Re: [Nmh-workers] show's checkmime

[Oliver Kiddle]

On 21 Jan,  wrote:
> 
> a user may have done 'export MM_CHARSET="$LANG";', so we might want to check
> if UTF-8 is found anywhere in the string:

User's really shouldn't set MM_CHARSET like that. That'd result in
outgoing mime messages having something like charset="fr_FR.UTF-8" in
the Content-Type header instead of one of the canonical forms. If we
want to make it easier for user's to set MM_CHARSET, using nl_langinfo
is better. It was a mistake that I used strncasecmp with 5, in the
patch, just using strcasecmp is I think better.

> The only reason it doesn't is because MM_CHARSET was created before 
> nl_langinfo()
> existed.  Also, we may still be advertising the ability to build it in some 
> old
> crufty environments that don't have nl_langinfo() *yet*, so we may need to 
> have
> an #ifdef/#else/#endif and some ./configure magic (or heave said environments
> from last century over the side and be done with it.. ;)

How about we continue to look in MM_CHARSET but check nl_langinfo where
possible if the environment variable isn't set. The simple patch for
that follows. The only difficult part is that nl_langinfo doesn't
necessarily return character sets in canonical MIME form. There's code
to convert it at http://www.cl.cam.ac.uk/~mgk25/ucs/norm_charmap.c
If nobody objects, the easiest is just to include that.

Oliver

Index: configure.in
===================================================================
RCS file: /cvsroot/nmh/nmh/configure.in,v
retrieving revision 1.65
diff -u -r1.65 configure.in
--- configure.in        6 Oct 2003 20:11:54 -0000       1.65
+++ configure.in        24 Jan 2005 14:49:48 -0000
@@ -444,8 +444,8 @@
 AC_HEADER_STAT
 AC_CHECK_HEADERS(string.h memory.h stdlib.h unistd.h errno.h fcntl.h \
                  limits.h crypt.h termcap.h termio.h termios.h locale.h \
-                 netdb.h sys/param.h sys/time.h sys/utsname.h arpa/inet.h \
-                 arpa/ftp.h)
+                 langinfo.h netdb.h sys/param.h sys/time.h sys/utsname.h \
+                 arpa/inet.h arpa/ftp.h)
 
 
 AC_CACHE_CHECK(POSIX termios, nmh_cv_sys_posix_termios,
@@ -499,7 +499,7 @@
 AC_CHECK_LIB(mkstemp,mkstemp)
 AC_CHECK_FUNCS(waitpid wait3 sigaction sigprocmask sigblock sigsetmask \
                sighold sigrelse writev lstat uname tzset killpg mkstemp \
-               sethostent getutent)
+               sethostent getutent nl_langinfo)
 
 dnl solaris screws this up
 AC_CHECK_FUNC(gethostbyname, [AC_DEFINE(HAVE_GETHOSTBYNAME)],
Index: h/prototypes.h
===================================================================
RCS file: /cvsroot/nmh/nmh/h/prototypes.h,v
retrieving revision 1.8
diff -u -r1.8 prototypes.h
--- h/prototypes.h      19 Nov 2004 05:06:16 -0000      1.8
+++ h/prototypes.h      24 Jan 2005 14:49:48 -0000
@@ -83,6 +83,7 @@
 void m_unknown(FILE *);
 int makedir (char *);
 char *nmh_getpass(const char *);
+char *norm_charmap(char *);
 char *new_fs (char *, char *, char *);
 char *path(char *, int);
 int peekc(FILE *ib);
Index: sbr/Makefile.in
===================================================================
RCS file: /cvsroot/nmh/nmh/sbr/Makefile.in,v
retrieving revision 1.15
diff -u -r1.15 Makefile.in
--- sbr/Makefile.in     21 Nov 2002 21:22:24 -0000      1.15
+++ sbr/Makefile.in     24 Jan 2005 14:49:48 -0000
@@ -65,7 +65,8 @@
        fmt_scan.c lock_file.c m_atoi.c m_backup.c                      \
        m_convert.c m_draft.c m_getfld.c m_gmprot.c                     \
        m_maildir.c m_name.c m_scratch.c m_tmpfil.c                     \
-       makedir.c mts.c path.c peekc.c pidwait.c pidstatus.c            \
+       makedir.c mts.c norm_charmap.c                                  \
+       path.c peekc.c pidwait.c pidstatus.c                            \
        print_help.c print_sw.c print_version.c push.c                  \
        putenv.c pwd.c refile.c remdir.c r1bindex.c                     \
        readconfig.c ruserpass.c seq_add.c seq_bits.c                   \
Index: sbr/check_charset.c
===================================================================
RCS file: /cvsroot/nmh/nmh/sbr/check_charset.c,v
retrieving revision 1.3
diff -u -r1.3 check_charset.c
--- sbr/check_charset.c 21 Jan 2005 19:25:44 -0000      1.3
+++ sbr/check_charset.c 24 Jan 2005 14:49:48 -0000
@@ -10,6 +10,25 @@
  */
 
 #include <h/mh.h>
+#ifdef HAVE_LANGINFO_H
+# include <langinfo.h>
+#endif
+
+
+/*
+ * Get the current character set
+ */
+char *
+get_charset ()
+{
+    char *charset = getenv ("MM_CHARSET");
+#if defined(HAVE_NL_LANGINFO) && defined(CODESET)
+    if (!charset)
+       charset = norm_charmap(nl_langinfo (CODESET));
+#endif
+    return charset;
+}
+
 
 /*
  * Check if we can display a given character set natively.
@@ -28,7 +47,7 @@
 
     /* Cache the name of our default character set */
     if (!mm_charset) {
-       if (!(mm_charset = getenv ("MM_CHARSET")))
+       if (!(mm_charset = get_charset ()))
            mm_charset = "US-ASCII";
        mm_len = strlen (mm_charset);
 
@@ -63,7 +82,7 @@
      * Cache the name of the character set to
      * use for 8bit text.
      */
-    if (!mm_charset && !(mm_charset = getenv ("MM_CHARSET")))
+    if (!mm_charset && !(mm_charset = get_charset ()))
            mm_charset = "x-unknown";
 
     return mm_charset;
Index: sbr/norm_charmap.c
===================================================================
RCS file: sbr/norm_charmap.c
diff -N sbr/norm_charmap.c
--- /dev/null   1 Jan 1970 00:00:00 -0000
+++ sbr/norm_charmap.c  24 Jan 2005 14:49:48 -0000
@@ -0,0 +1,112 @@
+/*
+ * 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.
+ *
+ * address@hidden -- 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 <string.h>
+
+#define digit(x) ((x) >= '0' && (x) <= '9')
+
+static char buf[16];
+
+char *
+norm_charmap(char *name)
+{
+  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";
+
+  /* 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;
+    }
+  }
+
+  /* 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;
+    }
+  }
+
+  /* TIS-620 comes in at least the following two forms */
+  if (!strcmp(name, "TIS-620") ||
+      !strcmp(name, "TIS620.2533"))
+    return "ISO-8859-11";
+
+  /* For some, uppercase/lowercase might differ */
+  if (!strcmp(name, "Big5") || !strcmp(name, "BIG5"))
+    return "Big5";
+  if (!strcmp(name, "Big5HKSCS") || !strcmp(name, "BIG5HKSCS"))
+    return "Big5HKSCS";
+
+  /* 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;
+}