Re: i18n for yacc backend messages

[Paul Eggert]

Jan Nieuwenhuizen <address@hidden> writes:

> I'll see if I can come up with something better, how's the patch below?

On further thought I'm leery of translating the token names, as they
are part of the input, and are not really English-language strings.

I installed the following more-conservative patch, as it is some
progress anyway.

Can you please explain why the token names should be translated?  It's
possible that this should be an option, rather than being an
unconditional translation.

2005-04-13  Paul Eggert  <address@hidden>

        * NEWS: Bison-generated C parsers now use the _ macro to
        translate strings.
        * data/yacc.c (_) [!defined _]: New macro.
        All English strings wrapped inside this macro.
        * doc/bison.texinfo (Bison Parser): Document _.
        * po/POTFILES.in: Include src/parse-gram.c, since it now
        includes translateable strings that parse-gram.y doesn't.

Index: NEWS
===================================================================
RCS file: /cvsroot/bison/bison/NEWS,v
retrieving revision 1.110
diff -p -u -r1.110 NEWS
--- NEWS        26 Dec 2004 05:49:52 -0000      1.110
+++ NEWS        14 Apr 2005 00:07:46 -0000
@@ -1,6 +1,12 @@
 Bison News
 ----------
 
+Changes in version 2.0a, ????-??-??:
+
+* Bison-generated C parsers use the _ macro to translate strings from
+  English to the user's language, e.g., _("syntax error").  By default,
+  _ is defined to be a no-op macro so the strings are not translated.
+
 Changes in version 2.0, 2004-12-25:
 
 * Possibly-incompatible changes
@@ -593,7 +599,7 @@ End:
 
 -----
 
-Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004
+Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
 Free Software Foundation, Inc.
 
 This file is part of Bison, the GNU Compiler Compiler.
Index: data/yacc.c
===================================================================
RCS file: /cvsroot/bison/bison/data/yacc.c,v
retrieving revision 1.83
diff -p -u -r1.83 yacc.c
--- data/yacc.c 17 Mar 2005 19:19:44 -0000      1.83
+++ data/yacc.c 14 Apr 2005 00:07:46 -0000
@@ -375,6 +375,11 @@ static const char *const yytname[] =
 };
 #endif
 
+/* INFRINGES ON USER NAME SPACE */
+#ifndef _
+# define _(msgid) msgid
+#endif
+
 # ifdef YYPRINT
 /* YYTOKNUM[YYLEX-NUM] -- Internal token number corresponding to
    token YYLEX-NUM.  */
@@ -492,7 +497,7 @@ do                                                          
\
     }                                                          \
   else                                                         \
     {                                                          \
-      yyerror (]b4_yyerror_args["syntax error: cannot back up");\
+      yyerror (]b4_yyerror_args[_("syntax error: cannot back up")); \
       YYERROR;                                                 \
     }                                                          \
 while (0)
@@ -586,7 +591,7 @@ do {                                                        
        \
                    [[short int *bottom], [bottom]],
                    [[short int *top],    [top]])[
 {
-  YYFPRINTF (stderr, "Stack now");
+  YYFPRINTF (stderr, _("Stack now"));
   for (/* Nothing. */; bottom <= top; ++bottom)
     YYFPRINTF (stderr, " %d", *bottom);
   YYFPRINTF (stderr, "\n");
@@ -608,7 +613,7 @@ do {                                                        
        \
 {
   int yyi;
   unsigned int yylno = yyrline[yyrule];
-  YYFPRINTF (stderr, "Reducing stack by rule %d (line %u), ",
+  YYFPRINTF (stderr, _("Reducing stack by rule %d (line %u), "),
              yyrule - 1, yylno);
   /* Print the symbols being reduced, and their result.  */
   for (yyi = yyprhs[yyrule]; 0 <= yyrhs[yyi]; yyi++)
@@ -877,7 +882,7 @@ b4_syncline(address@hidden@], address@hidden@])])dnl
           data in use in that stack, in bytes.  This used to be a
           conditional around just the two extra args, but that might
           be undefined if yyoverflow is a macro.  */
-       yyoverflow ("parser stack overflow",
+       yyoverflow (_("parser stack overflow"),
                    &yyss1, yysize * sizeof (*yyssp),
                    &yyvs1, yysize * sizeof (*yyvsp),
 ]b4_location_if([                  &yyls1, yysize * sizeof (*yylsp),])[
@@ -917,14 +922,14 @@ b4_syncline(address@hidden@], address@hidden@])])dnl
       yyvsp = yyvs + yysize - 1;
 ]b4_location_if([      yylsp = yyls + yysize - 1;])[
 
