path: root/gcc/ABOUT-GCC-NLS

Notes on GCC's Native Language Support

GCC's Native Language Support (NLS) is relatively new and
experimental, so NLS is currently disabled by default.

The main reason for it being buggy is, that GCC does not set the
locale categories correctly.  Currently only LC_MESSAGES is set if the
system supports it and else nothing.  To work correctly, GCC would have
to also set the character set used by the terminal by either setting
LC_CTYPE together with LC_MESSAGES or LC_ALL if LC_MESSAGES is
not supported.

This would change the behaviour of GCC in quite a few places because
a number of standard C functions and macros change their behaviour
depending on the locale.  These necessary changes have been done in the
development version, but these changes are beyond the scope
of a maintenance release such as this.  It is therefore recommended that
you leave it disabled.

If you still want to enable the feature, use configure's --enable-nls
option to enable it.  Eventually, NLS will be enabled by default, and
you'll need --disable-nls to disable it.  You must enable NLS in order
to make a GCC distribution.

By and large, only diagnostic messages have been internationalized.
Some work remains in other areas; for example, GCC does not yet allow
non-ASCII letters in identifiers.

Not all of GCC's diagnostic messages have been internationalized.
Programs like `enquire' and `genattr' are not internationalized, as
their users are GCC maintainers who typically need to be able to read
English anyway; internationalizing them would thus entail needless
work for the human translators.  And no one has yet gotten around to
internationalizing the messages in the C++ compiler, or in the
specialized MIPS-specific programs mips-tdump and mips-tfile.

The GCC library should not contain any messages that need
internationalization, because it operates below the
internationalization library.

Currently, the only language translation supplied is en_UK (British English).

Unlike some other GNU programs, the GCC sources contain few instances
of explicit translation calls like _("string").  Instead, the
diagnostic printing routines automatically translate their arguments.
For example, GCC source code should not contain calls like `error
(_("unterminated comment"))'; it should contain calls like `error
("unterminated comment")' instead, as it is the `error' function's
responsibility to translate the message before the user sees it.

By convention, any function parameter in the GCC sources whose name
ends in `msgid' is expected to be a message requiring translation.
For example, the `error' function's first parameter is named `msgid'.
GCC's exgettext script uses this convention to determine which
function parameter strings need to be translated.  The exgettext
script also assumes that any occurrence of `%eMSGID}' on a source
line, where MSGID does not contain `%' or `}', corresponds to a
message MSGID that requires translation; this is needed to identify
diagnostics in GCC spec strings.

If you enable NLS and modify source files, you'll need to use a
special version of the GNU gettext package to propagate the
modifications to the translation tables.  Apply the following patch
(use `patch -p0') to GNU gettext 0.10.35, which you can retrieve from:

ftp://alpha.gnu.org/gnu/gettext-0.10.35.tar.gz

This patch has been submitted to the GNU gettext maintainer, so
eventually we shouldn't need this special gettext version.

This patch is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.

This patch is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this patch; see the file COPYING.  If not, write to
the Free Software Foundation, 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA.

1998-07-26  Paul Eggert  <eggert@twinsun.com>

	* po/Makefile.in.in (maintainer-clean): Remove cat-id-tbl.c and
	stamp-cat-id.

1998-07-24  Paul Eggert  <eggert@twinsun.com>

	* po/Makefile.in.in (cat-id-tbl.o): Depend on
	$(top_srcdir)/intl/libgettext.h, not ../intl/libgettext.h.

1998-07-20  Paul Eggert  <eggert@twinsun.com>

	* po/Makefile.in.in (.po.pox, all-yes, $(srcdir)/cat-id-tbl.c,
	$(srcdir)/stamp-cat-id, update-po): Prepend `$(srcdir)/' to
	files built in the source directory; this is needed for
	VPATH-based make in Solaris 2.6.

1998-07-17  Paul Eggert  <eggert@twinsun.com>

	Add support for user-specified argument numbers for keywords.
	Extract all strings from a keyword arg, not just the first one.
	Handle parenthesized commas inside keyword args correctly.
	Warn about nested keywords.

	* doc/gettext.texi: Document --keyword=id:argnum.

