Add ICONV_{GET,SET}_ILSEQ_INVALID iconvctl. GNU iconv returns EILSEQ

when there is an invalid character in the output codeset while it is
valid in the input.  However, POSIX requires iconv() to perform an
implementation-defined conversion on the character.  So, Citrus iconv converts
such a character to a special character which means it is invalid in the
output codeset.

This is not a problem in most cases but some software like libxml2 depends
on GNU's behavior to determine if a character is output as-is or another form
such as a character entity (&#NNN;).

5 files changed, 45 insertions, 1 deletions

diff --git a/include/iconv.h b/include/iconv.h
index da1036a..c07d02e 100644
--- a/include/iconv.h
+++ b/include/iconv.h
@@ -86,6 +86,8 @@ void	 iconv_set_relocation_prefix(const char *, const char *);
 #define ICONV_SET_DISCARD_ILSEQ	4
 #define ICONV_SET_HOOKS		5
 #define ICONV_SET_FALLBACKS	6
+#define ICONV_GET_ILSEQ_INVALID	128
+#define ICONV_SET_ILSEQ_INVALID	129
 
 typedef void (*iconv_unicode_char_hook) (unsigned int mbr, void *data);
 typedef void (*iconv_wide_char_hook) (wchar_t wc, void *data);
diff --git a/lib/libc/iconv/citrus_iconv_local.h b/lib/libc/iconv/citrus_iconv_local.h
index e673c9a..12d2fa3 100644
--- a/lib/libc/iconv/citrus_iconv_local.h
+++ b/lib/libc/iconv/citrus_iconv_local.h
@@ -99,6 +99,7 @@ struct _citrus_iconv_shared {
 	char						*ci_convname;
 	bool						 ci_discard_ilseq;
 	struct iconv_hooks				*ci_hooks;
+	bool						 ci_ilseq_invalid;
 };
 
 struct _citrus_iconv {
diff --git a/lib/libc/iconv/iconv.c b/lib/libc/iconv/iconv.c
index 555efd8..40a1a4e 100644
--- a/lib/libc/iconv/iconv.c
+++ b/lib/libc/iconv/iconv.c
@@ -298,6 +298,12 @@ __bsd_iconvctl(iconv_t cd, int request, void *argument)
 	case ICONV_SET_FALLBACKS:
 		errno = EOPNOTSUPP;
 		return (-1);
+	case ICONV_GET_ILSEQ_INVALID:
+		*i = cv->cv_shared->ci_ilseq_invalid ? 1 : 0;
+		return (0);
+	case ICONV_SET_ILSEQ_INVALID:
+		cv->cv_shared->ci_ilseq_invalid = *i;
+		return (0);
 	default:
 		errno = EINVAL;
 		return (-1);
diff --git a/lib/libc/iconv/iconvctl.3 b/lib/libc/iconv/iconvctl.3
index f012157..5bd06ea 100644
--- a/lib/libc/iconv/iconvctl.3
+++ b/lib/libc/iconv/iconvctl.3
@@ -34,7 +34,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd November 7, 2009
+.Dd November 25, 2009
 .Dt ICONVCTL 3
 .Os
 .Sh NAME
@@ -110,6 +110,31 @@ variable, which is passed to
 via
 .Fa argument
 by its address.
+.It ICONV_GET_ILSEQ_INVALID
+Determines if a character in the input buffer that is valid,
+but for which an identical character does not exist in the target
+codeset returns
+.Er EILSEQ
+or not.
+The answer is stored in
+.Fa argument ,
+which is of
+.Ft int * .
+It will be set to 1 if this feature is enabled or set to 0 otherwise.
+.It ICONV_SET_ILSEQ_INVALID
+Sets whether a character in the input buffer that is valid,
+but for which an identical character does not exist in the target
+codeset returns
+.Er EILSEQ
+or not.
+If
+.Fa argument ,
+which is of
+.Ft int *
+is set to 1 it will be enabled,
+and if
+.Fa argument
+is set to 0 it will be disabled.
 .El
 .\" XXX: fallbacks are unimplemented and trying to set them will always
 .\"      return EOPNOTSUPP but definitions are provided for source-level
diff --git a/lib/libiconv_modules/iconv_std/citrus_iconv_std.c b/lib/libiconv_modules/iconv_std/citrus_iconv_std.c
index b30f099..54a00d0 100644
--- a/lib/libiconv_modules/iconv_std/citrus_iconv_std.c
+++ b/lib/libiconv_modules/iconv_std/citrus_iconv_std.c
@@ -543,6 +543,16 @@ _citrus_iconv_std_iconv_convert(struct _citrus_iconv * __restrict cv,
 		ret = do_conv(is, &csid, &idx);
 		if (ret) {
 			if (ret == E_NO_CORRESPONDING_CHAR) {
+				/*
+				 * GNU iconv returns EILSEQ when no
+				 * corresponding character in the output.
+				 * Some software depends on this behavior
+				 * though this is against POSIX specification.
+				 */
+				if (cv->cv_shared->ci_ilseq_invalid != 0) {
+					ret = EILSEQ;
+					goto err;
+				}
 				inval++;
 				szrout = 0;
 				if ((((flags & _CITRUS_ICONV_F_HIDE_INVALID) == 0) &&