sh: Apply locale vars on builtins, recognize LC_MESSAGES as a locale var.

This allows doing things like LC_ALL=C some_builtin to run a builtin under a
different locale, just like is possible with external programs. The
immediate reason is that this allows making printf(1) a builtin without
breaking things like LC_NUMERIC=C printf '%f\n' 1.2

This change also affects special builtins, as even though the assignment is
persistent, the export is only to the builtin (unless the variable was
already exported).

Note: for this to work for builtins that also exist as external programs
such as /bin/test, the setlocale() call must be under #ifndef SHELL. The
shell will do the setlocale() calls which may not agree with the environment
variables.

4 files changed, 207 insertions, 7 deletions

diff --git a/bin/sh/eval.c b/bin/sh/eval.c
index e93fb14..c4495d0 100644
--- a/bin/sh/eval.c
+++ b/bin/sh/eval.c
@@ -937,6 +937,8 @@ evalcommand(union node *cmd, int flags, struct backcmd *backcmd)
 			cmdentry.special = 1;
 		if (cmdentry.special)
 			listsetvar(cmdenviron);
+		if (argc > 0)
+			bltinsetlocale();
 		commandname = argv[0];
 		argptr = argv + 1;
 		nextopt_optptr = NULL;		/* initialize nextopt */
@@ -944,6 +946,8 @@ evalcommand(union node *cmd, int flags, struct backcmd *backcmd)
 		exitstatus = (*builtinfunc[cmdentry.u.index])(argc, argv);
 		flushall();
 cmddone:
+		if (argc > 0)
+			bltinunsetlocale();
 		cmdenviron = NULL;
 		out1 = &output;
 		out2 = &errout;
diff --git a/bin/sh/var.c b/bin/sh/var.c
index d46e2c3..75de239 100644
--- a/bin/sh/var.c
+++ b/bin/sh/var.c
@@ -122,6 +122,14 @@ STATIC const struct varinit varinit[] = {
 
 STATIC struct var *vartab[VTABSIZE];
 
+STATIC const char *const locale_names[7] = {
+	"LC_COLLATE", "LC_CTYPE", "LC_MONETARY",
+	"LC_NUMERIC", "LC_TIME", "LC_MESSAGES", NULL
+};
+STATIC const int locale_categories[7] = {
+	LC_COLLATE, LC_CTYPE, LC_MONETARY, LC_NUMERIC, LC_TIME, LC_MESSAGES, 0
+};
+
 STATIC struct var **hashvar(const char *);
 STATIC int varequal(const char *, const char *);
 STATIC int localevar(const char *);
@@ -258,11 +266,7 @@ setvar(const char *name, const char *val, int flags)
 STATIC int
 localevar(const char *s)
 {
-	static const char *lnames[7] = {
-		"ALL", "COLLATE", "CTYPE", "MONETARY",
-		"NUMERIC", "TIME", NULL
-	};
-	const char **ss;
+	const char *const *ss;
 
 	if (*s != 'L')
 		return 0;
@@ -270,8 +274,10 @@ localevar(const char *s)
 		return 1;
 	if (strncmp(s + 1, "C_", 2) != 0)
 		return 0;
-	for (ss = lnames; *ss ; ss++)
-		if (varequal(s + 3, *ss))
+	if (varequal(s + 3, "ALL"))
+		return 1;
+	for (ss = locale_names; *ss ; ss++)
+		if (varequal(s + 3, *ss + 3))
 			return 1;
 	return 0;
 }
@@ -437,6 +443,61 @@ bltinlookup(const char *name, int doall)
 }
 
 
+/*
+ * Set up locale for a builtin (LANG/LC_* assignments).
+ */
+void
+bltinsetlocale(void)
+{
+	struct strlist *lp;
+	int act = 0;
+	char *loc, *locdef;
+	int i;
+
+	for (lp = cmdenviron ; lp ; lp = lp->next) {
+		if (localevar(lp->text)) {
+			act = 1;
+			break;
+		}
+	}
+	if (!act)
+		return;
+	loc = bltinlookup("LC_ALL", 0);
+	INTOFF;
+	if (loc != NULL) {
+		setlocale(LC_ALL, loc);
+		INTON;
+		return;
+	}
+	locdef = bltinlookup("LANG", 0);
+	for (i = 0; locale_names[i] != NULL; i++) {
+		loc = bltinlookup(locale_names[i], 0);
+		if (loc == NULL)
+			loc = locdef;
+		if (loc != NULL)
+			setlocale(locale_categories[i], loc);
+	}
+	INTON;
+}
+
+/*
+ * Undo the effect of bltinlocaleset().
+ */
+void
+bltinunsetlocale(void)
+{
+	struct strlist *lp;
+
+	INTOFF;
+	for (lp = cmdenviron ; lp ; lp = lp->next) {
+		if (localevar(lp->text)) {
+			setlocale(LC_ALL, "");
+			return;
+		}
+	}
+	INTON;
+}
+
 
 /*
  * Generate a list of exported variables.  This routine is used to construct
diff --git a/bin/sh/var.h b/bin/sh/var.h
index 9c792c8..6a02630 100644
--- a/bin/sh/var.h
+++ b/bin/sh/var.h
@@ -107,6 +107,8 @@ struct strlist;
 void listsetvar(struct strlist *);
 char *lookupvar(const char *);
 char *bltinlookup(const char *, int);
+void bltinsetlocale(void);
+void bltinunsetlocale(void);
 char **environment(void);
 int showvarscmd(int, char **);
 int exportcmd(int, char **);
diff --git a/tools/regression/bin/sh/builtins/locale1.0 b/tools/regression/bin/sh/builtins/locale1.0
new file mode 100644
index 0000000..93a37df
--- /dev/null
+++ b/tools/regression/bin/sh/builtins/locale1.0
@@ -0,0 +1,133 @@
+# $FreeBSD$
+# Note: this test depends on strerror() using locale.
+
+failures=0
+
+check() {
+	if ! eval "[ $1 ]"; then
+		echo "Failed: $1 at $2"
+		: $((failures += 1))
+	fi
+}
+
+unset LANG LC_ALL LC_COLLATE LC_CTYPE LC_MONETARY LC_NUMERIC LC_TIME LC_MESSAGES
+
+msgeng="No such file or directory"
+msgdut="Bestand of map niet gevonden"
+
+# Verify C locale error message.
+case $(command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+# Various locale variables that should not affect the message.
+case $(LC_ALL=C command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_ALL=C LANG=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_ALL=C LC_MESSAGES=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_CTYPE=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+# Verify Dutch message.
+case $(export LANG=nl_NL.ISO8859-1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(export LC_MESSAGES=nl_NL.ISO8859-1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(export LC_ALL=nl_NL.ISO8859-1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LANG=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_MESSAGES=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_ALL=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+# Verify that command assignments do not set the locale persistently.
+case $(command . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LANG=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_MESSAGES=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(LC_ALL=nl_NL.ISO8859-1 command . /var/empty/foo 2>&1; command . /var/empty/foo 2>&1) in
+	*"$msgdut"*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+# Check special builtin; add colon invocation to avoid depending on certain fix.
+case $(LC_ALL=nl_NL.ISO8859-1 . /var/empty/foo 2>&1; :) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+# Assignments on special builtins are exported to that builtin; the export
+# is not persistent.
+case $(LC_ALL=nl_NL.ISO8859-1 . /dev/null; . /var/empty/foo 2>&1) in
+	*"$msgeng"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+case $(export LC_ALL; LC_ALL=nl_NL.ISO8859-1 . /dev/null; . /var/empty/foo 2>&1) in
+	*"$msgdut"*) ok=1 ;;
+	*) ok=0 ;;
+esac
+check '$ok -eq 1' $LINENO
+
+exit $((failures > 0))