	* src/xgettext.c (scan_c_file):
	Warn about nested keywords, e.g. _(_("xxx")).
	Warn also about not-yet-implemented but allowed nesting, e.g.
	dcgettext(..._("xxx")..., "yyy").
	Get all strings in a keyword arg, not just the first one.
	Handle parenthesized commas inside keyword args correctly.

	* src/xget-lex.h (enum xgettext_token_type_ty):
	Replace xgettext_token_type_keyword1 and
	xgettext_token_type_keyword2 with just plain
	xgettext_token_type_keyword; it now has argnum value.
	Add xgettext_token_type_rp.
	(struct xgettext_token_ty): Add argnum member.
	line_number and file_name are now also set for
	xgettext_token_type_keyword.
	(xgettext_lex_keyword): Arg is const char *.

	* src/xget-lex.c: Include "hash.h".
	(enum token_type_ty): Add token_type_rp.
	(keywords): Now a hash table.
	(phase5_get): Return token_type_rp for ')'.
	(xgettext_lex, xgettext_lex_keyword): Add support for keyword argnums.
	(xgettext_lex): Return xgettext_token_type_rp for ')'.
	Report keyword argnum, line number, and file name back to caller.

1998-07-09  Paul Eggert  <eggert@twinsun.com>

        * intl/Makefile.in (uninstall):
	Do nothing unless $(PACKAGE) is gettext.

===================================================================
RCS file: doc/gettext.texi,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.1
diff -pu -r0.10.35.0 -r0.10.35.1
--- doc/gettext.texi	1998/05/01 05:53:32	0.10.35.0
+++ doc/gettext.texi	1998/07/18 00:25:15	0.10.35.1
@@ -1854,13 +1854,19 @@ List of directories searched for input f
 Join messages with existing file.
 
 @item -k @var{word}
-@itemx --keyword[=@var{word}]
-Additonal keyword to be looked for (without @var{word} means not to
+@itemx --keyword[=@var{keywordspec}]
+Additonal keyword to be looked for (without @var{keywordspec} means not to
 use default keywords).
 
-The default keywords, which are always looked for if not explicitly
-disabled, are @code{gettext}, @code{dgettext}, @code{dcgettext} and
-@code{gettext_noop}.
+If @var{keywordspec} is a C identifer @var{id}, @code{xgettext} looks
+for strings in the first argument of each call to the function or macro
+@var{id}.  If @var{keywordspec} is of the form
+@samp{@var{id}:@var{argnum}}, @code{xgettext} looks for strings in the
+@var{argnum}th argument of the call.
+
+The default keyword specifications, which are always looked for if not
+explicitly disabled, are @code{gettext}, @code{dgettext:2},
+@code{dcgettext:2} and @code{gettext_noop}.
 
 @item -m [@var{string}]
 @itemx --msgstr-prefix[=@var{string}]
===================================================================
RCS file: intl/Makefile.in,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.1
diff -pu -r0.10.35.0 -r0.10.35.1
--- intl/Makefile.in	1998/04/27 21:53:18	0.10.35.0
+++ intl/Makefile.in	1998/07/09 21:39:18	0.10.35.1
@@ -143,10 +143,14 @@ install-data: all
 installcheck:
 
 uninstall:
-	dists="$(DISTFILES.common)"; \
-	for file in $$dists; do \
-	  rm -f $(gettextsrcdir)/$$file; \
-	done
+	if test "$(PACKAGE)" = "gettext"; then \
+	  dists="$(DISTFILES.common)"; \
+	  for file in $$dists; do \
+	    rm -f $(gettextsrcdir)/$$file; \
+	  done
+	else \
+	  : ; \
+	fi
 
 info dvi:
 
===================================================================
RCS file: src/xget-lex.c,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.1
diff -pu -r0.10.35.0 -r0.10.35.1
--- src/xget-lex.c	1998/07/09 22:49:48	0.10.35.0
+++ src/xget-lex.c	1998/07/18 00:25:15	0.10.35.1
@@ -33,6 +33,7 @@
 #include "error.h"
 #include "system.h"
 #include "libgettext.h"
+#include "hash.h"
 #include "str-list.h"
 #include "xget-lex.h"
 
