nokogiri 1.6.7.2-java → 1.6.8-java

data/bin/nokogiri

@@ -41,9 +41,9 @@ opts = OptionParser.new do |opts|
   opts.define_head "Usage: nokogiri <uri|path> [options]"
   opts.separator ""
   opts.separator "Examples:"
-  opts.separator "  nokogiri http://www.ruby-lang.org/"
+  opts.separator "  nokogiri https://www.ruby-lang.org/"
   opts.separator "  nokogiri ./public/index.html"
-  opts.separator "  curl -s http://nokogiri.org | nokogiri -e'p $_.css(\"h1\").length'"
+  opts.separator "  curl -s http://www.nokogiri.org | nokogiri -e'p $_.css(\"h1\").length'"
   opts.separator ""
   opts.separator "Options:"
 

data/dependencies.yml

@@ -1,4 +1,29 @@
-libxml2: "2.9.2"
-libxslt: "1.1.28"
-zlib: "1.2.8"
-libiconv: "1.14"
+libxml2:
+  version: "2.9.4"
+  md5: "ae249165c173b1ff386ee8ad676815f5" # manually confirmed via `gpg --verify`
+  # gpg: Signature made Mon 23 May 2016 04:02:13 AM EDT using DSA key ID DE95BC1F
+  # gpg: Good signature from "Daniel Veillard (Red Hat work email) <veillard@redhat.com>"
+  # gpg:                 aka "Daniel Veillard <Daniel.Veillard@w3.org>"
+  # gpg: WARNING: This key is not certified with a trusted signature!
+  # gpg:          There is no indication that the signature belongs to the owner.
+  # Primary key fingerprint: C744 15BA 7C9C 7F78 F02E  1DC3 4606 B8A5 DE95 BC1F
+
+libxslt:
+  version: "1.1.29"
+  md5: "a129d3c44c022de3b9dcf6d6f288d72e"
+  # gpg: Signature made Mon 23 May 2016 09:58:52 PM EDT using DSA key ID DE95BC1F
+  # gpg: Good signature from "Daniel Veillard (Red Hat work email) <veillard@redhat.com>"
+  # gpg:                 aka "Daniel Veillard <Daniel.Veillard@w3.org>"
+  # gpg: WARNING: This key is not certified with a trusted signature!
+  # gpg:          There is no indication that the signature belongs to the owner.
+  # Primary key fingerprint: C744 15BA 7C9C 7F78 F02E  1DC3 4606 B8A5 DE95 BC1F
+
+zlib:
+  version: "1.2.8"
+  md5: "44d667c142d7cda120332623eab69f40"
+
+libiconv:
+  version: "1.14"
+  md5: "e34509b1623cec449dfeb73d7ce9c6c6"
+  # gpg: Signature made Sun 07 Aug 2011 01:58:18 PM EDT using DSA key ID F059B1D1
+  # gpg: BAD signature from "Bruno Haible (Open Source Development) <bruno@clisp.org>"

data/ext/java/nokogiri/HtmlElementDescription.java

@@ -105,9 +105,12 @@ public class HtmlElementDescription extends RubyObject {
     public static IRubyObject get(ThreadContext context,
                                   IRubyObject klazz, IRubyObject name) {
 
-        HTMLElements.Element elem = HTMLElements.getElement(name.toString());
+        // nekohtml will return an element even for invalid names, see
+        // http://sourceforge.net/p/nekohtml/code/HEAD/tree/trunk/src/org/cyberneko/html/HTMLElements.java#l514
+        // which breaks `test_fetch_nonexistent'
+        HTMLElements.Element elem = HTMLElements.getElement(name.asJavaString(), HTMLElements.NO_SUCH_ELEMENT);
         if (elem == HTMLElements.NO_SUCH_ELEMENT)
-            return context.getRuntime().getNil();
+            return context.nil;
 
