rubys 01/08/17 19:12:17
Modified: proposal/gump gen.java
Log:
Arg! Now get rid of extra CR's!
Revision Changes Path
1.9 +284 -284 jakarta-alexandria/proposal/gump/gen.java
Index: gen.java
===================================================================
RCS file: /home/cvs/jakarta-alexandria/proposal/gump/gen.java,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -r1.8 -r1.9
--- gen.java 2001/08/18 02:08:51 1.8
+++ gen.java 2001/08/18 02:12:17 1.9
@@ -1,284 +1,284 @@
-// TRaX classes
-import javax.xml.transform.dom.DOMResult;
-import javax.xml.transform.dom.DOMSource;
-import javax.xml.transform.OutputKeys;
-import javax.xml.transform.stream.StreamResult;
-import javax.xml.transform.stream.StreamSource;
-import javax.xml.transform.Transformer;
-import javax.xml.transform.TransformerFactory;
-
-// Java API for XML Parsing classes
-import javax.xml.parsers.DocumentBuilder;
-import javax.xml.parsers.DocumentBuilderFactory;
-
-// DOM classes
-import org.w3c.dom.Attr;
-import org.w3c.dom.Document;
-import org.w3c.dom.Element;
-import org.w3c.dom.NamedNodeMap;
-import org.w3c.dom.Node;
-import org.w3c.dom.traversal.NodeIterator;
-import org.xml.sax.SAXParseException;
-
-
-// Java classes
-import java.io.FileOutputStream;
-import java.util.Enumeration;
-import java.util.Hashtable;
-import java.util.Vector;
-
-// Apache xpath
-import org.apache.xpath.XPathAPI;
-
-public class gen {
-
- DocumentBuilderFactory dFactory = DocumentBuilderFactory.newInstance();
- TransformerFactory tFactory = TransformerFactory.newInstance();
-
- /**
- * parse an XML source file into an DOM
- * @param name of source file
- * @return Node
- */
- private Document parse(String source) throws Exception {
- try {
- DocumentBuilder dBuilder = dFactory.newDocumentBuilder();
- return dBuilder.parse(new java.io.File(source));
- } catch (SAXParseException e) {
- System.err.print("Error parsing file " + source);
- System.err.println(" line " + e.getLineNumber() + ": ");
- throw e;
- }
- }
-
- /**
- * transform a DOM given a stylesheet
- * @param DOM to be transformed
- * @param sheet to be used
- * @return Node
- */
- private Node transform(Node dom, String sheet) throws Exception {
- StreamSource source = new StreamSource("stylesheet/"+sheet);
- Transformer transformer = tFactory.newTransformer(source);
- DOMResult output = new DOMResult();
- transformer.transform(new DOMSource(dom), output);
- return output.getNode();
- }
-
- /**
- * copy child nodes (attributes, elements, text, etc).
- * @param source element
- * @param target element
- */
- private void copyChildren(Element source, Element target) {
- Node child=source.getFirstChild();
- while (child != null) {
- Node next=child.getNextSibling();
- target.appendChild(child);
- child=next;
- }
-
- NamedNodeMap attrs = source.getAttributes();
- for (int i=0; i<attrs.getLength(); i++) {
- Node a = attrs.item(i);
- if (target.getAttributeNode(a.getNodeName())==null) {
- target.setAttribute(a.getNodeName(),a.getNodeValue());
- }
- }
- }
-
- /**
- * expand hrefs in place, recursively
- * @param source element
- */
- private void expand(Element node) throws Exception {
- // expand hrefs
- Attr href = node.getAttributeNode("href");
- if (href != null && !node.getNodeName().equals("url")) {
- String source=href.getValue();
- Node sub = parse(source);
-
- if (source.lastIndexOf(".")>0)
- source=source.substring(0,source.lastIndexOf("."));
-
- if (source.lastIndexOf("/")>0)
- source=source.substring(source.lastIndexOf("/")+1);
-
- node.removeAttribute("href");
- node.setAttribute("defined-in", source);
-
- Document doc = node.getOwnerDocument();
- Element copy=(Element)doc.importNode(sub.getFirstChild(), true);
- copyChildren(node, copy);
-
- Element parent = (Element)node.getParentNode();
- if (node.getNodeName().equals("profile")) {
- copyChildren(copy, parent);
- } else {
- parent.replaceChild(copy,node);
- }
- }
-
- // recurse through the children
- Node child=node.getFirstChild();
- while (child != null) {
- Node next=child.getNextSibling();
- if (child.getNodeType()==Node.ELEMENT_NODE) expand((Element)child);
- child=next;
- }
- }
-
- /**
- * Merge the contents of nodes with the same value for the name attribute.