-      YYDPRINTF ((stderr, "Stack size increased to %lu\n",
+      YYDPRINTF ((stderr, _("Stack size increased to %lu\n"),
                  (unsigned long int) yystacksize));
 
       if (yyss + yystacksize - 1 <= yyssp)
        YYABORT;
     }
 
-  YYDPRINTF ((stderr, "Entering state %d\n", yystate));
+  YYDPRINTF ((stderr, _("Entering state %d\n"), yystate));
 
   goto yybackup;
 
@@ -948,19 +953,19 @@ yybackup:
   /* YYCHAR is either YYEMPTY or YYEOF or a valid look-ahead symbol.  */
   if (yychar == YYEMPTY)
     {
-      YYDPRINTF ((stderr, "Reading a token: "));
+      YYDPRINTF ((stderr, _("Reading a token: ")));
       yychar = YYLEX;
     }
 
   if (yychar <= YYEOF)
     {
       yychar = yytoken = YYEOF;
-      YYDPRINTF ((stderr, "Now at end of input.\n"));
+      YYDPRINTF ((stderr, _("Now at end of input.\n")));
     }
   else
     {
       yytoken = YYTRANSLATE (yychar);
-      YY_SYMBOL_PRINT ("Next token is", yytoken, &yylval, &yylloc);
+      YY_SYMBOL_PRINT (_("Next token is"), yytoken, &yylval, &yylloc);
     }
 
   /* If the proper action on seeing token YYTOKEN is to reduce or to
@@ -981,7 +986,7 @@ yybackup:
     YYACCEPT;
 
   /* Shift the look-ahead token.  */
-  YY_SYMBOL_PRINT ("Shifting", yytoken, &yylval, &yylloc);
+  YY_SYMBOL_PRINT (_("Shifting"), yytoken, &yylval, &yylloc);
 
   /* Discard the token being shifted unless it is eof.  */
   if (yychar != YYEOF)
@@ -1091,7 +1096,7 @@ yyerrlab:
          int yyxend = yychecklim < YYNTOKENS ? yychecklim : YYNTOKENS;
          int yycount = 0;
 
-         yyprefix = ", expecting ";
+         yyprefix = _(", expecting ");
          for (yyx = yyxbegin; yyx < yyxend; ++yyx)
            if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR)
              {
@@ -1103,34 +1108,34 @@ yyerrlab:
                    break;
                  }
              }
-         yysize += (sizeof ("syntax error, unexpected ")
-                    + yystrlen (yytname[yytype]));
+         yysize += (strlen (_("syntax error, unexpected "))
+                    + yystrlen (yytname[yytype]) + 1);
          yymsg = (char *) YYSTACK_ALLOC (yysize);
          if (yymsg != 0)
            {
-             char *yyp = yystpcpy (yymsg, "syntax error, unexpected ");
+             char *yyp = yystpcpy (yymsg, _("syntax error, unexpected "));
              yyp = yystpcpy (yyp, yytname[yytype]);
 
              if (yycount < 5)
                {
-                 yyprefix = ", expecting ";
+                 yyprefix = _(", expecting ");
                  for (yyx = yyxbegin; yyx < yyxend; ++yyx)
                    if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR)
                      {
                        yyp = yystpcpy (yyp, yyprefix);
                        yyp = yystpcpy (yyp, yytname[yyx]);
-                       yyprefix = " or ";
+                       yyprefix = _(" or ");
                      }
                }
              yyerror (]b4_yyerror_args[yymsg);
              YYSTACK_FREE (yymsg);
            }
          else
-           yyerror (]b4_yyerror_args["syntax error; also virtual memory 
exhausted");
+           yyerror (]b4_yyerror_args[_("syntax error; also virtual memory 
exhausted"));
        }
       else
 #endif /* YYERROR_VERBOSE */
-       yyerror (]b4_yyerror_args["syntax error");
+       yyerror (]b4_yyerror_args[_("syntax error"));
     }
 
