From a8414720e43671e45d9b3c71a0f7833e9c7fad0d Mon Sep 17 00:00:00 2001
From: Evan Miller <emmiller@gmail.com>
Date: Sun, 20 Dec 2020 17:56:33 -0500
Subject: [PATCH] Use wcstombs when iconv is not present (#223)

Replace the jerry-built UTF-16 => Latin-1 code path with a cross-platform wcstombs solution that emits UTF-8.

This adds an element to the end of the MdbHandle struct, but should not break any existing code.

A run-time option could be added later to emit other encodings, but people who care about such things can just use the iconv code path.
---
 configure.ac       |  4 ++--
 include/mdbtools.h |  9 +++++++++
 src/libmdb/data.c  |  3 ++-
 src/libmdb/iconv.c | 42 ++++++++++++++++++++++++++++++++++--------
 4 files changed, 47 insertions(+), 11 deletions(-)

diff --git a/configure.ac b/configure.ac
index 93bd00c..6fa6e36 100644
--- a/configure.ac
+++ b/configure.ac
@@ -27,7 +27,7 @@ AC_PROG_YACC
 
 dnl Checks for header files.
 AC_HEADER_STDC
-AC_CHECK_HEADERS(fcntl.h limits.h unistd.h)
+AC_CHECK_HEADERS(fcntl.h limits.h unistd.h xlocale.h)
 AC_CHECK_HEADERS(wordexp.h)
 AC_CHECK_LIB(mswstr, DBLCMapStringW)
 AC_CHECK_DECLS([program_invocation_short_name], [], [], [[
@@ -36,7 +36,7 @@ AC_CHECK_DECLS([program_invocation_short_name], [], [], [[
 
 dnl Checks for library functions.
 VL_LIB_READLINE
-AC_CHECK_FUNCS(strptime fmemopen gmtime_r)
+AC_CHECK_FUNCS(strptime fmemopen gmtime_r wcstombs_l)
 
 dnl Checks for typedefs, structures, and compiler characteristics.
 AC_C_CONST
diff --git a/include/mdbtools.h b/include/mdbtools.h
index e17e414..8c19783 100644
--- a/include/mdbtools.h
+++ b/include/mdbtools.h
@@ -39,6 +39,11 @@
 
 #ifdef HAVE_ICONV
 #include <iconv.h>
+#else
+#ifdef HAVE_XLOCALE_H
+#include <xlocale.h>
+#endif
+#include <locale.h>
 #endif
 
 #ifdef _WIN32
@@ -291,6 +296,10 @@ typedef struct {
 #ifdef HAVE_ICONV
 	iconv_t	iconv_in;
 	iconv_t	iconv_out;
+#elif defined(_WIN32) || defined(WIN32) || defined(_WIN64) || defined(WIN64) || defined(WINDOWS)
+    _locale_t locale;
+#else
+    locale_t locale;
 #endif
 } MdbHandle; 
 
diff --git a/src/libmdb/data.c b/src/libmdb/data.c
index b32844b..73db744 100644
--- a/src/libmdb/data.c
+++ b/src/libmdb/data.c
@@ -16,10 +16,11 @@
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  */
 
+#include "mdbtools.h"
+
 #define _XOPEN_SOURCE
 #include <time.h>
 #include <math.h>
-#include "mdbtools.h"
 
 #define OFFSET_MASK 0x1fff
 #define OLE_BUFFER_SIZE (MDB_BIND_SIZE*64)
diff --git a/src/libmdb/iconv.c b/src/libmdb/iconv.c
index fb6e084..87264f1 100644
--- a/src/libmdb/iconv.c
+++ b/src/libmdb/iconv.c
@@ -78,14 +78,32 @@ static size_t decompressed2ascii_without_iconv(MdbHandle *mdb, const char *in_pt
 		snprintf(dest, dlen, "%.*s%n", (int)len_in, in_ptr, &count);
 		return count;
 	}
-	/* rough UCS-2LE to ISO-8859-1 conversion */
-	/* wcstombs would be better; see libxls implementation for 
-	 * a multi-platform solution */
-	unsigned int i;
-	for (i=0; 2*i+1<len_in && i<dlen-1; i++)
-		dest[i] = (in_ptr[2*i+1] == 0) ? in_ptr[2*i] : '?';
-	dest[i] = '\0';
-	return i;
+    size_t i;
+    size_t count = 0;
+    size_t len_out = dlen - 1;
+    wchar_t *w = malloc((len_in/2+1)*sizeof(wchar_t));
+
+    for(i=0; i<len_in/2; i++)
+    {
+        w[i] = (unsigned char)in_ptr[2*i] + ((unsigned char)in_ptr[2*i+1] << 8);
+    }
+    w[len_in/2] = '\0';
+
+#if defined(_WIN32) || defined(WIN32) || defined(_WIN64) || defined(WIN64) || defined(WINDOWS)
+    count = _wcstombs_l(dest, w, len_out, mdb->locale);
+#elif defined(HAVE_WCSTOMBS_L)
+    count = wcstombs_l(dest, w, len_out, mdb->locale);
+#else
+    locale_t oldlocale = uselocale(mdb->locale);
+    count = wcstombs(dest, w, len_out);
+    uselocale(oldlocale);
+#endif
+    free(w);
+    if (count == (size_t)-1)
+        return 0;
+
+    dest[count] = '\0';
+	return count;
 }
 #endif
 
@@ -247,6 +265,10 @@ void mdb_iconv_init(MdbHandle *mdb)
 		mdb->iconv_out = iconv_open(jet3_iconv_code, iconv_code);
 		mdb->iconv_in = iconv_open(iconv_code, jet3_iconv_code);
 	}
+#elif defined(_WIN32) || defined(WIN32) || defined(_WIN64) || defined(WIN64) || defined(WINDOWS)
+    mdb->locale = _create_locale(LC_CTYPE, ".65001");
+#else
+    mdb->locale = newlocale(LC_CTYPE_MASK, "C.UTF-8", NULL);
 #endif
 }
 void mdb_iconv_close(MdbHandle *mdb)
@@ -254,6 +276,10 @@ void mdb_iconv_close(MdbHandle *mdb)
 #ifdef HAVE_ICONV
     if (mdb->iconv_out != (iconv_t)-1) iconv_close(mdb->iconv_out);
     if (mdb->iconv_in != (iconv_t)-1) iconv_close(mdb->iconv_in);
+#elif defined(_WIN32) || defined(WIN32) || defined(_WIN64) || defined(WIN64) || defined(WINDOWS)
+    if (mdb->locale) _free_locale(mdb->locale);
+#else
+    if (mdb->locale) freelocale(mdb->locale);
 #endif
 }