- parser.c: fixed UTF8 BOM support in push mode

- test/utf8bom.xml result/utf8bom.xml result/noent/utf8bom.xml: added a specific testcase Daniel
ConradIrwin · Jun 28, 2001 · 0e4cd17 · 0e4cd17
1 parent 3e5bb8e
commit 0e4cd17
Show file tree

Hide file tree

Showing 5 changed files with 40 additions and 8 deletions.
diff --git a/ChangeLog b/ChangeLog
@@ -1,3 +1,9 @@
+Thu Jun 28 14:11:28 CEST 2001 Daniel Veillard <Daniel.Veillard@imag.fr>
+
+	* parser.c: fixed UTF8 BOM support in push mode
+	* test/utf8bom.xml result/utf8bom.xml result/noent/utf8bom.xml:
+	  added a specific testcase
+
 Wed Jun 27 18:33:13 CEST 2001 Daniel Veillard <Daniel.Veillard@imag.fr>
 
 	* Makefile.am: added --push regression tests

diff --git a/parser.c b/parser.c
@@ -7632,11 +7632,31 @@ xmlParseTryOrFinish(xmlParserCtxtPtr ctxt, int terminate) {
 		 */
 	        goto done;
             case XML_PARSER_START:
-	        /*
-		 * Very first chars read from the document flow.
-		 */
-		if (avail < 2)
-		    goto done;
+		if (ctxt->charset == XML_CHAR_ENCODING_NONE) {
+		    xmlChar start[4];
+		    xmlCharEncoding enc;
+
+		    /*
+		     * Very first chars read from the document flow.
+		     */
+		    if (avail < 4)
+			goto done;
+
+		    /* 
+		     * Get the 4 first bytes and decode the charset
+		     * if enc != XML_CHAR_ENCODING_NONE
+		     * plug some encoding conversion routines.
+		     */
+		    start[0] = RAW;
+		    start[1] = NXT(1);
+		    start[2] = NXT(2);
+		    start[3] = NXT(3);
+		    enc = xmlDetectCharEncoding(start, 4);
+		    if (enc != XML_CHAR_ENCODING_NONE) {
+			xmlSwitchEncoding(ctxt, enc);
+		    }
+		    break;
+		}
 
 		cur = ctxt->input->cur[0];
 		next = ctxt->input->cur[1];
@@ -8509,9 +8529,6 @@ xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax, void *user_data,
     inputStream->cur = inputStream->buf->buffer->content;
     inputStream->end = 
 	&inputStream->buf->buffer->content[inputStream->buf->buffer->use];
-    if (enc != XML_CHAR_ENCODING_NONE) {
-        xmlSwitchEncoding(ctxt, enc);
-    }
 
     inputPush(ctxt, inputStream);
 
@@ -8523,6 +8540,10 @@ xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax, void *user_data,
 #endif
     }
 
+    if (enc != XML_CHAR_ENCODING_NONE) {
+        xmlSwitchEncoding(ctxt, enc);
+    }
+
     return(ctxt);
 }
 

diff --git a/result/noent/utf8bom.xml b/result/noent/utf8bom.xml
@@ -0,0 +1,2 @@
+<?xml version="1.0"?>
+<foo/>
diff --git a/result/utf8bom.xml b/result/utf8bom.xml
@@ -0,0 +1,2 @@
+<?xml version="1.0"?>
+<foo/>
diff --git a/test/utf8bom.xml b/test/utf8bom.xml
@@ -0,0 +1 @@
+<foo/>