lib/util/charset Use push_string and talloc_strupper/strlower from common code

The only caller of push_string() (not to be confused with
push_string_check()) in the common code was encode_pw_buffer(), and it
didn't use the alignment or STR_UPPER flags.

The talloc_strupper() and talloc_strlower() functions are tested in
smbtorture, and are next_codepoint() based.

Andrew Bartlett

6 files changed, 12 insertions, 145 deletions

diff --git a/lib/util/charset/util_unistr.c b/lib/util/charset/util_unistr.c
index 1915c73ad5..e4ae65053c 100644
--- a/lib/util/charset/util_unistr.c
+++ b/lib/util/charset/util_unistr.c
@@ -194,7 +194,7 @@ _PUBLIC_ size_t count_chars_m(const char *s, char c)
  * @param dest_len the maximum length in bytes allowed in the
  * destination.  If @p dest_len is -1 then no maximum is used.
  **/
-static bool push_ascii(void *dest, const char *src, size_t dest_len, int flags, size_t *converted_size)
+static bool push_ascii_string(void *dest, const char *src, size_t dest_len, int flags, size_t *converted_size)
 {
 	size_t src_len;
 	bool ret;
@@ -204,7 +204,7 @@ static bool push_ascii(void *dest, const char *src, size_t dest_len, int flags,
 		if (tmpbuf == NULL) {
 			return false;
 		}
-		ret = push_ascii(dest, tmpbuf, dest_len, flags & ~STR_UPPER, converted_size);
+		ret = push_ascii_string(dest, tmpbuf, dest_len, flags & ~STR_UPPER, converted_size);
 		talloc_free(tmpbuf);
 		return ret;
 	}
@@ -232,7 +232,7 @@ static bool push_ascii(void *dest, const char *src, size_t dest_len, int flags,
  * @param src_len is the length of the source area in bytes.
  * @returns the number of bytes occupied by the string in @p src.
  **/
-static ssize_t pull_ascii(char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
+static ssize_t pull_ascii_string(char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
 {
 	size_t size = 0;
 
@@ -373,7 +373,7 @@ _PUBLIC_ ssize_t push_string(void *dest, const char *src, size_t dest_len, int f
 {
 	if (flags & STR_ASCII) {
 		size_t size = 0;
-		if (push_ascii(dest, src, dest_len, flags, &size)) {
+		if (push_ascii_string(dest, src, dest_len, flags, &size)) {
 			return (ssize_t)size;
 		} else {
 			return (ssize_t)-1;
@@ -404,7 +404,7 @@ _PUBLIC_ ssize_t push_string(void *dest, const char *src, size_t dest_len, int f
 _PUBLIC_ ssize_t pull_string(char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
 {
 	if (flags & STR_ASCII) {
-		return pull_ascii(dest, src, dest_len, src_len, flags);
+		return pull_ascii_string(dest, src, dest_len, src_len, flags);
 	} else if (flags & STR_UNICODE) {
 		return pull_ucs2(dest, src, dest_len, src_len, flags);
 	} else {
diff --git a/lib/util/charset/wscript_build b/lib/util/charset/wscript_build
index 771ff5dc24..7623131146 100644
--- a/lib/util/charset/wscript_build
+++ b/lib/util/charset/wscript_build
@@ -1,18 +1,11 @@
 #!/usr/bin/env python
 
-
-if bld.env._SAMBA_BUILD_ == 4:
-    bld.SAMBA_SUBSYSTEM('CHARSET',
-                        source='util_unistr.c',
-                        public_deps='CODEPOINTS',
-                        public_headers='charset.h',
-                        )
-
 bld.SAMBA_SUBSYSTEM('ICONV_WRAPPER',
                     source='iconv.c',
                     public_deps='iconv replace talloc')
 
