Part 1 of merging sbr/norm_charmap.c into sbr/check_charset.c.

[nmh] / sbr / check_charset.c

diff --git a/sbr/check_charset.c b/sbr/check_charset.c
index 545aab46c8312bf9e88ae2b7e83070aecba899e6..b10cde34a6429a9edc92712c683d04229f1c8a33 100644 (file)
--- a/sbr/check_charset.c
+++ b/sbr/check_charset.c
@@ -1,82 +1,113 @@
-/* check_charset.c -- routines for character sets
+/* norm_charmap.c -- map locale's encoding to MIME charset.

  *
  *

- * This code is Copyright (c) 2002, by the authors of nmh.  See the
- * COPYRIGHT file in the root directory of the nmh distribution for
- * complete copyright information.
+ * The Single Unix Specification function nl_langinfo(CODESET)
+ * returns the name of the encoding used by the currently selected
+ * locale:
+ *
+ *   http://www.opengroup.org/onlinepubs/7908799/xsh/langinfo.h.html
+ *
+ * Unfortunately the encoding names are not yet standardized.
+ * This function knows about the encoding names used on many
+ * different systems and converts them where possible into
+ * the corresponding MIME charset name registered in
+ *
+ *   http://www.iana.org/assignments/character-sets
+ *
+ * Please extend it as needed and suggest improvements to the author.
+ *
+ * Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11
+ * Permission to use, copy, modify, and distribute this software
+ * for any purpose and without fee is hereby granted. The author
+ * disclaims all warranties with regard to this software.
+ *
+ * Latest version:
+ *
+ *   http://www.cl.cam.ac.uk/~mgk25/ucs/norm_charmap.c

  */
 
  */
 

-#include <h/mh.h>
+#include <string.h>

 
 

-#include <langinfo.h>
+#define digit(x) ((x) >= '0' && (x) <= '9')

 
 

-/* Only get_charset() should use norm_charmap(), so hide its
-   declaration here. */
-char *norm_charmap(char *);
+static char buf[16];

 
 

-/*
- * Get the current character set
- */

 char *
 char *

-get_charset(void)
-{
-    return norm_charmap(nl_langinfo (CODESET));
-}
-
-
-/*
- * Check if we can display a given character set natively.
- * We are passed the length of the initial part of the
- * string to check, since we want to allow the name of the
- * character set to be a substring of a larger string.
- */
-
-int
-check_charset (char *str, int len) 
+norm_charmap(char *name)

 {
 {

-    static char *mm_charset = NULL;
-    static char *alt_charset = NULL;
-    static int mm_len;
-    static int alt_len;
+  char *p;
+  
+  if (!name)
+    return name;
+  
+  /* Many need no remapping, but they are listed here so you
+   * can see what output to expect, and modify for your needs
+   * as necessary. */
+  if (!strcmp(name, "UTF-8"))
+    return "UTF-8";
+  if (!strcmp(name, "EUC-JP"))
+    return "EUC-JP";
+  if (!strcmp(name, "EUC-KR"))
+    return "EUC-KR";
+  if (!strcmp(name, "EUC-TW"))
+    return "EUC-TW";
+  if (!strcmp(name, "KOI8-R"))
+    return "KOI8-R";
+  if (!strcmp(name, "KOI8-U"))
+    return "KOI8-U";
+  if (!strcmp(name, "GBK"))
+    return "GBK";
+  if (!strcmp(name, "GB2312"))
+    return "GB2312";
+  if (!strcmp(name, "GB18030"))
+    return "GB18030";
+  if (!strcmp(name, "VSCII"))
+    return "VSCII";
+  
+  /* ASCII comes in many names */
+  if (!strcmp(name, "ASCII") ||
+      !strcmp(name, "US-ASCII") ||
+      !strcmp(name, "ANSI_X3.4-1968") ||
+      !strcmp(name, "646") ||
+      !strcmp(name, "ISO646") ||
+      !strcmp(name, "ISO_646.IRV"))
+    return "US-ASCII";

 
 

-    /* Cache the name of our default character set */
-    if (!mm_charset) {
-       if (!(mm_charset = get_charset ()))
-           mm_charset = "US-ASCII";
-       mm_len = strlen (mm_charset);
-
-       /* US-ASCII is a subset of the ISO-8859-X and UTF-8 character sets */
-       if (!strncasecmp("ISO-8859-", mm_charset, 9) ||
-               !strcasecmp("UTF-8", mm_charset)) {
-           alt_charset = "US-ASCII";
-           alt_len = strlen (alt_charset);
-       }
+  /* ISO 8859 will be converted to "ISO-8859-x" */
+  if ((p = strstr(name, "8859-"))) {
+    memcpy(buf, "ISO-8859-\0\0", 12);
+    p += 5;
+    if (digit(*p)) {
+      buf[9] = *p++;
+      if (digit(*p)) buf[10] = *p++;
+      return buf;

     }
     }

+  }

 
 

-    /* Check if character set is OK */
-    if ((len == mm_len) && !strncasecmp(str, mm_charset, mm_len))
-       return 1;
-    if (alt_charset && (len == alt_len) && !strncasecmp(str, alt_charset, alt_len))
-       return 1;
-
-    return 0;
-}
-
+  /* Windows code pages will be converted to "WINDOWS-12xx" */
+  if ((p = strstr(name, "CP12"))) {
+    memcpy(buf, "WINDOWS-12\0\0", 13);
+    p += 4;
+    if (digit(*p)) {
+      buf[10] = *p++;
+      if (digit(*p)) buf[11] = *p++;
+      return buf;
+    }
+  }

 
 

-/*
- * Return the name of the character set we are
- * using for 8bit text.
- */
-char *
-write_charset_8bit (void)
-{
-    static char *mm_charset = NULL;
+  /* TIS-620 comes in at least the following two forms */
+  if (!strcmp(name, "TIS-620") ||
+      !strcmp(name, "TIS620.2533"))
+    return "ISO-8859-11";

 
 

-    /*
-     * Cache the name of the character set to
-     * use for 8bit text.
-     */
-    if (!mm_charset && !(mm_charset = get_charset ()))
-           mm_charset = "x-unknown";
+  /* For some, uppercase/lowercase might differ */
+  if (!strcmp(name, "Big5") || !strcmp(name, "BIG5"))
+    return "Big5";
+  if (!strcmp(name, "Big5HKSCS") || !strcmp(name, "BIG5HKSCS"))
+    return "Big5HKSCS";

 
 

-    return mm_charset;
+  /* I don't know of any implementation of nl_langinfo(CODESET) out
+   * there that returns anything else (and I'm not even certain all of
+   * the above occur in the wild), but just in case, as a fallback,
+   * return the unmodified name. */
+  return name;

 }
 }