From 43b511fa714df875dc4f40d108061eede0d4d76b Mon Sep 17 00:00:00 2001
From: Nick Wellnhofer <wellnhofer@aevum.de>
Date: Sun, 26 Nov 2023 14:31:39 +0100
Subject: [PATCH] parser: Make CRLF increment line number

Partial revert of cb927e85 fixing CRLFs not incrementing the line
number.

This requires to rework xmlParseQNameHashed. The original implementation
prompted the change to xmlCurrentChar which really shouldn't modify the
'cur' pointer as side effect. But the NEXTL macro relies on this
behavior.

Ultimately, we should reintroduce the change to xmlCurrentChar and fix
the NEXTL macro. This will lead to single CRs incrementing the line
number as well which seems more consistent.

Fixes #628.
---
 parser.c                    | 15 +++++++++------
 parserInternals.c           |  9 ++++++++-
 result/errors/name3.xml.ent |  3 +++
 result/errors/name3.xml.err |  3 +++
 result/errors/name3.xml.str |  4 ++++
 test/errors/name3.xml       |  1 +
 6 files changed, 28 insertions(+), 7 deletions(-)
 create mode 100644 result/errors/name3.xml.ent
 create mode 100644 result/errors/name3.xml.err
 create mode 100644 result/errors/name3.xml.str
 create mode 100644 test/errors/name3.xml

diff --git a/parser.c b/parser.c
index 6b0eb421..4843a728 100644
--- a/parser.c
+++ b/parser.c
@@ -8949,7 +8949,7 @@ xmlParseEndTag(xmlParserCtxtPtr ctxt) {
 static xmlHashedString
 xmlParseQNameHashed(xmlParserCtxtPtr ctxt, xmlHashedString *prefix) {
     xmlHashedString l, p;
-    int start;
+    int start, isNCName = 0;
 
     l.name = NULL;
     p.name = NULL;
@@ -8960,10 +8960,13 @@ xmlParseQNameHashed(xmlParserCtxtPtr ctxt, xmlHashedString *prefix) {
     start = CUR_PTR - BASE_PTR;
 
     l = xmlParseNCName(ctxt);
-    if ((l.name != NULL) && (CUR == ':')) {
-        NEXT;
-	p = l;
-	l = xmlParseNCName(ctxt);
+    if (l.name != NULL) {
+        isNCName = 1;
+        if (CUR == ':') {
+            NEXT;
+            p = l;
+            l = xmlParseNCName(ctxt);
+        }
     }
     if ((l.name == NULL) || (CUR == ':')) {
         xmlChar *tmp;
@@ -8972,7 +8975,7 @@ xmlParseQNameHashed(xmlParserCtxtPtr ctxt, xmlHashedString *prefix) {
         p.name = NULL;
         if (ctxt->instate == XML_PARSER_EOF)
             return(l);
-        if ((CUR != ':') && (CUR_PTR <= BASE_PTR + start))
+        if ((isNCName == 0) && (CUR != ':'))
             return(l);
         tmp = xmlParseNmtoken(ctxt);
         if (tmp != NULL)
diff --git a/parserInternals.c b/parserInternals.c
index 29deebe4..1eff6c71 100644
--- a/parserInternals.c
+++ b/parserInternals.c
@@ -849,7 +849,14 @@ xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
              *   the single character #xA.
              */
             if (c == '\r') {
-                *len = ((cur[1] == '\n') ? 2 : 1);
+                /*
+                 * TODO: This function shouldn't change the 'cur' pointer
+                 * as side effect, but the NEXTL macro in parser.c relies
+                 * on this behavior when incrementing line numbers.
+                 */
+                if (cur[1] == '\n')
+                    ctxt->input->cur++;
+                *len = 1;
                 c = '\n';
             } else if (c == 0) {
                 if (ctxt->input->cur >= ctxt->input->end) {
diff --git a/result/errors/name3.xml.ent b/result/errors/name3.xml.ent
new file mode 100644
index 00000000..3925f9da
--- /dev/null
+++ b/result/errors/name3.xml.ent
@@ -0,0 +1,3 @@
+./test/errors/name3.xml:1: parser error : StartTag: invalid element name
+<.name/>
+ ^
diff --git a/result/errors/name3.xml.err b/result/errors/name3.xml.err
new file mode 100644
index 00000000..3925f9da
--- /dev/null
+++ b/result/errors/name3.xml.err
@@ -0,0 +1,3 @@
+./test/errors/name3.xml:1: parser error : StartTag: invalid element name
+<.name/>
+ ^
diff --git a/result/errors/name3.xml.str b/result/errors/name3.xml.str
new file mode 100644
index 00000000..460b31a3
--- /dev/null
+++ b/result/errors/name3.xml.str
@@ -0,0 +1,4 @@
+./test/errors/name3.xml:1: parser error : StartTag: invalid element name
+<.name/>
+ ^
+./test/errors/name3.xml : failed to parse
diff --git a/test/errors/name3.xml b/test/errors/name3.xml
new file mode 100644
index 00000000..2f5f96e7
--- /dev/null
+++ b/test/errors/name3.xml
@@ -0,0 +1 @@
+<.name/>