Arrange to convert command-line arguments from the right encoding.

This is a temporary workaround for the 2.0 stable series.  The next
stable series should have an implicit `setlocale (LC_ALL, "")' call,
which will make this unnecessary.

* libguile/feature.c (progargs_fluid): Rename to...
  (scm_program_arguments_fluid): ... this.  Update users.

* libguile/feature.h (scm_program_arguments_fluid): New internal
  declaration.

* libguile/init.c (invoke_main_func): Call
  `scm_i_set_boot_program_arguments' instead of
  `scm_set_program_arguments'.

* libguile/script.c (locale_arguments_to_string_list,
  scm_i_set_boot_program_arguments): New functions.
  (scm_compile_shell_switches): Use `locale_arguments_to_string_list'.

* libguile/script.h (scm_i_set_boot_program_arguments): New internal
  declaration.

* test-suite/standalone/Makefile.am (check_SCRIPTS, TESTS): Add
  `test-command-line-encoding'.
* test-suite/standalone/test-command-line-encoding: New file.

libguile/feature.c

@@ -1,4 +1,5 @@
-/* Copyright (C) 1995,1996,1998,1999,2000,2001,2002, 2003, 2004, 2006, 2007, 2009 Free Software Foundation, Inc.
+/* Copyright (C) 1995, 1996, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
+ *   2006, 2007, 2009, 2011 Free Software Foundation, Inc.
  *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public License
@@ -36,7 +37,8 @@
 
 
 
-static SCM progargs_fluid;
+SCM scm_program_arguments_fluid;
+
 static SCM features_var;
 
 void
@@ -58,7 +60,7 @@ SCM_DEFINE (scm_program_arguments, "program-arguments", 0, 0, 0,
 	    "options like @code{-e} and @code{-l}.")
 #define FUNC_NAME s_scm_program_arguments
 {
-  return scm_fluid_ref (progargs_fluid);
+  return scm_fluid_ref (scm_program_arguments_fluid);
 }
 #undef FUNC_NAME
 
@@ -74,7 +76,7 @@ scm_set_program_arguments (int argc, char **argv, char *first)
   SCM args = scm_makfromstrs (argc, argv);
   if (first)
     args = scm_cons (scm_from_locale_string (first), args);
-  scm_fluid_set_x (progargs_fluid, args);
+  scm_fluid_set_x (scm_program_arguments_fluid, args);
 }
 
 SCM_DEFINE (scm_set_program_arguments_scm, "set-program-arguments", 1, 0, 0, 
@@ -89,7 +91,7 @@ SCM_DEFINE (scm_set_program_arguments_scm, "set-program-arguments", 1, 0, 0,
 	    "strings within it are copied, so should not be modified later.")
 #define FUNC_NAME s_scm_set_program_arguments_scm
 {
-  return scm_fluid_set_x (progargs_fluid, lst);
+  return scm_fluid_set_x (scm_program_arguments_fluid, lst);
 }
 #undef FUNC_NAME
 
@@ -99,7 +101,7 @@ SCM_DEFINE (scm_set_program_arguments_scm, "set-program-arguments", 1, 0, 0,
 void
 scm_init_feature()
 {
-  progargs_fluid = scm_make_fluid ();
+  scm_program_arguments_fluid = scm_make_fluid ();
 
   features_var = scm_c_define ("*features*", SCM_EOL);
 #ifndef _Windows

libguile/feature.h

@@ -3,7 +3,8 @@
 #ifndef SCM_FEATURE_H
 #define SCM_FEATURE_H
 
-/* Copyright (C) 1995,1996,1999,2000,2001, 2006, 2007, 2008 Free Software Foundation, Inc.
+/* Copyright (C) 1995, 1996, 1999, 2000, 2001, 2006, 2007, 2008,
+ *   2011 Free Software Foundation, Inc.
  *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public License
@@ -29,6 +30,8 @@ SCM_API void scm_add_feature (const char* str);
 SCM_API SCM scm_program_arguments (void);
 SCM_API void scm_set_program_arguments (int argc, char **argv, char *first);
 SCM_API SCM scm_set_program_arguments_scm (SCM lst);
+
+SCM_INTERNAL SCM scm_program_arguments_fluid;
 SCM_INTERNAL void scm_init_feature (void);
 
 #endif  /* SCM_FEATURE_H */

libguile/init.c

@@ -332,7 +332,7 @@ invoke_main_func (void *body_data)
 {
   struct main_func_closure *closure = (struct main_func_closure *) body_data;
 