         HtmlElementDescription desc =
             new HtmlElementDescription(context.getRuntime(), (RubyClass)klazz);

data/ext/java/nokogiri/NokogiriService.java

@@ -36,6 +36,7 @@ import java.util.Collections;
 import java.util.HashMap;
 import java.util.Map;
 
+import org.cyberneko.html.HTMLElements;
 import org.jruby.Ruby;
 import org.jruby.RubyArray;
 import org.jruby.RubyClass;
@@ -53,6 +54,24 @@ import org.jruby.runtime.load.BasicLibraryService;
  * @author Yoko Harada <yokolet@gmail.com>
  */
 public class NokogiriService implements BasicLibraryService {
+
+    // nekohtml from version 1.9.13 they autocomplete tbody around
+    // tr tags of a table - http://sourceforge.net/p/nekohtml/code/241/
+    // this monkey patch undoes this autocompletion
+    static class MonkeyPatchHTMLElements extends HTMLElements {
+        static void patchIt() {
+            Element[] array = ELEMENTS_ARRAY['T'-'A'];
+            for(int i = 0; i < array.length; i++) {
+                if (array[i].name.equals("TR")) {
+                    array[i] = new Element(TR, "TR", Element.BLOCK, TABLE, new short[]{TD,TH,TR,COLGROUP,DIV});
+                }
+            }
+        }
+    }
+    static {
+        MonkeyPatchHTMLElements.patchIt();
+    }
+
     public static final String nokogiriClassCacheGvarName = "$NOKOGIRI_CLASS_CACHE";
 
     public boolean basicLoad(Ruby ruby) {

data/ext/java/nokogiri/XmlAttr.java

@@ -164,7 +164,9 @@ public class XmlAttr extends XmlNode{
     protected IRubyObject getNodeName(ThreadContext context) {
         if (name != null) return name;
         String attrName = ((Attr)node).getName();
-        if (!(doc instanceof HtmlDocument)) attrName = NokogiriHelpers.getLocalPart(attrName);
+        if (!(doc instanceof HtmlDocument) && node.getNamespaceURI() != null) {
+            attrName = NokogiriHelpers.getLocalPart(attrName);
+        }
         return attrName == null ? context.getRuntime().getNil() : RubyString.newString(context.getRuntime(), attrName);
     }
 

data/ext/java/nokogiri/XmlDocumentFragment.java

@@ -54,7 +54,6 @@ import org.jruby.javasupport.util.RuntimeHelpers;
 import org.jruby.runtime.ThreadContext;
 import org.jruby.runtime.builtin.IRubyObject;
 import org.w3c.dom.Attr;
-import org.w3c.dom.Document;
 import org.w3c.dom.NamedNodeMap;
 
 /**
@@ -90,7 +89,6 @@ public class XmlDocumentFragment extends XmlNode {
         
         // make wellformed fragment, ignore invalid namespace, or add appropriate namespace to parse
         if (args.length > 1 && args[1] instanceof RubyString) {
-            args[1] = trim(context, doc, (RubyString)args[1]);
             if (XmlDocumentFragment.isTag((RubyString)args[1])) {
                 args[1] = RubyString.newString(context.getRuntime(), addNamespaceDeclIfNeeded(doc, rubyStringToString(args[1])));
             }
@@ -107,18 +105,6 @@ public class XmlDocumentFragment extends XmlNode {
         RuntimeHelpers.invoke(context, fragment, "initialize", args);
         return fragment;
     }
-    
-    private static IRubyObject trim(ThreadContext context, XmlDocument xmlDocument, RubyString str) {
-        // checks whether schema is given. if exists, allows whitespace processing to a parser
-        Document document = (Document)xmlDocument.node;
-        if (document.getDoctype() != null) return str;
-        // strips trailing \n off forcefully
-        // not to return new object in case of no chomp needed, chomp! is used here.
-        IRubyObject result;
-        if (context.getRuntime().is1_9()) result = str.chomp_bang19(context);
-        else result = str.chomp_bang(context);
-        return result.isNil() ? str : result;
-    }
 
