More efficient/robust implementation of function TclNativeCreateNativeRep(). rfe_3389978

- No more intermediate results in a Tcl_DString, just allocate space directly. 
- Let MultiByteToWideChar() do all the difficult work, inclusive checking for invalid byte sequences.
- Handled extended win32 paths, inclusive UNC paths.
Implementation for a great deal taken over from fossil.

1 files changed, 63 insertions, 46 deletions

diff --git a/win/tclWinFile.c b/win/tclWinFile.c
index e32cd94..5761eeb 100644
--- a/win/tclWinFile.c
+++ b/win/tclWinFile.c
@@ -2897,10 +2897,10 @@ ClientData
 TclNativeCreateNativeRep(
     Tcl_Obj *pathPtr)
 {
-    char *nativePathPtr, *str;
-    Tcl_DString ds;
+    WCHAR *nativePathPtr;
+    const char *str;
     Tcl_Obj *validPathPtr;
-    int len, i = 2;
+    int len;
     WCHAR *wp;
 
     if (TclFSCwdIsNative()) {
@@ -2927,55 +2927,72 @@ TclNativeCreateNativeRep(
     }
 
     str = Tcl_GetStringFromObj(validPathPtr, &len);
-    Tcl_WinUtfToTChar(str, len, &ds);
-    len = Tcl_DStringLength(&ds) + sizeof(WCHAR);
-    wp = (WCHAR *) Tcl_DStringValue(&ds);
-    i=sizeof(WCHAR);
-    if ((wp[0]=='/'||wp[0]=='\\') && (wp[1]=='/'||wp[1]=='\\')) {
-	if (wp[2]=='?'){
-	    /* Extended path prefix: convert slashes but not the '?' */
-	    wp[0] = wp[1] = wp[3] = '\\';
-	    i += 8; wp+=4;
-	    if (((wp[0]>='A'&&wp[0]<='Z') || (wp[0]>='a'&&wp[0]<='z'))
-		    && (wp[1]==':') && (wp[2]=='/' || wp[2]=='\\')) {
-		/* With drive, don't convert the ':' */
-		i += 4; wp+=2;
-	    }
-	}
-    } else {
-	if (((wp[0]>='A'&&wp[0]<='Z') || (wp[0]>='a'&&wp[0]<='z'))
-		&& (wp[1]==':') && (wp[2]=='/' || wp[2]=='\\')) {
-	    /* With drive, don't convert the ':' */
-	    i += 4; wp+=2;
-	    if (len > (MAX_PATH * sizeof(WCHAR))){
-		/* Path is too long, needs an extended path prefix. */
-		Tcl_DStringSetLength(&ds, len+=8);
-		Tcl_DStringSetLength(&ds, len+1); /* Must end with two NUL bytes */
-		wp = (WCHAR *) Tcl_DStringValue(&ds); /* wp might be re-allocated */
-		memmove(wp+4, wp, len-8);
-		memcpy(wp, L"\\\\?\\", 8);
-		i+=12; wp += 6;
-	    }
+
+    if (strlen(str)!=len) {
+	/* String contains NUL-bytes. This is invalid. */
+	return 0;
+    }
+    /* Let MultiByteToWideChar check for other invalid sequences, like
+     * 0xC0 0x80 (== overlong NUL). See bug [3118489]: NUL in filenames */
+    len = MultiByteToWideChar(CP_UTF8, MB_ERR_INVALID_CHARS, str, -1, 0, 0);
+    if (len==0) {
+	return 0;
+    }
+    /* Overallocate 6 chars, making some room for extended paths */
+    wp = nativePathPtr = ckalloc( (len+6) * sizeof(WCHAR) );
+    if (nativePathPtr==0) {
+      return 0;
+    }
+    MultiByteToWideChar(CP_UTF8, MB_ERR_INVALID_CHARS, str, -1, nativePathPtr, len);
+    /*
+    ** If path starts with "//?/" or "\\?\" (extended path), translate
+    ** any slashes to backslashes but leave the '?' intact
+    */
+    if ((str[0]=='\\' || str[0]=='/') && (str[1]=='\\' || str[1]=='/')
+	    && str[2]=='?' && (str[3]=='\\' || str[3]=='/')) {
+	wp[0] = wp[1] = wp[3] = '\\';
+	str += 4;
+	wp += 4;
+    }
+    /*
+    ** If there is no "\\?\" prefix but there is a drive or UNC
+    ** path prefix and the path is larger than MAX_PATH chars,
+    ** no Win32 API function can handle that unless it is
+    ** prefixed with the extended path prefix. See:
+    ** <http://msdn.microsoft.com/en-us/library/aa365247(VS.85).aspx#maxpath>
+    **/
+    if (((str[0]>='A'&&str[0]<='Z') || (str[0]>='a'&&str[0]<='z'))
+	    && str[1]==':' && (str[2]=='\\' || str[2]=='/')) {
+	if (wp==nativePathPtr && len>MAX_PATH) {
+	    memmove(wp+4, wp, len*sizeof(WCHAR));
+	    memcpy(wp, L"\\\\?\\", 4*sizeof(WCHAR));
+	    wp += 4;
 	}
+	/*
+	 ** If (remainder of) path starts with "<drive>:/" or "<drive>:\",
+	 ** leave the ':' intact but translate the backslash to a slash.
+	 */
+	wp[2] = '\\';
+	wp += 3;
+    } else if (wp==nativePathPtr && len>MAX_PATH
+	    && (str[0]=='\\' || str[0]=='/')
+	    && (str[1]=='\\' || str[1]=='/') && str[2]!='?') {
+	memmove(wp+6, wp, len*sizeof(WCHAR));
+	memcpy(wp, L"\\\\?\\UNC", 7*sizeof(WCHAR));
+	wp += 7;
     }
-    for (; i<len; ++wp,i+=sizeof(WCHAR)) {
-	if ( (*wp < ' ') || wcschr(L"\"*:<>?|", *wp) ){
-	    if (!*wp){
-		/* See bug [3118489]: NUL in filenames */
-		Tcl_DecrRefCount(validPathPtr);
-		Tcl_DStringFree(&ds);
-		return NULL;
-	    }
+    /*
+    ** In the remainder of the path, translate invalid characters to
+    ** characters in the Unicode private use area.
+    */
+    while (*wp != '\0') {
+	if ((*wp < ' ') || wcschr(L"\"*:<>?|", *wp)) {
 	    *wp |= 0xF000;
-	}else if (*wp=='/') {
+	} else if (*wp == '/') {
 	    *wp = '\\';
 	}
+	++wp;
     }
-    Tcl_DecrRefCount(validPathPtr);
-    nativePathPtr = ckalloc(len);
-    memcpy(nativePathPtr, Tcl_DStringValue(&ds), (size_t) len);
-
-    Tcl_DStringFree(&ds);
     return nativePathPtr;
 }