fix multi-byte parsing

JuliaIO · Nov 10, 2016 · 564adfe · 564adfe
1 parent c1ce997
commit 564adfe
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 1 deletion.
diff --git a/src/document.jl b/src/document.jl
@@ -74,7 +74,7 @@ function parsexml(xmlstring::AbstractString)
         (:xmlParseMemory, libxml2),
         Ptr{_Node},
         (Cstring, Cint),
-        xmlstring, length(xmlstring))
+        xmlstring, sizeof(xmlstring))
     if ptr == C_NULL
         throw_xml_error()
     end

diff --git a/test/runtests.jl b/test/runtests.jl
@@ -87,6 +87,18 @@ end
         @test nodetype(parsexml("<xml/>".data).node) === EzXML.XML_DOCUMENT_NODE
         @test nodetype(parsexml("<html/>".data).node) === EzXML.XML_DOCUMENT_NODE
 
+        # This includes multi-byte characters.
+        doc = parse(Document, """
+        <?xml version="1.0" encoding="UTF-8" ?>
+        <Link>
+            <Name>pubmed_pubmed</Name>
+            <Menu>Similar articles</Menu>
+            <Description>... “linked from” ...</Description>
+            <DbTo>pubmed</DbTo>
+        </Link>
+        """)
+        @test nodetype(doc.node) === EzXML.XML_DOCUMENT_NODE
+
         @test_throws ArgumentError parse(Document, "")
         @test_throws XMLError parse(Document, " ")
         @test_throws XMLError parse(Document, "abracadabra")