Merge from 3_0: Build a lookup table of valid DOS characters just

once, when init_iconv is first called.  Use this in check_dos_char
rather than doing the conversion every time.  Much faster.
(This used to be commit e8f1399323745a93ba9df97553effcf2d7629850)

2 files changed, 59 insertions, 5 deletions

diff --git a/source3/lib/charcnv.c b/source3/lib/charcnv.c
index f025e93718..d727628e41 100644
--- a/source3/lib/charcnv.c
+++ b/source3/lib/charcnv.c
@@ -59,7 +59,7 @@ static const char *charset_name(charset_t ch)
 	return ret;
 }
 
-static void lazy_initialize_conv(void)
+void lazy_initialize_conv(void)
 {
 	static int initialized = False;
 
@@ -67,14 +67,16 @@ static void lazy_initialize_conv(void)
 		initialized = True;
 		load_case_tables();
 		init_iconv();
-		init_valid_table();
 	}
 }
 
 /**
- Initialize iconv conversion descriptors.
-**/
-
+ * Initialize iconv conversion descriptors.
+ *
+ * This is called the first time it is needed, and also called again
+ * every time the configuration is reloaded, because the charset or
+ * codepage might have changed.
+ **/
 void init_iconv(void)
 {
 	int c1, c2;
@@ -112,6 +114,10 @@ void init_iconv(void)
 	}
 
 	if (did_reload) {
+		/* XXX: Does this really get called every time the dos
+		 * codepage changes? */
+		/* XXX: Is the did_reload test too strict? */
+		init_doschar_table();
 		init_valid_table();
 	}
 }
diff --git a/source3/lib/util_unistr.c b/source3/lib/util_unistr.c
index c666155f9f..812859000a 100644
--- a/source3/lib/util_unistr.c
+++ b/source3/lib/util_unistr.c
@@ -31,6 +31,14 @@ static smb_ucs2_t *upcase_table;
 static smb_ucs2_t *lowcase_table;
 static uint8 *valid_table;
 
+/**
+ * This table says which Unicode characters are valid dos
+ * characters.
+ *
+ * Each value is just a single bit.
+ **/
+static uint8 doschar_table[8192]; /* 65536 characters / 8 bits/byte */
+
 
 /*******************************************************************
 load the case handling tables
@@ -85,6 +93,21 @@ void load_case_tables(void)
 */
 static int check_dos_char(smb_ucs2_t c)
 {
+	static int initialized = False;
+
+	if (!initialized) {
+		initialized = True;
+		init_doschar_table();
+	}
+	
+	/* Find the right byte, and right bit within the byte; return
+	 * 1 or 0 */
+	return (doschar_table[(c & 0xffff) / 8] & (1 << (c & 7))) != 0;
+}
+
+
+static int check_dos_char_slowly(smb_ucs2_t c)
+{
 	char buf[10];
 	smb_ucs2_t c2 = 0;
 	int len1, len2;
@@ -95,6 +118,31 @@ static int check_dos_char(smb_ucs2_t c)
 	return (c == c2);
 }
 
+
+/**
+ * Fill out doschar table the hard way, by examining each character
+ **/
+void init_doschar_table(void)
+{
+	int i, j, byteval;
+
+	/* For each byte of packed table */
+	
+	for (i = 0; i <= 0xffff; i += 8) {
+		byteval = 0;
+		for (j = 0; j <= 7; j++) {
+			smb_ucs2_t c;
+
+			c = i + j;
+			
+			if (check_dos_char_slowly(c))
+				byteval |= 1 << j;
+		}
+		doschar_table[i/8] = byteval;
+	}
+}
+
+
 /**
  * Load the valid character map table from <tt>valid.dat</tt> or
  * create from the configured codepage.