[Libreoffice-commits] core.git: sax/source

[Caolán McNamara]

 sax/source/expatwrap/saxwriter.cxx |   22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

New commits:
commit 167fb3f7800e342b2009b98579a1d42d528b9b5a
Author: Caolán McNamara <caolanm at redhat.com>
Date:   Thu Mar 15 16:17:51 2018 +0000

    crashtesting: asserts on legal xml names on export
    
    e.g. ooo82358-1.odt which has Chinese characters used in tags inside
    a parent xforms:instance tag
    
    Change-Id: If2edf9cc13e5a4cc969f5a46618a6534c52f2877
    Reviewed-on: https://gerrit.libreoffice.org/51352
    Reviewed-by: Michael Stahl <mstahl at redhat.com>
    Tested-by: Michael Stahl <mstahl at redhat.com>
    Tested-by: Jenkins <ci at libreoffice.org>

diff --git a/sax/source/expatwrap/saxwriter.cxx b/sax/source/expatwrap/saxwriter.cxx
index b7de667bcf05..c9d0a9b48cce 100644
--- a/sax/source/expatwrap/saxwriter.cxx
+++ b/sax/source/expatwrap/saxwriter.cxx
@@ -566,6 +566,13 @@ inline void SaxWriterHelper::startDocument()
         nCurrentPos = writeSequence();
 }
 
+#ifndef NDEBUG
+bool inrange(sal_Unicode c, sal_Unicode start, sal_Unicode end)
+{
+    return c >= start && c <= end;
+}
+#endif
+
 void CheckValidName(OUString const& rName)
 {
 #ifdef NDEBUG
@@ -583,10 +590,17 @@ void CheckValidName(OUString const& rName)
             else
                 hasColon = true;
         }
-        else if (!rtl::isAsciiAlphanumeric(c) && c != '_' && c != '-' && c != '.')
-        {   // note: this will also warn about non-ASCII characters which
-            // are allowed by XML but surely unexpected in LO filters
-            // (OTOH we don't warn about invalid start chars)
+        else if (!rtl::isAsciiAlphanumeric(c) && c != '_' && c != '-' && c != '.' &&
+                 !inrange(c, 0x00C0, 0x00D6) && !inrange(c, 0x00D8, 0x00F6) &&
+                 !inrange(c, 0x00F8, 0x02FF) && !inrange(c, 0x0370, 0x037D) &&
+                 !inrange(c, 0x037F, 0x1FFF) && !inrange(c, 0x200C, 0x200D) &&
+                 !inrange(c, 0x2070, 0x218F) && !inrange(c, 0x2C00, 0x2FEF) &&
+                 !inrange(c, 0x3001, 0xD7FF) && !inrange(c, 0xF900, 0xFDCF) &&
+                 !inrange(c, 0xFDF0, 0xFFFD) && c != 0x00B7 &&
+                 !inrange(c, 0x0300, 0x036F) && !inrange(c, 0x203F, 0x2040))
+        {
+            // https://www.w3.org/TR/xml11/#NT-NameChar
+            // (currently we don't warn about invalid start chars)
             assert(!"unexpected character in attribute name");
         }
     }