[patch] man-1.6b-i18n_nroff.patch

man-1.6b-i18n_nroff.patch (text/plain), 7.38 KB, created by SpanKY on 2005-11-10 11:55:44 UTC

--- man-1.6b/configure.nroff	2005-11-10 10:02:58.000000000 +0100
+++ man-1.6b/configure	2005-11-10 10:29:55.971053960 +0100
@@ -405,10 +405,8 @@
   done
   troff=""
   nroff=""
-  jnroff=""
   eqn=""
   neqn=""
-  jneqn=""
   tbl=""
   col=""
   vgrind=""
@@ -467,25 +465,23 @@
   then
     if test $Fnroff = "missing"
     then
-      nroff="nroff -Tlatin1 -mandoc"
+      nroff="nroff --legacy NROFF_OLD_CHARSET -mandoc 2>/dev/null"
     else
-      nroff="$Fnroff -Tlatin1 -mandoc"
+      nroff="$Fnroff --legacy NROFF_OLD_CHARSET -mandoc 2>/dev/null"
     fi
     troff="troff -mandoc"
     echo "Warning: could not find groff"
   else
     if test $Fnroff = "missing"
     then
-      nroff="$Fgroff -Tlatin1 -mandoc"
+      nroff="$Fgroff -Tutf8 -mandoc"
     else
-      nroff="$Fnroff -Tlatin1 -mandoc"
+      nroff="$Fnroff --legacy NROFF_OLD_CHARSET -mandoc 2>/dev/null"
     fi
     troff="$Fgroff -Tps -mandoc"
-    jnroff="$Fgroff -Tnippon -mandocj"
   fi
   eqn="$Fgeqn -Tps"
-  neqn="$Fgeqn -Tlatin1"
-  jneqn="$Fgeqn -Tnippon"
+  neqn="$Fgeqn -Tutf8"
   tbl="$Fgtbl"
   col="$Fcol"
   vgrind="$Fvgrind"
@@ -518,7 +514,7 @@
   cat="$Fcat"
   awk="$Fawk"
 
-  FILTERS="troff nroff jnroff eqn neqn jneqn tbl col vgrind refer grap pic pager browser htmlpager cmp cat awk"
+  FILTERS="troff nroff eqn neqn tbl col vgrind refer grap pic pager browser htmlpager cmp cat awk"
 fi
 
 # Note: older nroff gives an error message for -c
@@ -526,7 +522,6 @@
 if [ "x$set_compatibility_mode_for_colored_groff" = "xtrue" ]; then
     troff="$troff -c"
     nroff="$nroff -c"
-    jnroff="$jnroff -c"
 fi
 
 if [ x$default = x ]; then
@@ -1281,10 +1276,8 @@
 s,@LIBOBJS@,$LIBOBJS,
 s,@troff@,$troff,
 s,@nroff@,$nroff,
-s,@jnroff@,$jnroff,
 s,@eqn@,$eqn,
 s,@neqn@,$neqn,
-s,@jneqn@,$jneqn,
 s,@tbl@,$tbl,
 s,@nocol@,$nocol,
 s,@pcol@,$pcol,
--- man-1.6b/src/man.c.nroff	2005-08-21 01:26:06.000000000 +0200
+++ man-1.6b/src/man.c	2005-11-10 10:53:04.778923072 +0100
@@ -80,8 +80,7 @@
 int fsstnd;
 int noautopath;
 int nocache;
-static int is_japanese;
-static char *language;
+
 static char **section_list;
 
 #ifdef DO_COMPRESS
@@ -442,7 +441,7 @@
 }
 
 static int
-is_lang_page (char *lang, const char *file) {
+is_lang_page (const char *lang, const char *file) {
 	char lang_path[16] = "";
 
 	snprintf(lang_path, sizeof(lang_path), "/%s/", lang);
@@ -457,22 +456,75 @@
 	return 0;
 }
 
+/* we need to pass, as a parameter, what the character set of a man page
+ * is likely to be if it is not utf-8, so that nroff can iconv it to utf-8.
+ *
+ * the character set/encoding is "guessed" based on the most common non-Unicode
+ * encoding used for man pages.
+ */
+
+typedef struct {
+    const char *lang;
+    const char *encoding;
+} lc2enc_t;
+
+static const lc2enc_t lc2enc[] = {
+    { "C",     "ASCII"      },
+    { "de",    "ISO-8859-1" },
+    { "en",    "ISO-8859-1" },
+    { "es",    "ISO-8859-1" },
+    { "fr",    "ISO-8859-1" },
+    { "it",    "ISO-8859-1" },
+    { "pt",    "ISO-8859-1" },
+    { "hr",    "ISO-8859-2" },
+    { "cs",    "ISO-8859-2" },
+    { "et",    "ISO-8859-2" },
+    { "hu",    "ISO-8859-2" },
+    { "lv",    "ISO-8859-2" },
+    { "lt",    "ISO-8859-2" },
+    { "pl",    "ISO-8859-2" },
+    { "ro",    "ISO-8859-2" },
+    { "sk",    "ISO-8859-2" },
+    { "sl",    "ISO-8859-2" },
+    { "bg",    "KOI8-R"     },
+    { "be",    "KOI8-R"     },
+    { "mk",    "KOI8-R"     },
+    { "ru",    "KOI8-R"     },
+    { "sh",    "KOI8-R"     },
+    { "uk",    "KOI8-R"     },
+    { "el",    "ISO-8859-7" },
+    { "tr",    "ISO-8859-9" },
+    { "ja",    "EUC-JP"     },
+    { "ko",    "EUC-KR"     },
+    { "zh_CN", "GB2312"     },
+    { "zh_HK", "BIG5"       },
+    { "zh_TW", "BIG5"       },
+    { NULL }
+};
+
+static const char *
+get_legacy_encoding(const char *file) {
+    const lc2enc_t *ptr;
+
+    for (ptr = lc2enc; ptr->lang != NULL; ptr++) {
+        if (is_lang_page(ptr->lang, file)) {
+            return ptr->encoding;
+        }
+    }
+    return "ISO-8859-1";
+}
+
 static int
 parse_roff_directive (char *cp, const char *file, char *buf, int buflen) {
      char c;
      int tbl_found = 0;
-     int use_jroff;
-
-     use_jroff = (is_japanese &&
-		   (strstr(file, "/jman/") || is_lang_page(language, file)));
 