 ]b4_location_if([[  yyerror_range[0] = yylloc;]])[
@@ -1151,13 +1156,13 @@ yyerrlab:
                 YYPOPSTACK;
                 if (yyssp == yyss)
                   YYABORT;
-                yydestruct ("Error: popping",
+                yydestruct (_("Error: popping"),
                              yystos[*yyssp], yyvsp]b4_location_if([, yylsp])[);
               }
         }
       else
        {
-         yydestruct ("Error: discarding", yytoken, &yylval]b4_location_if([, 
&yylloc])[);
+         yydestruct (_("Error: discarding"), yytoken, 
&yylval]b4_location_if([, &yylloc])[);
          yychar = YYEMPTY;
        }
     }
@@ -1211,7 +1216,7 @@ yyerrlab1:
        YYABORT;
 
 ]b4_location_if([[      yyerror_range[0] = *yylsp;]])[
-      yydestruct ("Error: popping", yystos[yystate], yyvsp]b4_location_if([, 
yylsp])[);
+      yydestruct (_("Error: popping"), yystos[yystate], 
yyvsp]b4_location_if([, yylsp])[);
       YYPOPSTACK;
       yystate = *yyssp;
       YY_STACK_PRINT (yyss, yyssp);
@@ -1229,7 +1234,7 @@ yyerrlab1:
   *++yylsp = yyloc;]])[
 
   /* Shift the error token. */
-  YY_SYMBOL_PRINT ("Shifting", yystos[yyn], yyvsp, yylsp);
+  YY_SYMBOL_PRINT (_("Shifting"), yystos[yyn], yyvsp, yylsp);
 
   yystate = yyn;
   goto yynewstate;
@@ -1246,7 +1251,7 @@ yyacceptlab:
 | yyabortlab -- YYABORT comes here.  |
 `-----------------------------------*/
 yyabortlab:
-  yydestruct ("Error: discarding lookahead",
+  yydestruct (_("Error: discarding lookahead"),
               yytoken, &yylval]b4_location_if([, &yylloc])[);
   yychar = YYEMPTY;
   yyresult = 1;
@@ -1257,7 +1262,7 @@ yyabortlab:
 | yyoverflowlab -- parser overflow comes here.  |
 `----------------------------------------------*/
 yyoverflowlab:
-  yyerror (]b4_yyerror_args["parser stack overflow");
+  yyerror (]b4_yyerror_args[_("parser stack overflow"));
   yyresult = 2;
   /* Fall through.  */
 #endif
Index: doc/bison.texinfo
===================================================================
RCS file: /cvsroot/bison/bison/doc/bison.texinfo,v
retrieving revision 1.142
diff -p -u -r1.142 bison.texinfo
--- doc/bison.texinfo   1 Mar 2005 00:41:33 -0000       1.142
+++ doc/bison.texinfo   14 Apr 2005 00:07:47 -0000
@@ -1166,7 +1166,14 @@ start with a function called @code{main}
 arrange for it to call @code{yyparse} or the parser will never run.
 @xref{Interface, ,Parser C-Language Interface}.
 
-Aside from the token type names and the symbols in the actions you
+If your code defines a C preprocessor macro @code{_} (a single
+underscore), Bison assumes that it can be used to translate
+English-language strings to the user's preferred language using a
+function-like syntax, e.g., @code{_("syntax error")}.  Otherwise,
+Bison defines a no-op macro by that name that merely returns its
+argument, so strings are not translated.
+
+Aside from @code{_} and the token type names and the symbols in the actions you
 write, all symbols defined in the Bison parser file itself
 begin with @samp{yy} or @samp{YY}.  This includes interface functions
 such as the lexical analyzer function @code{yylex}, the error reporting
Index: po/POTFILES.in
===================================================================
RCS file: /cvsroot/bison/bison/po/POTFILES.in,v
retrieving revision 1.19
diff -p -u -r1.19 POTFILES.in
--- po/POTFILES.in      13 Nov 2002 06:40:06 -0000      1.19
+++ po/POTFILES.in      14 Apr 2005 00:07:47 -0000
@@ -4,6 +4,7 @@ src/files.c
 src/getargs.c
 src/gram.c
 src/main.c
+src/parse-gram.c
 src/parse-gram.y
 src/print.c
 src/reader.c