@@ -83,6 +84,7 @@ enum token_type_ty
   token_type_eoln,
   token_type_hash,
   token_type_lp,
+  token_type_rp,
   token_type_comma,
   token_type_name,
   token_type_number,
@@ -109,7 +111,7 @@ static FILE *fp;
 static int trigraphs;
 static int cplusplus_comments;
 static string_list_ty *comment;
-static string_list_ty *keywords;
+static hash_table keywords;
 static int default_keywords = 1;
 
 /* These are for tracking whether comments count as immediately before
@@ -941,6 +943,10 @@ phase5_get (tp)
       tp->type = token_type_lp;
       return;
 
+    case ')':
+      tp->type = token_type_rp;
+      return;
+
     case ',':
       tp->type = token_type_comma;
       return;
@@ -1179,6 +1185,7 @@ xgettext_lex (tp)
   while (1)
     {
       token_ty token;
+      void *keyword_value;
 
       phase8_get (&token);
       switch (token.type)
@@ -1213,17 +1220,20 @@ xgettext_lex (tp)
 	  if (default_keywords)
 	    {
 	      xgettext_lex_keyword ("gettext");
-	      xgettext_lex_keyword ("dgettext");
-	      xgettext_lex_keyword ("dcgettext");
+	      xgettext_lex_keyword ("dgettext:2");
+	      xgettext_lex_keyword ("dcgettext:2");
 	      xgettext_lex_keyword ("gettext_noop");
 	      default_keywords = 0;
 	    }
 
-	  if (string_list_member (keywords, token.string))
-	    {
-	      tp->type = (strcmp (token.string, "dgettext") == 0
-			  || strcmp (token.string, "dcgettext") == 0)
-		? xgettext_token_type_keyword2 : xgettext_token_type_keyword1;
+	  if (find_entry (&keywords, token.string, strlen (token.string),
+			  &keyword_value)
+	      == 0)
+	    {
+	      tp->type = xgettext_token_type_keyword;
+	      tp->argnum = (int) keyword_value;
+	      tp->line_number = token.line_number;
+	      tp->file_name = logical_file_name;
 	    }
 	  else
 	    tp->type = xgettext_token_type_symbol;
@@ -1236,6 +1246,12 @@ xgettext_lex (tp)
 	  tp->type = xgettext_token_type_lp;
 	  return;
 
+	case token_type_rp:
+	  last_non_comment_line = newline_count;
+
+	  tp->type = xgettext_token_type_rp;
+	  return;
+
 	case token_type_comma:
 	  last_non_comment_line = newline_count;
 
@@ -1263,16 +1279,32 @@ xgettext_lex (tp)
 
 void
 xgettext_lex_keyword (name)
-     char *name;
+     const char *name;
 {
   if (name == NULL)
     default_keywords = 0;
   else
     {
-      if (keywords == NULL)
-	keywords = string_list_alloc ();
+      int argnum;
+      size_t len;
+      const char *sp;
+
+      if (keywords.table == NULL)
+	init_hash (&keywords, 100);
+
+      sp = strchr (name, ':');
+      if (sp)
+	{
+	  len = sp - name;
+	  argnum = atoi (sp + 1);
+	}
+      else
+	{
+	  len = strlen (name);
+	  argnum = 1;
+	}
 
-      string_list_append_unique (keywords, name);
+      insert_entry (&keywords, name, len, (void *) argnum);
     }
 }
 
===================================================================
RCS file: src/xget-lex.h,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.1
diff -pu -r0.10.35.0 -r0.10.35.1
--- src/xget-lex.h	1998/07/09 22:49:48	0.10.35.0
+++ src/xget-lex.h	1998/07/18 00:25:15	0.10.35.1
@@ -23,9 +23,9 @@ Foundation, Inc., 59 Temple Place - Suit
 enum xgettext_token_type_ty
 {
   xgettext_token_type_eof,
-  xgettext_token_type_keyword1,
-  xgettext_token_type_keyword2,
+  xgettext_token_type_keyword,
   xgettext_token_type_lp,
+  xgettext_token_type_rp,
   xgettext_token_type_comma,
   xgettext_token_type_string_literal,
   xgettext_token_type_symbol
@@ -37,8 +37,14 @@ struct xgettext_token_ty
 {
   xgettext_token_type_ty type;
 