     private static boolean isTag(RubyString ruby_string) {
         String str = rubyStringToString(ruby_string);

data/ext/java/nokogiri/XmlNode.java

@@ -333,15 +333,11 @@ public class XmlNode extends RubyObject {
         Element element = null;
         String node_name = rubyStringToString(name);
         String prefix = NokogiriHelpers.getPrefix(node_name);
-        if (prefix == null) {
-            element = document.createElement(node_name);
-        } else {
-            String namespace_uri = null;
-            if (document.getDocumentElement() != null) {
-                namespace_uri = document.getDocumentElement().lookupNamespaceURI(prefix);
-            }
-            element = document.createElementNS(namespace_uri, node_name);
+        String namespace_uri = null;
+        if (document.getDocumentElement() != null) {
+            namespace_uri = document.getDocumentElement().lookupNamespaceURI(prefix);
         }
+        element = document.createElementNS(namespace_uri, node_name);
         setNode(context, element);
     }
 
@@ -454,57 +450,77 @@ public class XmlNode extends RubyObject {
     public void post_add_child(ThreadContext context, XmlNode current, XmlNode child) {
     }
 
+    /**
+     * This method should be called after a node has been adopted in a new
+     * document. This method will ensure that the node is renamed with the
+     * appriopriate NS uri. First the prefix of the node is extracted, then is
+     * used to lookup the namespace uri in the new document starting at the
+     * current node and traversing the ancestors. If the namespace uri wasn't
+     * empty (or null) all children and the node has attributes and/or children
+     * then the algorithm is recursively applied to the children.
+     */
     public void relink_namespace(ThreadContext context) {
-        if (node instanceof Element) {
-            clearCachedNode(node);
-            Element e = (Element) node;
-            String prefix = e.getPrefix();
-            String currentNS = e.getNamespaceURI();
-            if (prefix == null && currentNS == null) {
-                prefix = NokogiriHelpers.getPrefix(e.getNodeName());
-            } else if (currentNS != null) {
-                prefix = e.lookupPrefix(currentNS);
-            }
-            e.getOwnerDocument().setStrictErrorChecking(false);
-            String nsURI = e.lookupNamespaceURI(prefix);
-            this.node = NokogiriHelpers.renameNode(e, nsURI, e.getNodeName());
-
-            if (e.hasAttributes()) {
-                NamedNodeMap attrs = e.getAttributes();
-
-                for (int i = 0; i < attrs.getLength(); i++) {
-                    Attr attr = (Attr) attrs.item(i);
-                    String nsUri = "";
-                    String attrPrefix = attr.getPrefix();
-                    if (attrPrefix == null) {
-                        attrPrefix = NokogiriHelpers.getPrefix(attr.getNodeName());
-                    }
-                    String nodeName = attr.getNodeName();
-                    if ("xml".equals(attrPrefix)) {
-                        nsUri = "http://www.w3.org/XML/1998/namespace";
-                    } else if ("xmlns".equals(attrPrefix) || nodeName.equals("xmlns")) {
-                        nsUri = "http://www.w3.org/2000/xmlns/";
-                    } else {
-                        nsUri = attr.lookupNamespaceURI(attrPrefix);
-                    }
+        if (!(node instanceof Element)) {
+            return;
+        }
 
-                    if (nsUri == e.getNamespaceURI()) {
-                        nsUri = null;
-                    }
+        // TODO: this feels kind of weird, why are we clearing the XmlNode
+        // cache here !!!
+        clearCachedNode(node);
+        Element e = (Element) node;
 