      while ((c = *cp++) != '\0') {
 	  switch (c) {
 	  case 'e':
 	       if (debug)
 		    gripe (FOUND_EQN);
-	       add_directive((do_troff ? "EQN" : use_jroff ? "JNEQN": "NEQN"),
-			     file, buf, buflen);
+	       add_directive((do_troff ? "EQN" : "NEQN"), file, buf, buflen);
 	       break;
 
 	  case 'g':
@@ -520,9 +572,27 @@
      if (*buf == 0)
 	  return 1;
 
-     add_directive (do_troff ? "TROFF" : use_jroff ? "JNROFF" : "NROFF",
-		    "", buf, buflen);
+     add_directive (do_troff ? "TROFF" : "NROFF", "", buf, buflen);
 
+     if (!do_troff && strstr(buf, "NROFF_OLD_CHARSET") != NULL) {
+         const char *encoding = NULL;
+         size_t len = strlen("NROFF_OLD_CHARSET");
+         char *p = strstr(buf, "NROFF_OLD_CHARSET");
+
+         if (debug) {
+             fprintf(stderr, "\nfound '%s' in path\n", "NROFF_OLD_CHARSET");
+         }
+         encoding = get_legacy_encoding(file);
+         if (debug) {
+             fprintf(stderr, "\nold charset of '%s' is '%s'\n", file, encoding);
+         }
+         if (strlen(encoding) < len) {
+             memmove(p, p + len, strlen(p + len) + 1);
+             memmove(p + strlen(encoding), p, strlen(p) + 1);
+             memcpy(p, encoding, strlen(encoding));
+         }
+     }
+     
      if (tbl_found && !do_troff && *getval("COL"))
 	  add_directive ("COL", "", buf, buflen);
 
@@ -1181,22 +1251,6 @@
      return status;
 }
 
-/* Special code for Japanese (to pick jnroff instead of nroff, etc.) */
-static void
-setlang(void) {
-	char *lang;
-
-	/* We use getenv() instead of setlocale(), because of
-	   glibc 2.1.x security policy for SetUID/SetGID binary. */
-	if ((lang = getenv("LANG")) == NULL &&
-	    (lang = getenv("LC_ALL")) == NULL &&
-	    (lang = getenv("LC_CTYPE")) == NULL)
-		/* nothing */;
-
-	language = lang;
-	is_japanese = (lang && strncmp(lang, "ja", 2) == 0);
-}
-
 /*
  * Handle the apropos option.  Cheat by using another program.
  */
@@ -1263,10 +1317,6 @@
      setlocale(LC_MESSAGES, "");
 #endif
 
-     /* No doubt we'll need some generic language code here later.
-	For the moment only Japanese support. */
-     setlang();
-
      /* Handle /usr/man/man1.Z/name.1 nonsense from HP */
      dohp = getenv("MAN_HP_DIREXT");		/* .Z */
 
--- man-1.6b/src/man.conf.in.nroff	2005-11-10 10:02:58.000000000 +0100
+++ man-1.6b/src/man.conf.in	2005-11-10 10:39:04.442673600 +0100
@@ -86,14 +86,12 @@
 # (Maybe - but today I need -Tlatin1 to prevent double conversion to utf8.)
 #
 # If you have a new troff (version 1.18.1?) and its colored output
-# causes problems, add the -c option to TROFF, NROFF, JNROFF.
+# causes problems, add the -c option to TROFF, NROFF.
 #
 TROFF		@troff@
 NROFF		@nroff@
-JNROFF		@jnroff@
 EQN		@eqn@
 NEQN		@neqn@
-JNEQN		@jneqn@
 TBL		@tbl@
 @nocol@COL		@col@
 REFER		@refer@
--- man-1.6b/src/paths.h.in.nroff	2005-08-21 01:26:06.000000000 +0200
+++ man-1.6b/src/paths.h.in	2005-11-10 10:38:09.156078440 +0100
@@ -16,10 +16,8 @@
     { "WHATIS",   "@whatis@" },
     { "TROFF",	"@troff@" },
     { "NROFF",	"@nroff@" },
-    { "JNROFF",	"@jnroff@" },
     { "EQN",	"@eqn@" },
     { "NEQN",	"@neqn@" },
-    { "JNEQN",	"@jneqn@" },
     { "TBL",	"@tbl@" },
     { "COL",	"@pcol@" },
     { "REFER",	"@refer@" },