-  scm_set_program_arguments (closure->argc, closure->argv, 0);
+  scm_i_set_boot_program_arguments (closure->argc, closure->argv);
   (*closure->main_func) (closure->closure, closure->argc, closure->argv);
 
   scm_restore_signals ();

libguile/script.c

@@ -22,10 +22,12 @@
 #  include <config.h>
 #endif
 
+#include <localcharset.h>
 #include <stdlib.h>
 #include <stdio.h>
 #include <errno.h>
 #include <ctype.h>
+#include <uniconv.h>
 
 #include "libguile/_scm.h"
 #include "libguile/eval.h"
@@ -368,6 +370,41 @@ scm_shell_usage (int fatal, char *message)
                : SCM_BOOL_F));
 }
 
+/* Return a list of strings from ARGV, which contains ARGC strings
+   assumed to be encoded in the current locale.  Use
+   `environ_locale_charset' instead of relying on
+   `scm_from_locale_string' because the user hasn't had a change to call
+   (setlocale LC_ALL "") yet.
+
+   XXX: This hack is for 2.0 and will be removed in the next stable
+   series where the `setlocale' call will be implicit.  See
+   <http://lists.gnu.org/archive/html/guile-devel/2011-11/msg00040.html>
+   for details.  */
+static SCM
+locale_arguments_to_string_list (int argc, char **const argv)
+{
+  int i;
+  SCM lst;
+  const char *encoding;
+
+  encoding = environ_locale_charset ();
+  for (i = argc - 1, lst = SCM_EOL;
+       i >= 0;
+       i--)
+    lst = scm_cons (scm_from_stringn (argv[i], (size_t) -1, encoding,
+				      SCM_FAILED_CONVERSION_ESCAPE_SEQUENCE),
+		    lst);
+
+  return lst;
+}
+
+/* Set the value returned by `program-arguments', given ARGC and ARGV.  */
+void
+scm_i_set_boot_program_arguments (int argc, char *argv[])
+{
+  scm_fluid_set_x (scm_program_arguments_fluid,
+		   locale_arguments_to_string_list (argc, argv));
+}
 
 /* Given an array of command-line switches, return a Scheme expression
    to carry out the actions specified by the switches.
@@ -378,7 +415,7 @@ scm_compile_shell_switches (int argc, char **argv)
 {
   return scm_call_2 (scm_c_public_ref ("ice-9 command-line",
                                        "compile-shell-switches"),
-                     scm_makfromstrs (argc, argv),
+		     locale_arguments_to_string_list (argc, argv),
                      (scm_usage_name
                       ? scm_from_locale_string (scm_usage_name)
                       : scm_from_latin1_string ("guile")));

libguile/script.h

@@ -3,7 +3,7 @@
 #ifndef SCM_SCRIPT_H
 #define SCM_SCRIPT_H
 
-/* Copyright (C) 1997,1998,2000, 2006, 2008 Free Software Foundation, Inc.
+/* Copyright (C) 1997,1998,2000, 2006, 2008, 2011 Free Software Foundation, Inc.
  *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public License
@@ -37,6 +37,7 @@ SCM_API void scm_shell_usage (int fatal, char *message);
 SCM_API SCM scm_compile_shell_switches (int argc, char **argv);
 SCM_API void scm_shell (int argc, char **argv);
 SCM_API char *scm_usage_name;
+SCM_INTERNAL void scm_i_set_boot_program_arguments (int argc, char *argv[]);
 SCM_INTERNAL void scm_init_script (void);
 
 #endif  /* SCM_SCRIPT_H */

test-suite/standalone/Makefile.am

@@ -82,6 +82,9 @@ TESTS += test-import-order
 EXTRA_DIST += test-import-order-a.scm test-import-order-b.scm \
   test-import-order-c.scm test-import-order-d.scm
 
+check_SCRIPTS += test-command-line-encoding
+TESTS += test-command-line-encoding
+
 # test-num2integral
 test_num2integral_SOURCES = test-num2integral.c
 test_num2integral_CFLAGS = ${test_cflags}

test-suite/standalone/test-command-line-encoding

@@ -0,0 +1,24 @@
+#!/bin/sh
+
+# Choose a UTF-8 locale.  The locale doesn't have to be available on the
+# system since `environ_locale_charset' does not actually try to set it.
+LC_ALL="en_US.UTF-8"
+export LC_ALL
+unset LANG
+unset LC_CTYPE
+
+exec guile -q -s "$0" "λ"
+!#
+
+;; Make sure our first argument is a lower-case lambda.
+;;
+;; Up to Guile 2.0.3 included, command-line arguments would not be converted
+;; according to the locale settings; see
+;; <http://lists.gnu.org/archive/html/guile-devel/2011-11/msg00026.html> for
+;; details.
+(exit (string=? (cadr (program-arguments)) "λ"))
+
+;; Local Variables:
+;; mode: scheme
+;; coding: utf-8
+;; End: