Merge pull request #453 from libc/master

Add Nokogiri::XML::SAX::ParserContext#recovery
sparklemotion · Nov 13, 2013 · 4b7a7fb · 4b7a7fb
2 parents 68e1e98 + 804df68
commit 4b7a7fb
Show file tree

Hide file tree

Showing 4 changed files with 98 additions and 0 deletions.
diff --git a/ext/java/nokogiri/XmlSaxParserContext.java b/ext/java/nokogiri/XmlSaxParserContext.java
@@ -75,18 +75,22 @@ public class XmlSaxParserContext extends ParserContext {
         "http://xml.org/sax/features/namespace-prefixes";
     protected static final String FEATURE_LOAD_EXTERNAL_DTD =
         "http://apache.org/xml/features/nonvalidating/load-external-dtd";
+    protected static final String FEATURE_CONTINUE_AFTER_FATAL_ERROR =
+        "http://apache.org/xml/features/continue-after-fatal-error";
 
     protected AbstractSAXParser parser;
 
     protected NokogiriHandler handler = null;
     private IRubyObject replaceEntities;
+    private IRubyObject recovery;
 
     public XmlSaxParserContext(final Ruby ruby, RubyClass rubyClass) {
         super(ruby, rubyClass);
     }
 
     protected void initialize(Ruby runtime) {
         replaceEntities = runtime.getTrue();
+        recovery = runtime.getFalse();
         try {
             parser = createParser();
         } catch (SAXException se) {
@@ -199,6 +203,13 @@ protected void preParse(ThreadContext context,
                             IRubyObject handlerRuby,
                             NokogiriHandler handler) {
         ((XmlSaxParser) parser).setXmlDeclHandler(handler);
+        if(recovery.isTrue()) {
+          try {
+            ((XmlSaxParser) parser).setFeature(FEATURE_CONTINUE_AFTER_FATAL_ERROR, true);
+          } catch(Exception e) {
+            throw RaiseException.createNativeRaiseException(context.getRuntime(), e);
+          }
+        }
     }
 
     protected void postParse(ThreadContext context,
@@ -290,6 +301,28 @@ public IRubyObject get_replace_entities(ThreadContext context) {
         return replaceEntities;
     }
 
+    /**
+     * Can take a boolean assignment.
+     *
+     * @param context
+     * @param value
+     * @return
+     */
+    @JRubyMethod(name = "recovery=")
+    public IRubyObject set_recovery(ThreadContext context,
+                                            IRubyObject value) {
+        if (!value.isTrue()) recovery = context.getRuntime().getFalse();
+        else recovery = context.getRuntime().getTrue();
+
+        return this;
+    }
+
+    @JRubyMethod(name="recovery")
+    public IRubyObject get_recovery(ThreadContext context) {
+        return recovery;
+    }
+
+
 
     /**
      * If the handler's document is a FragmentHandler, attempt to trim

diff --git a/ext/nokogiri/xml_sax_parser_context.c b/ext/nokogiri/xml_sax_parser_context.c
@@ -201,6 +201,44 @@ static VALUE column(VALUE self)
   return Qnil;
 }
 
+/*
+ * call-seq:
+ *  recovery=(boolean)
+ *
+ * Should this parser recover from structural errors? It will not stop processing
+ * file on structural errors if if set to true
+ */
+static VALUE set_recovery(VALUE self, VALUE value)
+{
+  xmlParserCtxtPtr ctxt;
+  Data_Get_Struct(self, xmlParserCtxt, ctxt);
+
+  if(value == Qfalse)
+    ctxt->recovery = 0;
+  else
+    ctxt->recovery = 1;
+
+  return value;
+}
+
+/*
+ * call-seq:
+ *  recovery
+ *
+ * Should this parser recover from structural errors? It will not stop processing
+ * file on structural errors if if set to true
+ */
+static VALUE get_recovery(VALUE self)
+{
+  xmlParserCtxtPtr ctxt;
+  Data_Get_Struct(self, xmlParserCtxt, ctxt);
+
+  if(ctxt->recovery == 0)
+    return Qfalse;
+  else
+    return Qtrue;
+}
+
 void init_xml_sax_parser_context()
 {
   VALUE nokogiri  = rb_define_module("Nokogiri");
@@ -217,6 +255,8 @@ void init_xml_sax_parser_context()
   rb_define_method(klass, "parse_with", parse_with, 1);
   rb_define_method(klass, "replace_entities=", set_replace_entities, 1);
   rb_define_method(klass, "replace_entities", get_replace_entities, 0);
+  rb_define_method(klass, "recovery=", set_recovery, 1);
+  rb_define_method(klass, "recovery", get_recovery, 0);
   rb_define_method(klass, "line", line, 0);
   rb_define_method(klass, "column", column, 0);
 }
diff --git a/test/xml/sax/test_parser.rb b/test/xml/sax/test_parser.rb
@@ -360,6 +360,22 @@ def test_parser_attributes
 
           assert_equal [['root', []], ['foo', [['a', '&b'], ['c', '>d']]]], @parser.document.start_elements
         end
+
+        def test_recovery_from_incorrect_xml
+          xml = <<-eoxml
+<?xml version="1.0" ?><Root><Data><?xml version='1.0'?><Item>hey</Item></Data><Data><Item>hey yourself</Item></Data></Root>
+          eoxml
+
+          block_called = false
+          @parser.parse(xml) { |ctx|
+            block_called = true
+            ctx.recovery = true
+          }
+
+          assert block_called
+
+          assert_equal [['Root', []], ['Data', []], ['Item', []], ['Data', []], ['Item', []]], @parser.document.start_elements
+        end
       end
     end
   end

diff --git a/test/xml/sax/test_parser_context.rb b/test/xml/sax/test_parser_context.rb
@@ -65,6 +65,15 @@ def test_replace_entities
           assert_equal true, pc.replace_entities
         end
 
+        def test_recovery
+          pc = ParserContext.new StringIO.new('<root />'), 'UTF-8'
+          pc.recovery = false
+          assert_equal false, pc.recovery
+
+          pc.recovery = true
+          assert_equal true, pc.recovery
+        end
+
         def test_from_io
           ctx = ParserContext.new StringIO.new('fo'), 'UTF-8'
           assert ctx