-bld.SAMBA_SUBSYSTEM('CODEPOINTS',
-	source='codepoints.c convert_string.c util_str.c util_unistr_w.c charcnv.c pull_push.c',
-	deps='DYNCONFIG ICONV_WRAPPER'
-	)
+bld.SAMBA_SUBSYSTEM('CHARSET',
+                    public_headers='charset.h',
+                    source='codepoints.c convert_string.c util_str.c util_unistr_w.c charcnv.c pull_push.c util_unistr.c',
+                    deps='DYNCONFIG ICONV_WRAPPER'
+                    )
diff --git a/source3/Makefile.in b/source3/Makefile.in
index 6893e01f2c..5e7b63dae9 100644
--- a/source3/Makefile.in
+++ b/source3/Makefile.in
@@ -458,6 +458,7 @@ LIB_OBJ = $(LIBSAMBAUTIL_OBJ) $(UTIL_OBJ) $(CRYPTO_OBJ) \
 	  lib/wins_srv.o \
 	  lib/util_str.o ../lib/util/util_str_common.o \
 	  ../lib/util/base64.o lib/util_sid.o \
+	  ../lib/util/charset/util_unistr.o \
 	  ../lib/util/charset/util_unistr_w.o ../lib/util/charset/codepoints.o ../lib/util/charset/util_str.o lib/util_file.o \
 	  lib/util.o lib/util_cmdline.o lib/util_names.o \
 	  lib/util_sock.o lib/sock_exec.o lib/util_sec.o \
diff --git a/source3/include/proto.h b/source3/include/proto.h
index 086ba6c557..e8ce418ab2 100644
--- a/source3/include/proto.h
+++ b/source3/include/proto.h
@@ -82,10 +82,6 @@ bool convert_string_error(charset_t from, charset_t to,
 			    void const *src, size_t srclen,
 			    void *dest, size_t destlen,
 			    size_t *converted_size);
-char *talloc_strdup_upper(TALLOC_CTX *ctx, const char *s);
-char *strupper_talloc(TALLOC_CTX *ctx, const char *s);
-char *talloc_strdup_lower(TALLOC_CTX *ctx, const char *s);
-char *strlower_talloc(TALLOC_CTX *ctx, const char *s);
 size_t ucs2_align(const void *base_ptr, const void *p, int flags);
 size_t push_ascii(void *dest, const char *src, size_t dest_len, int flags);
 size_t push_ascii_fstring(void *dest, const char *src);
diff --git a/source3/lib/charcnv.c b/source3/lib/charcnv.c
index 6b667502da..765e3943ef 100644
--- a/source3/lib/charcnv.c
+++ b/source3/lib/charcnv.c
@@ -45,98 +45,6 @@ void init_iconv(void)
 }
 
 /**
- talloc_strdup() a unix string to upper case.
-**/
-
-char *talloc_strdup_upper(TALLOC_CTX *ctx, const char *s)
-{
-	char *out_buffer = talloc_strdup(ctx,s);
-	const unsigned char *p = (const unsigned char *)s;
-	unsigned char *q = (unsigned char *)out_buffer;
-
-	if (!q) {
-		return NULL;
-	}
-
-	/* this is quite a common operation, so we want it to be
-	   fast. We optimise for the ascii case, knowing that all our
-	   supported multi-byte character sets are ascii-compatible
-	   (ie. they match for the first 128 chars) */
-
-	while (*p) {
-		if (*p & 0x80)
-			break;
-		*q++ = toupper_ascii_fast(*p);
-		p++;
-	}
-
-	if (*p) {
-		/* MB case. */
-		size_t converted_size, converted_size2;
-		smb_ucs2_t *ubuf = NULL;
-
-		/* We're not using the ascii buffer above. */
-		TALLOC_FREE(out_buffer);
-
-		if (!convert_string_talloc(ctx, CH_UNIX, CH_UTF16LE, s,
-					   strlen(s)+1, (void *)&ubuf,
-					   &converted_size))
-		{
-			return NULL;
-		}
-
-		strupper_w(ubuf);
-
-		if (!convert_string_talloc(ctx, CH_UTF16LE, CH_UNIX, ubuf,
-					   converted_size, (void *)&out_buffer,
-					   &converted_size2))
-		{
-			TALLOC_FREE(ubuf);
-			return NULL;
-		}
-
-		/* Don't need the intermediate buffer
- 		 * anymore.
- 		 */
-		TALLOC_FREE(ubuf);
-	}
-
-	return out_buffer;
-}
-
-char *strupper_talloc(TALLOC_CTX *ctx, const char *s) {
-	return talloc_strdup_upper(ctx, s);
-}
-
-
-char *talloc_strdup_lower(TALLOC_CTX *ctx, const char *s)
-{
-	size_t converted_size;
-	smb_ucs2_t *buffer = NULL;
-	char *out_buffer;
-
-	if (!push_ucs2_talloc(ctx, &buffer, s, &converted_size)) {
-		return NULL;
-	}
-
-	strlower_w(buffer);
-
-	if (!pull_ucs2_talloc(ctx, &out_buffer, buffer, &converted_size)) {
-		TALLOC_FREE(buffer);
-		return NULL;
-	}
-
-	TALLOC_FREE(buffer);
-
-	return out_buffer;
-}
-
-char *strlower_talloc(TALLOC_CTX *ctx, const char *s) {
-	return talloc_strdup_lower(ctx, s);
-}
-
-
-/**
  * Copy a string from a char* unix src to a dos codepage string destination.
  *
  * @return the number of bytes occupied by the string in the destination.
@@ -627,36 +535,6 @@ size_t push_string_base(const char *base, uint16 flags2,
 }
 
 /**
- Copy a string from a char* src to a unicode or ascii
- dos codepage destination choosing unicode or ascii based on the 
- flags supplied
- Return the number of bytes occupied by the string in the destination.
- flags can have:
-  STR_TERMINATE means include the null termination.
-  STR_UPPER     means uppercase in the destination.
-  STR_ASCII     use ascii even with unicode packet.
-  STR_NOALIGN   means don't do alignment.
- dest_len is the maximum length allowed in the destination. If dest_len
- is -1 then no maxiumum is used.
-**/
-
-ssize_t push_string(void *dest, const char *src, size_t dest_len, int flags)
-{
-	size_t ret;
-
-	if (!(flags & STR_ASCII) && \
-	    (flags & STR_UNICODE)) {
-		ret = push_ucs2(NULL, dest, src, dest_len, flags);
-	} else {
-		ret = push_ascii(dest, src, dest_len, flags);
-	}
-	if (ret == (size_t)-1) {
-		return -1;
-	}
-	return ret;
-}
-
-/**
  Copy a string from a unicode or ascii source (depending on
  the packet flags) to a char* destination.
  Flags can have:
diff --git a/source3/wscript_build b/source3/wscript_build
index 83cc39ebe7..fbc6693346 100755
--- a/source3/wscript_build
+++ b/source3/wscript_build
@@ -964,7 +964,7 @@ bld.SAMBA3_SUBSYSTEM('tdb-wrap3',
 
 bld.SAMBA3_SUBSYSTEM('CHARSET3',
                     source='''lib/util_str.c lib/charcnv.c lib/fstring.c''',
-                    public_deps='ICONV_WRAPPER CODEPOINTS',
+                    public_deps='ICONV_WRAPPER CHARSET',
                     deps='samba-util')
 
 bld.SAMBA3_SUBSYSTEM('ldb3',
@@ -1317,7 +1317,6 @@ if not bld.env.toplevel_build:
     bld.SAMBA3_SUBSYSTEM('tdb-wrap', source='', deps='tdb-wrap3')
     bld.SAMBA3_SUBSYSTEM('errors', source='', deps='errors3')
     bld.SAMBA3_SUBSYSTEM('samba-util', source='', deps='DYNCONFIG')
-    bld.SAMBA3_SUBSYSTEM('CHARSET', source='', deps='CHARSET3')
     bld.SAMBA3_SUBSYSTEM('ldb', source='', deps='ldb3')
     bld.SAMBA3_SUBSYSTEM('dcerpc', '', deps='UTIL_TEVENT')
     bld.SAMBA3_SUBSYSTEM('cli-ldap', '', deps='UTIL_TEVENT')