {
                (*o_base64conv)(EOF,0);
                OCONV_NEWLINE((*o_base64conv));
                (*o_base64conv)(0,SP);
@@ -4849,14 +4856,17 @@ mime_putc(nkf_char c)
                    return;
                }
            }
-           (*o_mputc)(c);
-           base64_count++;
+           if (c != 0x1B) {
+               (*o_mputc)(c);
+               base64_count++;
+               return;
+           }
        }
-       return;
     }
 
     if (mimeout_mode <= 0) {
-       if (c <= DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) {
+       if (c <= DEL && (output_mode==ASCII || output_mode == ISO_8859_1 ||
+                   output_mode == UTF_8)) {
            if (nkf_isspace(c)) {
                int flag = 0;
                if (mimeout_mode == -1) {
@@ -4947,14 +4957,15 @@ mime_putc(nkf_char c)
        }
     }else{
        /* mimeout_mode == 'B', 1, 2 */
-       if ( c<=DEL && (output_mode==ASCII ||output_mode == ISO_8859_1)) {
+       if (c <= DEL && (output_mode==ASCII || output_mode == ISO_8859_1 ||
+                   output_mode == UTF_8)) {
            if (lastchar == CR || lastchar == LF){
                if (nkf_isblank(c)) {
                    for (i=0;i<mimeout_state.count;i++) {
                        mimeout_addchar(mimeout_state.buf[i]);
                    }
                    mimeout_state.count = 0;
-               } else if (SP<c && c<DEL) {
+               } else {
                    eof_mime();
                    for (i=0;i<mimeout_state.count;i++) {
                        (*o_mputc)(mimeout_state.buf[i]);
@@ -5250,6 +5261,8 @@ module_connection(void)
     set_output_encoding(output_encoding);
     oconv = nkf_enc_to_oconv(output_encoding);
     o_putc = std_putc;
+    if (nkf_enc_unicode_p(output_encoding))
+       output_mode = UTF_8;
 
     /* replace continucation module, from output side */
 
@@ -5398,7 +5411,7 @@ kanji_convert(FILE *f)
               (c4 = (*i_getc)(f)) != EOF) {
            nkf_iconv_utf_32(c1, c2, c3, c4);
        }
-       (*i_ungetc)(EOF, f);
+       goto finished;
     }
     else if (iconv == w_iconv16) {
        while ((c1 = (*i_getc)(f)) != EOF &&
@@ -5409,7 +5422,7 @@ kanji_convert(FILE *f)
                nkf_iconv_utf_16(c1, c2, c3, c4);
            }
        }
-       (*i_ungetc)(EOF, f);
+       goto finished;
     }
 #endif
 
@@ -5791,6 +5804,7 @@ kanji_convert(FILE *f)
        /* goto next_word */
     }
 
+finished:
     /* epilogue */
     (*iconv)(EOF, 0, 0);
     if (!input_codename)