shift read_hex_bytes() and parse_guid_string() into lib/util

read_hex_bytes() is going to be used in lib/util/rfc1738.c.

parse_guid_string() is shifted for two reasons: Firstly, it is called
very often in some operations, sometimes constituting a few percent of
the CPU load, and it makes several calls to read_hex_bytes(). We want
the compiler to be able to inline those calls if it thinks that is
wise. Secondly, there are other places that could do with fast GUID
parsing.

Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>

3 files changed, 116 insertions, 1 deletions

diff --git a/lib/util/util_str_hex.c b/lib/util/util_str_hex.c
new file mode 100644
index 00000000000..148735bd5f3
--- /dev/null
+++ b/lib/util/util_str_hex.c
@@ -0,0 +1,100 @@
+#include "replace.h"
+#include "libcli/util/ntstatus.h"
+#include "util_str_hex.h"
+
+NTSTATUS read_hex_bytes(const char *s, uint hexchars, uint64_t *dest)
+{
+	uint64_t x = 0;
+	uint i;
+	char c;
+
+	if ((hexchars & 1) || hexchars > 16) {
+		return NT_STATUS_INVALID_PARAMETER;
+	}
+
+	for (i = 0; i < hexchars; i++) {
+		x <<= 4;
+		c = s[i];
+		if (c >= '0' && c <= '9') {
+			x += c - '0';
+		}
+		else if (c >= 'a' && c <= 'f') {
+			x += c - 'a' + 10;
+		}
+		else if (c >= 'A' && c <= 'F') {
+			x += c - 'A' + 10;
+		}
+		else {
+			/* BAD character (including '\0') */
+			return NT_STATUS_INVALID_PARAMETER;
+		}
+	}
+	*dest = x;
+	return NT_STATUS_OK;
+}
+
+
+NTSTATUS parse_guid_string(const char *s,
+			   uint32_t *time_low,
+			   uint32_t *time_mid,
+			   uint32_t *time_hi_and_version,
+			   uint32_t *clock_seq,
+			   uint32_t *node)
+{
+	uint64_t tmp;
+	NTSTATUS status;
+	int i;
+	/* "e12b56b6-0a95-11d1-adbb-00c04fd8d5cd"
+                |     |    |    |    |
+                |     |    |    |    \ node[6]
+                |     |    |    \_____ clock_seq[2]
+                |     |    \__________ time_hi_and_version
+		|     \_______________ time_mid
+		\_____________________ time_low
+	*/
+	status = read_hex_bytes(s, 8, &tmp);
+	if (!NT_STATUS_IS_OK(status) || s[8] != '-') {
+		return NT_STATUS_INVALID_PARAMETER;
+	}
+	*time_low = tmp;
+	s += 9;
+
+	status = read_hex_bytes(s, 4, &tmp);
+	if (!NT_STATUS_IS_OK(status) || s[4] != '-') {
+		return NT_STATUS_INVALID_PARAMETER;
+	}
+	*time_mid = tmp;
+	s += 5;
+
+	status = read_hex_bytes(s, 4, &tmp);
+	if (!NT_STATUS_IS_OK(status) || s[4] != '-') {
+		return NT_STATUS_INVALID_PARAMETER;
+	}
+	*time_hi_and_version = tmp;
+	s += 5;
+
+	for (i = 0; i < 2; i++) {
+		status = read_hex_bytes(s, 2, &tmp);
+		if (!NT_STATUS_IS_OK(status)) {
+			return NT_STATUS_INVALID_PARAMETER;
+		}
+		clock_seq[i] = tmp;
+		s += 2;
+	}
+	if (s[0] != '-') {
+		return NT_STATUS_INVALID_PARAMETER;
+	}
+
+	s++;
+
+	for (i = 0; i < 6; i++) {
+		status = read_hex_bytes(s, 2, &tmp);
+		if (!NT_STATUS_IS_OK(status)) {
+			return NT_STATUS_INVALID_PARAMETER;
+		}
+		node[i] = tmp;
+		s += 2;
+	}
+
+	return NT_STATUS_OK;
+}
diff --git a/lib/util/util_str_hex.h b/lib/util/util_str_hex.h
new file mode 100644
index 00000000000..2d0ba65636b
--- /dev/null
+++ b/lib/util/util_str_hex.h
@@ -0,0 +1,10 @@
+#include "../libcli/util/ntstatus.h"
+
+NTSTATUS read_hex_bytes(const char *s, uint hexchars, uint64_t *dest);
+
+NTSTATUS parse_guid_string(const char *s,
+			   uint32_t *time_low,
+			   uint32_t *time_mid,
+			   uint32_t *time_hi_and_version,
+			   uint32_t *clock_seq,
+			   uint32_t *node);
diff --git a/lib/util/wscript_build b/lib/util/wscript_build
index 30face34381..8dbe1636e3b 100644
--- a/lib/util/wscript_build
+++ b/lib/util/wscript_build
@@ -123,7 +123,7 @@ else:
                     util_str.c util_str_common.c ms_fnmatch.c
                     server_id.c dprintf.c
                     tevent_debug.c memcache.c unix_match.c tfork.c''',
-                  deps='samba-util-core DYNCONFIG close-low-fd tini tiniparser genrand',
+                  deps='samba-util-core DYNCONFIG close-low-fd tini tiniparser genrand util_str_hex',
                   public_deps='talloc tevent execinfo pthread LIBCRYPTO charset util_setid systemd systemd-daemon',
                   public_headers='debug.h attr.h byteorder.h data_blob.h memory.h safe_string.h time.h talloc_stack.h string_wrappers.h idtree.h idtree_random.h blocking.h signal.h substitute.h fault.h genrand.h tfork.h',
                   header_path= [ ('dlinklist.h samba_util.h', '.'), ('*', 'util') ],
@@ -211,3 +211,8 @@ else:
                         source='util_str_escape.c',
                         deps='talloc',
                         local_include=False)
+
+    bld.SAMBA_SUBSYSTEM('util_str_hex',
+                        source='util_str_hex.c',
+                        deps='talloc',
+                        local_include=False)