- * Attributes from later definitions get added (or overlay) prior
- * definitions. Elements get appended.
- * @param type Element localname. Typically project or repository.
- * @param list Hashtable used for recursion. Must initially be empty.
- * @param Node Starting point for search.
- */
- private void merge(String type, Hashtable list, Node parent) {
- Node child=parent.getFirstChild();
- while (child != null) {
- Node next=child.getNextSibling();
- if (child.getNodeName().equals(type)) {
- Element element = (Element) child;
- String name = element.getAttributeNode("name").getValue();
-
- Element priorDefinition = (Element)list.get(name);
- if (priorDefinition == null) {
- list.put(name, element);
- merge(type, list, element);
- } else if (priorDefinition != element) {
- element.getParentNode().removeChild(element);
- copyChildren(element, priorDefinition);
- element=priorDefinition;
- }
- }
- child=next;
- }
- }
-
- /**
- * Unnest all elements of a given type by moving them all to become
- * direct children of the specified root node. In the process, merge
- * all matching nodes which contain the same value for the name attribute.
- * For elements that get "hoisted", an additional "defined-in" attribute
- * is added indicating where the element was originally defined.
- * @param type Element localname. Typically project or repository.
- * @param Node Root (workspace) node
- */
- private void flatten(String type, Node root) {
- Hashtable list = new Hashtable();
- merge(type, list, root);
- for (Enumeration e=list.keys(); e.hasMoreElements();) {
- Element element = (Element)list.get(e.nextElement());
- Element parent = (Element)element.getParentNode();
- if (parent != root) {
- parent.removeChild(element);
- String definedIn = parent.getAttributeNode("name").getValue();
- element.setAttribute("defined-in",definedIn);
- root.appendChild(element);
- }
- }
- }
-
- /**
- * serialize a DOM tree to file
- * @param DOM to be transformed
- * @param destination file
- */
- private void output(Node dom, String dest) throws Exception {
- DOMSource in = new DOMSource(dom);
- StreamResult out = new StreamResult(dest);
- tFactory.newTransformer().transform(in, out);
- }
-
- /**
- * Replace ant "depend" elements with "property" elements. This is
- * a convenience "syntatic sugar" that makes for simpler project
- * definitions. Attribute "property" becomes name. Attributes
- * reference="jarpath" and classpath="true" are added.
- * @param document to be transformed
- */
- private void antDependsToProperties(Document document) throws Exception {
- NodeIterator nl = XPathAPI.selectNodeIterator(document, "//ant/depend");
- for (Node depend=nl.nextNode(); depend!=null; depend=nl.nextNode()) {
-
- // create a new element based on existing element
- Element property = document.createElement("property");
- property.setAttribute("reference", "jarpath");
- property.setAttribute("classpath", "add");
- copyChildren((Element)depend, property);
-
- // change property attribute to name attribute
- if (property.getAttributeNode("name")==null) {
- Attr pname = property.getAttributeNode("property");
- if (pname != null) {
- property.setAttribute("name",pname.getValue());
- property.removeAttributeNode(pname);
- }
- }
-
- // replace existing element with new one
- depend.getParentNode().replaceChild(property, depend);
- }
- }
-
- /**
- * Rename <module> to <project>. This is a transitional convenience
- * as I move from the nested project approach to a simple declaration
- * of the projects (or perhaps, instead the targets) contained within
- * a module.
