r2159: converted samba4 over to UTF-16.

I had previously thought this was unnecessary, as windows doesn't use
standards compliant UTF-16, and for filesystem operations treats bytes
as UCS-2, but Bjoern Jacke has pointed out to me that this means we
don't correctly store extended UTF-16 characters as UTF-8 on
disk. This can be seen with (for example) the gothic characters with
codepoints above 64k.

This commit also adds a LOCAL-ICONV torture test that tests the first
1 million codepoints against the system iconv library, and tests 5
million random UTF-16LE buffers for identical error handling to the
system iconv library.

the lib/iconv.c changes need backporting to samba3
(This used to be commit 756f28ac95feaa84b42402723d5f7286865c78db)

1 files changed, 298 insertions, 0 deletions

diff --git a/source4/torture/local/iconv.c b/source4/torture/local/iconv.c
new file mode 100644
index 0000000000..0867be0bee
--- /dev/null
+++ b/source4/torture/local/iconv.c
@@ -0,0 +1,298 @@
+/* 
+   Unix SMB/CIFS implementation.
+
+   local testing of iconv routines. This tests the system iconv code against
+   the built-in iconv code
+
+   Copyright (C) Andrew Tridgell 2004
+   
+   This program 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 of the License, or
+   (at your option) any later version.
+   
+   This program 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 program; if not, write to the Free Software
+   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#include "includes.h"
+
+/*
+  generate a UTF-16LE buffer for a given unicode codepoint
+*/
+static int gen_codepoint(unsigned int codepoint,
+			  char *buf, size_t *size)
+{
+	static iconv_t cd;
+	uint8_t in[4];
+	char *ptr_in;
+	size_t size_in, size_out, ret;
+	if (!cd) {
+		cd = iconv_open("UTF-16LE", "UCS-4LE");
+	}
+
+	in[0] = codepoint & 0xFF;
+	in[1] = (codepoint>>8) & 0xFF;
+	in[2] = (codepoint>>16) & 0xFF;
+	in[3] = (codepoint>>24) & 0xFF;
+
+	ptr_in = in;
+	size_in = 4;
+	size_out = 8;
+
+	ret = iconv(cd, &ptr_in, &size_in, &buf, &size_out);
+
+	*size = 8 - size_out;
+
+	return ret;
+}
+
+
+/*
+  work out the unicode codepoint of the first UTF-8 character in the buffer
+*/
+static unsigned int get_codepoint(char *buf, size_t size)
+{
+	static iconv_t cd;
+	uint8_t out[4];
+	char *ptr_out;
+	size_t size_out, size_in, ret;
+	if (!cd) {
+		cd = iconv_open("UCS-4LE", "UTF-8");
+	}
+
+	size_in = size;
+	ptr_out = out;
+	size_out = sizeof(out);
+	memset(out, 0, sizeof(out));
+
+	ret = iconv(cd, &buf, &size_in, &ptr_out, &size_out);
+
+	return out[0] | (out[1]<<8) | (out[2]<<16) | (out[3]<<24);
+}
+
+/*
+  display a buffer with name prefix
+*/
+static void show_buf(const char *name, uint8_t *buf, size_t size)
+{
+	int i;
+	printf("%s ", name);
+	for (i=0;i<size;i++) {
+		printf("%02x ", buf[i]);
+	}
+	printf("\n");
+}
+
+/*
+  given a UTF-16LE buffer, test the system and built-in iconv code
+  to make sure they do exactly the same thing in converting the buffer
+  to UTF-8, then convert it back again and ensure we get the same buffer
+  back
+*/
+static int test_buffer(uint8_t *inbuf, size_t size)
+{
+	uint8_t buf1[1000], buf2[1000], buf3[1000];
+	size_t outsize1, outsize2, outsize3;
+	const char *ptr_in;
+	char *ptr_out;
+	size_t size_in1, size_in2, size_in3;
+	size_t ret1, ret2, ret3, len1, len2;
+	int ok = 1;
+	int errno1, errno2;
+	static iconv_t cd;
+	static smb_iconv_t cd2, cd3;
+	
+	if (!cd) {
+		cd = iconv_open("UTF-8", "UTF-16LE");
+		cd2 = smb_iconv_open("UTF-8", "UTF-16LE");
+		cd3 = smb_iconv_open("UTF-16LE", "UTF-8");
+	}
+
+#if 0
+	int i;
+	for (i=0;i<50;i++) {
+		ptr_in = inbuf;
+		ptr_out = buf1;
+		size_in1 = size;
+		outsize1 = sizeof(buf1);
+
+		memset(ptr_out, 0, outsize1);
+		errno = 0;