-  /* These 3 are only set for xgettext_token_type_string_literal.  */
+  /* This 1 is set only for xgettext_token_type_keyword.  */
+  int argnum;
+
+  /* This 1 is set only for xgettext_token_type_string_literal.  */
   char *string;
+
+  /* These 2 are set only for xgettext_token_type_keyword and
+     xgettext_token_type_string_literal.  */
   int line_number;
   char *file_name;
 };
@@ -50,7 +56,7 @@ void xgettext_lex PARAMS ((xgettext_toke
 const char *xgettext_lex_comment PARAMS ((size_t __n));
 void xgettext_lex_comment_reset PARAMS ((void));
 /* void xgettext_lex_filepos PARAMS ((char **, int *)); FIXME needed?  */
-void xgettext_lex_keyword PARAMS ((char *__name));
+void xgettext_lex_keyword PARAMS ((const char *__name));
 void xgettext_lex_cplusplus PARAMS ((void));
 void xgettext_lex_trigraphs PARAMS ((void));
 
===================================================================
RCS file: src/xgettext.c,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.1
diff -pu -r0.10.35.0 -r0.10.35.1
--- src/xgettext.c	1998/07/09 22:49:48	0.10.35.0
+++ src/xgettext.c	1998/07/18 00:25:15	0.10.35.1
@@ -835,6 +835,8 @@ scan_c_file(filename, mlp, is_cpp_file)
      int is_cpp_file;
 {
   int state;
+  int commas_to_skip;	/* defined only when in states 1 and 2 */
+  int paren_nesting;	/* defined only when in state 2 */
 
   /* Inform scanner whether we have C++ files or not.  */
   if (is_cpp_file)
@@ -854,63 +856,79 @@ scan_c_file(filename, mlp, is_cpp_file)
    {
      xgettext_token_ty token;
 
-     /* A simple state machine is used to do the recognising:
+     /* A state machine is used to do the recognising:
         State 0 = waiting for something to happen
-        State 1 = seen one of our keywords with string in first parameter
-        State 2 = was in state 1 and now saw a left paren
-	State 3 = seen one of our keywords with string in second parameter
-	State 4 = was in state 3 and now saw a left paren
-	State 5 = waiting for comma after being in state 4
-	State 6 = saw comma after being in state 5  */
+        State 1 = seen one of our keywords
+        State 2 = waiting for part of an argument */
      xgettext_lex (&token);
      switch (token.type)
        {
-       case xgettext_token_type_keyword1:
+       case xgettext_token_type_keyword:
+	 if (!extract_all && state == 2)
+	   {
+	     if (commas_to_skip == 0)
+	       {
+		 error (0, 0,
+			_("%s:%d: warning: keyword nested in keyword arg"),
+			token.file_name, token.line_number);
+		 continue;
+	       }
+
+	     /* Here we should nest properly, but this would require a
+		potentially unbounded stack.  We haven't run across an
+		example that needs this functionality yet.  For now,
+		we punt and forget the outer keyword.  */
+	     error (0, 0,
+		    _("%s:%d: warning: keyword between outer keyword and its arg"),
+		    token.file_name, token.line_number);
+	   }
+	 commas_to_skip = token.argnum - 1;
 	 state = 1;
 	 continue;
 
-       case xgettext_token_type_keyword2:
-	 state = 3;
-	 continue;
-
        case xgettext_token_type_lp:
 	 switch (state)
 	   {
 	   case 1:
+	     paren_nesting = 0;
 	     state = 2;
 	     break;
-	   case 3:
-	     state = 4;
+	   case 2:
+	     paren_nesting++;
 	     break;
-	   default:
-	     state = 0;
 	   }
 	 continue;
 
