From 8cc0d1f88966694e6562ee5bb2ca4406da11d64b Mon Sep 17 00:00:00 2001
From: Daniel Veillard <veillard@src.gnome.org>
Date: Mon, 16 Nov 1998 01:04:26 +0000
Subject: [PATCH] Better entities and char ref encoding, error msg formatting,
 Daniel.

---
 ChangeLog  |  5 +++++
 entities.c | 40 +++++++++++++++++++++++++++++++++++-----
 parser.c   | 15 +++++++++------
 3 files changed, 49 insertions(+), 11 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index e51225f3..15040f5b 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+Sun Nov 15 19:59:47 EST 1998 Daniel Veillard <Daniel.Veillard@w3.org>
+
+	* parser.c, entities.c: improve entities and char ref encoding,
+	  and cleanups of error messages.
+
 Fri Nov 13 13:03:10 EST 1998 Daniel Veillard <Daniel.Veillard@w3.org>
 
 	* parser.c, entities.c: simple bug hunting done during rpm2html and
diff --git a/entities.c b/entities.c
index 0b418001..28a35826 100644
--- a/entities.c
+++ b/entities.c
@@ -248,6 +248,15 @@ xmlEntityPtr xmlGetDocEntity(xmlDocPtr doc, const CHAR *name) {
     return(NULL);
 }
 
+/*
+ * [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD]
+ *                  | [#x10000-#x10FFFF]
+ * any Unicode character, excluding the surrogate blocks, FFFE, and FFFF.
+ */
+#define IS_CHAR(c)							\
+    (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) ||			\
+     (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF)))
+
 /*
  * xmlEncodeEntities : do a global encoding of a string, replacing the
  *                     predefined entities and non ASCII values with their
@@ -255,6 +264,7 @@ xmlEntityPtr xmlGetDocEntity(xmlDocPtr doc, const CHAR *name) {
  * TODO !!!! Once moved to UTF-8 internal encoding, the encoding of non-ascii
  *           get erroneous.
  */
+
 CHAR *xmlEncodeEntities(xmlDocPtr doc, const CHAR *input) {
     const CHAR *cur = input;
     CHAR *out = buffer;
@@ -310,6 +320,12 @@ CHAR *xmlEncodeEntities(xmlDocPtr doc, const CHAR *input) {
 	    *out++ = 'o';
 	    *out++ = 's';
 	    *out++ = ';';
+	} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
+	    (*cur == '\n') || (*cur == '\r') || (*cur == '\t')) {
+	    /*
+	     * default case, just copy !
+	     */
+	    *out++ = *cur;
 #ifndef USE_UTF_8
 	} else if ((sizeof(CHAR) == 1) && (*cur >= 0x80)) {
 	    char buf[10], *ptr;
@@ -321,12 +337,26 @@ CHAR *xmlEncodeEntities(xmlDocPtr doc, const CHAR *input) {
             ptr = buf;
 	    while (*ptr != 0) *out++ = *ptr++;
 #endif
-	} else {
-	    /*
-	     * default case, just copy !
-	     */
-	    *out++ = *cur;
+	} else if (IS_CHAR(*cur)) {
+	    char buf[10], *ptr;
+
+#ifdef HAVE_SNPRINTF
+	    snprintf(buf, 9, "&#%d;", *cur);
+#else
+	    sprintf(buf, "&#%d;", *cur);
+#endif
+            ptr = buf;
+	    while (*ptr != 0) *out++ = *ptr++;
 	}
+#if 0
+	else {
+	    /*
+	     * default case, this is not a valid char !
+	     * Skip it...
+	     */
+	    fprintf(stderr, "xmlEncodeEntities: invalid char %d\n", (int) *cur);
+	}
+#endif
 	cur++;
     }
     *out++ = 0;
diff --git a/parser.c b/parser.c
index 2b1b0081..5464b263 100644
--- a/parser.c
+++ b/parser.c
@@ -1080,7 +1080,7 @@ xmlParseQuotedString(xmlParserCtxtPtr ctxt) {
 	while (IS_CHAR(CUR) && (CUR != '"')) NEXT;
 	if (CUR != '"') {
 	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-	        ctxt->sax->error(ctxt, "String not closed\"%.50s\n", q);
+	        ctxt->sax->error(ctxt, "String not closed \"%.50s\"\n", q);
         } else {
             ret = xmlStrndup(q, CUR_PTR - q);
 	    NEXT;
@@ -1091,7 +1091,7 @@ xmlParseQuotedString(xmlParserCtxtPtr ctxt) {
 	while (IS_CHAR(CUR) && (CUR != '\'')) NEXT;
 	if (CUR != '\'') {
 	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-	        ctxt->sax->error(ctxt, "String not closed\"%.50s\n", q);
+	        ctxt->sax->error(ctxt, "String not closed \"%.50s\"\n", q);
         } else {
             ret = xmlStrndup(q, CUR_PTR - q);
 	    NEXT;
@@ -2253,7 +2253,8 @@ xmlParseCharRef(xmlParserCtxtPtr ctxt) {
 	        val = val * 16 + (CUR - 'A') + 10;
 	    else {
 	        if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-		    ctxt->sax->error(ctxt, "xmlParseCharRef: invalid value\n");
+		    ctxt->sax->error(ctxt, 
+		         "xmlParseCharRef: invalid hexa value\n");
 		val = 0;
 		break;
 	    }
@@ -2268,7 +2269,8 @@ xmlParseCharRef(xmlParserCtxtPtr ctxt) {
 	        val = val * 10 + (CUR - '0');
 	    else {
 	        if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-		    ctxt->sax->error(ctxt, "xmlParseCharRef: invalid value\n");
+		    ctxt->sax->error(ctxt, 
+		         "xmlParseCharRef: invalid decimal value\n");
 		val = 0;
 		break;
 	    }
@@ -2289,7 +2291,8 @@ xmlParseCharRef(xmlParserCtxtPtr ctxt) {
 	return(xmlStrndup(buf, 1));
     } else {
 	if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-	    ctxt->sax->error(ctxt, "xmlParseCharRef: invalid value");
+	    ctxt->sax->error(ctxt, "xmlParseCharRef: invalid CHAR value %d\n",
+	                     val);
     }
     return(NULL);
 }
@@ -3932,7 +3935,7 @@ xmlParserAddNodeInfo(xmlParserCtxtPtr ctxt,
 
       if ( tmp_buffer == NULL ) {
         if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
-	    ctxt->sax->error(ctxt, "Out of memory");
+	    ctxt->sax->error(ctxt, "Out of memory\n");
         return;
       }
       ctxt->node_seq.buffer = tmp_buffer;