-                    if (!(nsUri == null || "".equals(nsUri) || "http://www.w3.org/XML/1998/namespace".equals(nsUri))) {
-                        XmlNamespace.createFromAttr(context.getRuntime(), attr);
-                    }
-                    clearCachedNode(attr);
-                    NokogiriHelpers.renameNode(attr, nsUri, nodeName);
+        // disable error checking to prevent lines like the following
+        // from throwing a `NAMESPACE_ERR' exception:
+        // Nokogiri::XML::DocumentFragment.parse("<o:div>a</o:div>")
+        // since the `o' prefix isn't defined anywhere.
+        e.getOwnerDocument().setStrictErrorChecking(false);
+
+        String prefix = e.getPrefix();
+        String nsURI = e.lookupNamespaceURI(prefix);
+        this.node = NokogiriHelpers.renameNode(e, nsURI, e.getNodeName());
+
+        if (nsURI == null || nsURI == "") {
+            return;
+        }
+
+        if (e.hasAttributes()) {
+            NamedNodeMap attrs = e.getAttributes();
+
+            for (int i = 0; i < attrs.getLength(); i++) {
+                Attr attr = (Attr) attrs.item(i);
+                String nsUri = "";
+                String attrPrefix = attr.getPrefix();
+                if (attrPrefix == null) {
+                    attrPrefix = NokogiriHelpers.getPrefix(attr.getNodeName());
+                }
+                String nodeName = attr.getNodeName();
+                if ("xml".equals(attrPrefix)) {
+                    nsUri = "http://www.w3.org/XML/1998/namespace";
+                } else if ("xmlns".equals(attrPrefix) || nodeName.equals("xmlns")) {
+                    nsUri = "http://www.w3.org/2000/xmlns/";
+                } else {
+                    nsUri = attr.lookupNamespaceURI(attrPrefix);
                 }
-            }
 
-            if (this.node.hasChildNodes()) {
-                XmlNodeSet nodeSet = (XmlNodeSet)(children(context));
-                nodeSet.relink_namespace(context);
+                if (nsUri == e.getNamespaceURI()) {
+                    nsUri = null;
+                }
+
+                if (!(nsUri == null || "".equals(nsUri) || "http://www.w3.org/XML/1998/namespace".equals(nsUri))) {
+                    // Create a new namespace object and add it to the document
+                    // namespace cache.
+                    // TODO: why do we need the namespace cache ?
+                    XmlNamespace.createFromAttr(context.getRuntime(), attr);
+                }
+                clearCachedNode(attr);
+                NokogiriHelpers.renameNode(attr, nsUri, nodeName);
             }
         }
+
+        if (this.node.hasChildNodes()) {
+            XmlNodeSet nodeSet = (XmlNodeSet)(children(context));
+            nodeSet.relink_namespace(context);
+        }
     }
 
     // Users might extend XmlNode. This method works for such a case.
@@ -605,6 +621,7 @@ public class XmlNode extends RubyObject {
 
         NokogiriNamespaceCache nsCache = NokogiriHelpers.getNamespaceCacheFormNode(node);
         XmlNamespace cachedNamespace = nsCache.get(prefixString, hrefString);
+
         if (cachedNamespace != null) return cachedNamespace;
 
         Node namespaceOwner;
@@ -615,12 +632,14 @@ public class XmlNode extends RubyObject {
             final String uri = "http://www.w3.org/2000/xmlns/";
             String qName =
                 prefix.isNil() ? "xmlns" : "xmlns:" + prefixString;
+
             element.setAttributeNS(uri, qName, hrefString);
         }
         else if (node.getNodeType() == Node.ATTRIBUTE_NODE) namespaceOwner = ((Attr)node).getOwnerElement();
         else namespaceOwner = node.getParentNode();
         XmlNamespace ns = XmlNamespace.createFromPrefixAndHref(namespaceOwner, prefix, href);
         if (node != namespaceOwner) {
+
             this.node = NokogiriHelpers.renameNode(node, ns.getHref(), ns.getPrefix() + ":" + node.getLocalName());
         }
         updateNodeNamespaceIfNecessary(context, ns);
@@ -806,6 +825,13 @@ public class XmlNode extends RubyObject {
         }
 
         ctx.setInputSource(istream);
+        // TODO: for some reason, document.getEncoding() can be null or nil (don't know why)
+        // run `test_parse_with_unparented_html_text_context_node' few times to see this happen
+        if (document instanceof HtmlDocument && !(document.getEncoding() == null || document.getEncoding().isNil())) {
+        	HtmlDomParserContext htmlCtx= (HtmlDomParserContext) ctx;
+        	htmlCtx.setEncoding(document.getEncoding().asJavaString());
+        }
+        
         XmlDocument doc = ctx.parse(context, klass, runtime.getNil());
 
         RubyArray documentErrors = getErrorArray(document);
@@ -904,6 +930,10 @@ public class XmlNode extends RubyObject {
         return doc;
     }
 
+    public IRubyObject dup() {
+        return this.dup_implementation(getMetaClass().getClassRuntime(), true);
+    }
+
     @JRubyMethod
     public IRubyObject dup(ThreadContext context) {
         return this.dup_implementation(context, true);
@@ -917,13 +947,17 @@ public class XmlNode extends RubyObject {
     }
 
     protected IRubyObject dup_implementation(ThreadContext context, boolean deep) {
+       return dup_implementation(context.getRuntime(), deep);
+    }
+
+    protected IRubyObject dup_implementation(Ruby runtime, boolean deep) {
         XmlNode clone;
         try {
             clone = (XmlNode) clone();
         } catch (CloneNotSupportedException e) {
-            throw context.getRuntime().newRuntimeError(e.toString());
+            throw runtime.newRuntimeError(e.toString());
         }
-        if (node == null) throw context.getRuntime().newRuntimeError("FFFFFFFFFUUUUUUU");
+        if (node == null) throw runtime.newRuntimeError("FFFFFFFFFUUUUUUU");
         Node newNode = node.cloneNode(deep);
         clone.node = newNode;
         return clone;
@@ -1074,12 +1108,24 @@ public class XmlNode extends RubyObject {
 
     @JRubyMethod
     public IRubyObject namespace(ThreadContext context) {
-        if (doc instanceof HtmlDocument) return context.getRuntime().getNil();
+        Ruby runtime = context.getRuntime();
+        if (doc instanceof HtmlDocument) return runtime.getNil();
         NokogiriNamespaceCache nsCache = NokogiriHelpers.getNamespaceCacheFormNode(node);
+        String namespaceURI = node.getNamespaceURI();
+        if (namespaceURI == null || namespaceURI == "") {
+            return runtime.getNil();
+        }
+
         String prefix = node.getPrefix();
-        XmlNamespace namespace = nsCache.get(prefix == null ? "" : prefix, node.getNamespaceURI());
+        XmlNamespace namespace = nsCache.get(prefix == null ? "" : prefix, namespaceURI);
         if (namespace == null || namespace.isEmpty()) {
-            return context.getRuntime().getNil();
+          // if it's not in the cache, create an unowned, uncached namespace and
+          // return that. XmlReader can't insert namespaces into the cache, so
+          // this is necessary for XmlReader to work correctly.
+          namespace = new XmlNamespace(runtime, getNokogiriClass(runtime, "Nokogiri::XML::Namespace"));
+          IRubyObject rubyPrefix = NokogiriHelpers.stringOrNil(runtime, prefix);
+          IRubyObject rubyUri = NokogiriHelpers.stringOrNil(runtime, namespaceURI);
+          namespace.init(null, rubyPrefix, rubyUri, doc);
         }
 
         return namespace;
@@ -1296,7 +1342,7 @@ public class XmlNode extends RubyObject {
             } else if (prefix.equals("xmlns")) {
                 uri = "http://www.w3.org/2000/xmlns/";
             } else {
-                uri = findNamespaceHref(context, prefix);
+                uri = node.lookupNamespaceURI(prefix);
             }
         }
 