+		ret1 = smb_iconv(cd2, &ptr_in, &size_in1, &ptr_out, &outsize1);
+		errno1 = errno;
+	}
+#endif
+
+	ptr_in = inbuf;
+	ptr_out = buf1;
+	size_in1 = size;
+	outsize1 = sizeof(buf1);
+
+	memset(ptr_out, 0, outsize1);
+	errno = 0;
+	ret1 = smb_iconv(cd2, &ptr_in, &size_in1, &ptr_out, &outsize1);
+	errno1 = errno;
+
+	ptr_in = inbuf;
+	ptr_out = buf2;
+	size_in2 = size;
+	outsize2 = sizeof(buf2);
+	
+	memset(ptr_out, 0, outsize2);
+	errno = 0;
+	ret2 = iconv(cd, &ptr_in, &size_in2, &ptr_out, &outsize2);
+	errno2 = errno;
+
+	len1 = sizeof(buf1) - outsize1;
+	len2 = sizeof(buf2) - outsize2;
+
+	/* codepoints above 1M are not interesting for now */
+	if (len2 > len1 && 
+	    memcmp(buf1, buf2, len1) == 0 && 
+	    get_codepoint(buf2+len1, len2-len1) >= (1<<20)) {
+		return ok;
+	}
+	if (len1 > len2 && 
+	    memcmp(buf1, buf2, len2) == 0 && 
+	    get_codepoint(buf1+len2, len1-len2) >= (1<<20)) {
+		return ok;
+	}
+
+	if (ret1 != ret2) {
+		printf("ret1=%d ret2=%d\n", ret1, ret2);
+		ok = 0;
+	}
+
+	if (errno1 != errno2) {
+		printf("e1=%s e2=%s\n", strerror(errno1), strerror(errno2));
+		show_buf(" rem1:", inbuf+(size-size_in1), size_in1);
+		show_buf(" rem2:", inbuf+(size-size_in2), size_in2);
+		ok = 0;
+	}
+	
+	if (outsize1 != outsize2) {
+		printf("\noutsize mismatch outsize1=%d outsize2=%d\n",
+		       outsize1, outsize2);
+		ok = 0;
+	}
+	
+	if (size_in1 != size_in2) {
+		printf("\nsize_in mismatch size_in1=%d size_in2=%d\n",
+		       size_in1, size_in2);
+		ok = 0;
+	}
+
+	if (!ok ||
+	    len1 != len2 ||
+	    memcmp(buf1, buf2, len1) != 0) {
+		printf("\nsize=%d ret1=%d ret2=%d\n", size, ret1, ret2);
+		show_buf(" IN1:", inbuf, size-size_in1);
+		show_buf(" IN2:", inbuf, size-size_in2);
+		show_buf("OUT1:", buf1, len1);
+		show_buf("OUT2:", buf2, len2);
+		if (len2 > len1 && memcmp(buf1, buf2, len1) == 0) {
+			printf("next codepoint is %u\n", get_codepoint(buf2+len1, len2-len1));
+		}
+		if (len1 > len2 && memcmp(buf1, buf2, len2) == 0) {
+			printf("next codepoint is %u\n", get_codepoint(buf1+len2,len1-len2));
+		}
+
+		ok = 0;
+	}
+	
+	if (!ok) return ok;
+
+	size = size - size_in1;
+	ptr_in = buf1;
+	ptr_out = buf3;
+	size_in3 = len1;
+	outsize3 = sizeof(buf3);
+
+	memset(ptr_out, 0, outsize3);
+	ret3 = smb_iconv(cd3, &ptr_in, &size_in3, &ptr_out, &outsize3);
+	
+	if (ret3 != 0) {
+		printf("pull failed - %s\n", strerror(errno));
+		ok = 0;
+	}
+	
+	if (outsize3 != sizeof(buf3) - size) {
+		printf("wrong outsize3 - %d should be %d\n", 
+		       outsize3, sizeof(buf3) - size);
+		ok = 0;
+	}
+	
+	if (memcmp(buf3, inbuf, size) != 0) {
+		int i;
+		printf("pull bytes mismatch:\n");
+		for (i=0;i<size;i++) {
+			printf("%02x ", inbuf[i]);
+		}
+		printf("\n");
+		for (i=0;i<size;i++) {
+			printf("%02x ", buf3[i]);
+		}
+		printf("\n");
+		ok = 0;
+	}
+
+	return ok;
+}
+
+BOOL torture_local_iconv(int dummy) 
+{
+	size_t size;
+	unsigned char inbuf[1000];
+	int ok = 1;
+	unsigned int codepoint, i, c;
+
+	srandom(time(NULL));
+
+	printf("Testing first 1M codepoints\n");
+	for (codepoint=0;ok && codepoint<(1<<20);codepoint++) {
+		if (gen_codepoint(codepoint, inbuf, &size) != 0) {
+			continue;
+		}
+
+		if (codepoint % 100 == 0) {
+			printf("codepoint=%u   \r", codepoint);
+		}
+
+		ok = test_buffer(inbuf, size);
+	}
+
+
+	printf("Testing 5M random UTF-16LE sequences\n");
+	for (i=0;ok && i<500000;i++) {
+		if (i % 100 == 0) {
+			printf("i=%u              \r", i);
+		}
+
+		size = random() % 100;
+		for (c=0;c<size;c++) {
+			if (random() % 100 < 80) {
+				inbuf[c] = random() % 128;
+			} else {
+				inbuf[c] = random();
+			}
+			if (random() % 10 == 0) {
+				inbuf[c] |= 0xd8;
+			}
+			if (random() % 10 == 0) {
+				inbuf[c] |= 0xdc;
+			}
+		}
+		ok = test_buffer(inbuf, size);
+	}
+
+	return ok == 1;
+}
+
+