Stop using xmlStringCurrentChar

xmlStringCurrentChar has been deprecated.

4 files changed, 80 insertions, 14 deletions

diff --git a/libxslt/numbers.c b/libxslt/numbers.c
index 1764abba..3cd881e3 100644
--- a/libxslt/numbers.c
+++ b/libxslt/numbers.c
@@ -353,8 +353,7 @@ xsltNumberFormatTokenize(const xmlChar *format,
      * Insert initial non-alphanumeric token.
      * There is always such a token in the list, even if NULL
      */
-    while (!xsltIsLetterDigit(val = xmlStringCurrentChar(NULL, format+ix,
-                                                         &len))) {
+    while (!xsltIsLetterDigit(val = xsltGetUTF8CharZ(format+ix, &len))) {
 	if (format[ix] == 0)		/* if end of format string */
 	    break; /* while */
 	ix += len;
@@ -377,19 +376,19 @@ xsltNumberFormatTokenize(const xmlChar *format,
 	    tokens->end = NULL;
 	}
 
-	val = xmlStringCurrentChar(NULL, format+ix, &len);
+	val = xsltGetUTF8CharZ(format+ix, &len);
 	if (IS_DIGIT_ONE(val) ||
 		 IS_DIGIT_ZERO(val)) {
 	    tokens->tokens[tokens->nTokens].width = 1;
 	    while (IS_DIGIT_ZERO(val)) {
 		tokens->tokens[tokens->nTokens].width++;
 		ix += len;
-		val = xmlStringCurrentChar(NULL, format+ix, &len);
+		val = xsltGetUTF8CharZ(format+ix, &len);
 	    }
 	    if (IS_DIGIT_ONE(val)) {
 		tokens->tokens[tokens->nTokens].token = val - 1;
 		ix += len;
-		val = xmlStringCurrentChar(NULL, format+ix, &len);
+		val = xsltGetUTF8CharZ(format+ix, &len);
 	    } else {
                 tokens->tokens[tokens->nTokens].token = '0';
                 tokens->tokens[tokens->nTokens].width = 1;
@@ -400,7 +399,7 @@ xsltNumberFormatTokenize(const xmlChar *format,
 		    (val == 'i') ) {
 	    tokens->tokens[tokens->nTokens].token = val;
 	    ix += len;
-	    val = xmlStringCurrentChar(NULL, format+ix, &len);
+	    val = xsltGetUTF8CharZ(format+ix, &len);
 	} else {
 	    /* XSLT section 7.7
 	     * "Any other format token indicates a numbering sequence
@@ -422,7 +421,7 @@ xsltNumberFormatTokenize(const xmlChar *format,
 	 */
 	while (xsltIsLetterDigit(val)) {
 	    ix += len;
-	    val = xmlStringCurrentChar(NULL, format+ix, &len);
+	    val = xsltGetUTF8CharZ(format+ix, &len);
 	}
 
 	/*
@@ -433,7 +432,7 @@ xsltNumberFormatTokenize(const xmlChar *format,
 	    if (val == 0)
 		break; /* while */
 	    ix += len;
-	    val = xmlStringCurrentChar(NULL, format+ix, &len);
+	    val = xsltGetUTF8CharZ(format+ix, &len);
 	}
 	if (ix > j)
 	    tokens->end = xmlStrndup(&format[j], ix - j);
diff --git a/libxslt/pattern.c b/libxslt/pattern.c
index 6e99b124..581f1791 100644
--- a/libxslt/pattern.c
+++ b/libxslt/pattern.c
@@ -1226,10 +1226,10 @@ xsltScanLiteral(xsltParserContextPtr ctxt) {
     if (CUR == '"') {
         NEXT;
 	cur = q = CUR_PTR;
-	val = xmlStringCurrentChar(NULL, cur, &len);
+	val = xsltGetUTF8CharZ(cur, &len);
 	while ((xmlIsCharQ(val)) && (val != '"')) {
 	    cur += len;
-	    val = xmlStringCurrentChar(NULL, cur, &len);
+	    val = xsltGetUTF8CharZ(cur, &len);
 	}
 	if (!xmlIsCharQ(val)) {
 	    ctxt->error = 1;
@@ -1242,10 +1242,10 @@ xsltScanLiteral(xsltParserContextPtr ctxt) {
     } else if (CUR == '\'') {
         NEXT;
 	cur = q = CUR_PTR;
-	val = xmlStringCurrentChar(NULL, cur, &len);
+	val = xsltGetUTF8CharZ(cur, &len);
 	while ((xmlIsCharQ(val)) && (val != '\'')) {
 	    cur += len;
-	    val = xmlStringCurrentChar(NULL, cur, &len);
+	    val = xsltGetUTF8CharZ(cur, &len);
 	}
 	if (!xmlIsCharQ(val)) {
 	    ctxt->error = 1;
@@ -1280,7 +1280,7 @@ xsltScanNCName(xsltParserContextPtr ctxt) {
     SKIP_BLANKS;
 