+       case xgettext_token_type_rp:
+	 if (state == 2 && paren_nesting != 0)
+	   paren_nesting--;
+	 else
+	   state = 0;
+	 continue;
+
        case xgettext_token_type_comma:
-	 state = state == 5 ? 6 : 0;
+	 if (state == 2 && commas_to_skip != 0)
+	   commas_to_skip -= paren_nesting == 0;
+	 else
+	   state = 0;
 	 continue;
 
        case xgettext_token_type_string_literal:
-	 if (extract_all || state == 2 || state == 6)
-	   {
-	     remember_a_message (mlp, &token);
-	     state = 0;
-	   }
+	 if (extract_all || (state == 2 && commas_to_skip == 0))
+	   remember_a_message (mlp, &token);
 	 else
 	   {
 	     free (token.string);
-	     state = (state == 4 || state == 5) ? 5 : 0;
+	     state = state == 2 ? 2 : 0;
 	   }
 	 continue;
 
        case xgettext_token_type_symbol:
-	 state = (state == 4 || state == 5) ? 5 : 0;
+	 state = state == 2 ? 2 : 0;
 	 continue;
 
        default:
-	 state = 0;
-	 continue;
+	 abort ();
 
        case xgettext_token_type_eof:
 	 break;
===================================================================
RCS file: po/Makefile.in.in,v
retrieving revision 0.10.35.0
retrieving revision 0.10.35.5
diff -u -r0.10.35.0 -r0.10.35.5
--- po/Makefile.in.in	1998/07/20 20:20:38	0.10.35.0
+++ po/Makefile.in.in	1998/07/26 09:07:52	0.10.35.5
@@ -62,7 +62,7 @@
 	$(COMPILE) $<
 
 .po.pox:
-	$(MAKE) $(PACKAGE).pot
+	$(MAKE) $(srcdir)/$(PACKAGE).pot
 	$(MSGMERGE) $< $(srcdir)/$(PACKAGE).pot -o $*.pox
 
 .po.mo:
@@ -79,7 +79,7 @@
 
 all: all-@USE_NLS@
 
-all-yes: cat-id-tbl.c $(CATALOGS)
+all-yes: $(srcdir)/cat-id-tbl.c $(CATALOGS)
 all-no:
 
 $(srcdir)/$(PACKAGE).pot: $(POTFILES)
@@ -90,8 +90,8 @@
 	   || ( rm -f $(srcdir)/$(PACKAGE).pot \
 		&& mv $(PACKAGE).po $(srcdir)/$(PACKAGE).pot )
 
-$(srcdir)/cat-id-tbl.c: stamp-cat-id; @:
-$(srcdir)/stamp-cat-id: $(PACKAGE).pot
+$(srcdir)/cat-id-tbl.c: $(srcdir)/stamp-cat-id; @:
+$(srcdir)/stamp-cat-id: $(srcdir)/$(PACKAGE).pot
 	rm -f cat-id-tbl.tmp
 	sed -f ../intl/po2tbl.sed $(srcdir)/$(PACKAGE).pot \
 		| sed -e "s/@PACKAGE NAME@/$(PACKAGE)/" > cat-id-tbl.tmp
@@ -180,7 +180,8 @@
 
 check: all
 
-cat-id-tbl.o: ../intl/libgettext.h
+cat-id-tbl.o: $(srcdir)/cat-id-tbl.c $(top_srcdir)/intl/libgettext.h
+	$(COMPILE) $(srcdir)/cat-id-tbl.c
 
 dvi info tags TAGS ID:
 
@@ -196,7 +197,7 @@
 maintainer-clean: distclean
 	@echo "This command is intended for maintainers to use;"
 	@echo "it deletes files that may require special tools to rebuild."
-	rm -f $(GMOFILES)
+	rm -f $(GMOFILES) cat-id-tbl.c stamp-cat-id
 
 distdir = ../$(PACKAGE)-$(VERSION)/$(subdir)
 dist distdir: update-po $(DISTFILES)
@@ -207,7 +208,7 @@
 	done
 
 update-po: Makefile
-	$(MAKE) $(PACKAGE).pot
+	$(MAKE) $(srcdir)/$(PACKAGE).pot
 	PATH=`pwd`/../src:$$PATH; \
 	cd $(srcdir); \
 	catalogs='$(CATALOGS)'; \