@@ -1386,13 +1432,10 @@ public class XmlNode extends RubyObject {
 
     @JRubyMethod(name = {"unlink", "remove"})
     public IRubyObject unlink(ThreadContext context) {
-        if(node.getParentNode() == null) {
-            throw context.getRuntime().newRuntimeError("TYPE: " + node.getNodeType()+ " PARENT NULL");
-        } else {
+        if (node.getParentNode() != null) {
             clearXpathContext(node.getParentNode());
             node.getParentNode().removeChild(node);
         }
-
         return this;
     }
 
@@ -1555,7 +1598,7 @@ public class XmlNode extends RubyObject {
             ((XmlDocument) this).resetNamespaceCache(context);
         }
 
-        relink_namespace(context);
+        other.relink_namespace(context);
         // post_add_child(context, this, other);
 
         return nodeOrTags;

data/ext/java/nokogiri/XmlXpathContext.java

@@ -17,10 +17,10 @@
  * distribute, sublicense, and/or sell copies of the Software, and to
  * permit persons to whom the Software is furnished to do so, subject to
  * the following conditions:
- * 
+ *
  * The above copyright notice and this permission notice shall be
  * included in all copies or substantial portions of the Software.
- * 
+ *
  * THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND,
  * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
@@ -58,13 +58,13 @@ import org.jruby.runtime.builtin.IRubyObject;
 import org.w3c.dom.Node;
 import org.w3c.dom.NodeList;
 
-import com.sun.org.apache.xml.internal.dtm.DTM;
-import com.sun.org.apache.xml.internal.utils.PrefixResolver;
-import com.sun.org.apache.xpath.internal.XPathContext;
-import com.sun.org.apache.xpath.internal.jaxp.JAXPExtensionsProvider;
-import com.sun.org.apache.xpath.internal.jaxp.JAXPPrefixResolver;
-import com.sun.org.apache.xpath.internal.jaxp.JAXPVariableStack;
-import com.sun.org.apache.xpath.internal.objects.XObject;
+import org.apache.xml.dtm.DTM;
+import org.apache.xml.utils.PrefixResolver;
+import org.apache.xpath.XPathContext;
+import org.apache.xpath.jaxp.JAXPExtensionsProvider;
+import org.apache.xpath.jaxp.JAXPPrefixResolver;
+import org.apache.xpath.jaxp.JAXPVariableStack;
+import org.apache.xpath.objects.XObject;
 
 /**
  * Class for Nokogiri::XML::XpathContext
@@ -110,7 +110,7 @@ public class XmlXpathContext extends RubyObject {
     }
 
     private JAXPExtensionsProvider getProviderInstance() throws ClassNotFoundException, IllegalArgumentException, InstantiationException, IllegalAccessException, InvocationTargetException {
-        Class<?> clazz = Class.forName("com.sun.org.apache.xpath.internal.jaxp.JAXPExtensionsProvider");
+        Class<?> clazz = Class.forName("org.apache.xpath.jaxp.JAXPExtensionsProvider");
         Constructor[] constructors = clazz.getDeclaredConstructors();
         for (int i = 0; i < constructors.length; i++) {
             Class[] parameterTypes = constructors[i].getParameterTypes();
@@ -184,8 +184,8 @@ public class XmlXpathContext extends RubyObject {
         Node contextNode = context.node;
 
         try {
-          com.sun.org.apache.xpath.internal.XPath xpathInternal = new com.sun.org.apache.xpath.internal.XPath (expr, null,
-                      prefixResolver, com.sun.org.apache.xpath.internal.XPath.SELECT );
+          org.apache.xpath.XPath xpathInternal = new org.apache.xpath.XPath (expr, null,
+                      prefixResolver, org.apache.xpath.XPath.SELECT );
 
           // We always need to have a ContextNode with Xalan XPath implementation
           // To allow simple expression evaluation like 1+1 we are setting