     cur = q = CUR_PTR;
-    val = xmlStringCurrentChar(NULL, cur, &len);
+    val = xsltGetUTF8CharZ(cur, &len);
     if (!xmlIsBaseCharQ(val) && !xmlIsIdeographicQ(val) && (val != '_'))
 	return(NULL);
 
@@ -1291,7 +1291,7 @@ xsltScanNCName(xsltParserContextPtr ctxt) {
 	   xmlIsCombiningQ(val) ||
 	   xmlIsExtenderQ(val)) {
 	cur += len;
-	val = xmlStringCurrentChar(NULL, cur, &len);
+	val = xsltGetUTF8CharZ(cur, &len);
     }
     ret = xmlStrndup(q, cur - q);
     CUR_PTR = cur;
diff --git a/libxslt/xsltutils.c b/libxslt/xsltutils.c
index 1d3e82e5..3705d28f 100644
--- a/libxslt/xsltutils.c
+++ b/libxslt/xsltutils.c
@@ -305,6 +305,66 @@ error:
     return(-1);
 }
 
+/**
+ * xsltGetUTF8CharZ:
+ * @utf:  a sequence of UTF-8 encoded bytes
+ * @len:  a pointer to @bytes len
+ *
+ * Read one UTF8 Char from a null-terminated string.
+ *
+ * Returns the char value or -1 in case of error and update @len with the
+ *        number of bytes used
+ */
+int
+xsltGetUTF8CharZ(const unsigned char *utf, int *len) {
+    unsigned int c;
+
+    if (utf == NULL)
+	goto error;
+    if (len == NULL)
+	goto error;
+
+    c = utf[0];
+    if (c & 0x80) {
+	if ((utf[1] & 0xc0) != 0x80)
+	    goto error;
+	if ((c & 0xe0) == 0xe0) {
+	    if ((utf[2] & 0xc0) != 0x80)
+		goto error;
+	    if ((c & 0xf0) == 0xf0) {
+		if ((c & 0xf8) != 0xf0 || (utf[3] & 0xc0) != 0x80)
+		    goto error;
+		*len = 4;
+		/* 4-byte code */
+		c = (utf[0] & 0x7) << 18;
+		c |= (utf[1] & 0x3f) << 12;
+		c |= (utf[2] & 0x3f) << 6;
+		c |= utf[3] & 0x3f;
+	    } else {
+	      /* 3-byte code */
+		*len = 3;
+		c = (utf[0] & 0xf) << 12;
+		c |= (utf[1] & 0x3f) << 6;
+		c |= utf[2] & 0x3f;
+	    }
+	} else {
+	  /* 2-byte code */
+	    *len = 2;
+	    c = (utf[0] & 0x1f) << 6;
+	    c |= utf[1] & 0x3f;
+	}
+    } else {
+	/* 1-byte code */
+	*len = 1;
+    }
+    return(c);
+
+error:
+    if (len != NULL)
+	*len = 0;
+    return(-1);
+}
+
 #ifdef XSLT_REFACTORED
 
 /**
diff --git a/libxslt/xsltutils.h b/libxslt/xsltutils.h
index 0c5523ee..2514774b 100644
--- a/libxslt/xsltutils.h
+++ b/libxslt/xsltutils.h
@@ -91,6 +91,13 @@ XSLTPUBFUN const xmlChar * XSLTCALL
 XSLTPUBFUN int XSLTCALL
 		xsltGetUTF8Char	(const unsigned char *utf,
 				 int *len);
+#ifdef IN_LIBXSLT
+/** DOC_DISABLE */
+XSLTPUBFUN int XSLTCALL
+		xsltGetUTF8CharZ (const unsigned char *utf,
+				  int *len);
+/** DOC_ENABLE */
+#endif
 
 /*
  * XSLT Debug Tracing Tracing Types