- * @param document to be transformed
- */
- private void renameModuleToProject(Document document) throws Exception {
-
- // safely get a list of all modules
- NodeIterator nl = XPathAPI.selectNodeIterator(document, "//module");
- Vector list = new Vector();
- for (Node module=nl.nextNode(); module!=null; module=nl.nextNode()) {
- list.add(module);
- }
-
- // replace all elements in that list with projects
- for (Enumeration e=list.elements(); e.hasMoreElements();) {
- Element module = (Element)e.nextElement();
- Element project = document.createElement("project");
- copyChildren(module, project);
- module.getParentNode().replaceChild(project, module);
- }
- }
-
- /**
- * merge, sort, and insert defaults into a workspace
- * @param DOM to be transformed
- * @param sheet to be used
- * @return Node
- */
- private gen(String source) throws Exception {
- Document workspace = parse(source);
- expand((Element)workspace.getFirstChild());
- renameModuleToProject(workspace);
- flatten("project", workspace.getFirstChild());
- flatten("repository", workspace.getFirstChild());
- antDependsToProperties(workspace);
-
- Node resolved = transform(workspace, "defaults.xsl");
- output (resolved, "work/merge.xml");
-
- Node sorted = transform(resolved, "sortdep.xsl");
- output (sorted, "work/sorted.xml");
- }
-
- public static void main(String[] args) {
- try {
- if (args.length == 1)
- new gen(args[0]);
- else
- System.out.println("Usage: gen workspace.xml");
- } catch (Exception e) {
- e.printStackTrace();
- System.exit(99);
- }
- }
-}
+// TRaX classes
+import javax.xml.transform.dom.DOMResult;
+import javax.xml.transform.dom.DOMSource;
+import javax.xml.transform.OutputKeys;
+import javax.xml.transform.stream.StreamResult;
+import javax.xml.transform.stream.StreamSource;
+import javax.xml.transform.Transformer;
+import javax.xml.transform.TransformerFactory;
+
+// Java API for XML Parsing classes
+import javax.xml.parsers.DocumentBuilder;
+import javax.xml.parsers.DocumentBuilderFactory;
+
+// DOM classes
+import org.w3c.dom.Attr;
+import org.w3c.dom.Document;
+import org.w3c.dom.Element;
+import org.w3c.dom.NamedNodeMap;
+import org.w3c.dom.Node;
+import org.w3c.dom.traversal.NodeIterator;
+import org.xml.sax.SAXParseException;
+
+
+// Java classes
+import java.io.FileOutputStream;
+import java.util.Enumeration;
+import java.util.Hashtable;
+import java.util.Vector;
+
+// Apache xpath
+import org.apache.xpath.XPathAPI;
+
+public class gen {
+
+ DocumentBuilderFactory dFactory = DocumentBuilderFactory.newInstance();
+ TransformerFactory tFactory = TransformerFactory.newInstance();
+
+ /**
+ * parse an XML source file into an DOM
+ * @param name of source file
+ * @return Node
+ */
+ private Document parse(String source) throws Exception {
+ try {
+ DocumentBuilder dBuilder = dFactory.newDocumentBuilder();
+ return dBuilder.parse(new java.io.File(source));
+ } catch (SAXParseException e) {
+ System.err.print("Error parsing file " + source);
+ System.err.println(" line " + e.getLineNumber() + ": ");
+ throw e;
+ }
+ }
+
+ /**
+ * transform a DOM given a stylesheet
+ * @param DOM to be transformed
+ * @param sheet to be used
+ * @return Node
+ */
+ private Node transform(Node dom, String sheet) throws Exception {
+ StreamSource source = new StreamSource("stylesheet/"+sheet);
+ Transformer transformer = tFactory.newTransformer(source);
+ DOMResult output = new DOMResult();
+ transformer.transform(new DOMSource(dom), output);
+ return output.getNode();
+ }
+
+ /**
+ * copy child nodes (attributes, elements, text, etc).
+ * @param source element
+ * @param target element
+ */
+ private void copyChildren(Element source, Element target) {
+ Node child=source.getFirstChild();
+ while (child != null) {
+ Node next=child.getNextSibling();
+ target.appendChild(child);
+ child=next;
+ }
+
+ NamedNodeMap attrs = source.getAttributes();
+ for (int i=0; i<attrs.getLength(); i++) {
+ Node a = attrs.item(i);
+ if (target.getAttributeNode(a.getNodeName())==null) {
+ target.setAttribute(a.getNodeName(),a.getNodeValue());
+ }
+ }
+ }
+
+ /**
+ * expand hrefs in place, recursively
+ * @param source element
+ */
+ private void expand(Element node) throws Exception {
+ // expand hrefs
+ Attr href = node.getAttributeNode("href");
+ if (href != null && !node.getNodeName().equals("url")) {
+ String source=href.getValue();
+ Node sub = parse(source);
+
+ if (source.lastIndexOf(".")>0)
+ source=source.substring(0,source.lastIndexOf("."));
+
+ if (source.lastIndexOf("/")>0)
+ source=source.substring(source.lastIndexOf("/")+1);
+
+ node.removeAttribute("href");
+ node.setAttribute("defined-in", source);
+
+ Document doc = node.getOwnerDocument();
+ Element copy=(Element)doc.importNode(sub.getFirstChild(), true);
+ copyChildren(node, copy);
+
+ Element parent = (Element)node.getParentNode();
+ if (node.getNodeName().equals("profile")) {
+ copyChildren(copy, parent);
+ } else {
+ parent.replaceChild(copy,node);
+ }
+ }
+
+ // recurse through the children
+ Node child=node.getFirstChild();
+ while (child != null) {
+ Node next=child.getNextSibling();
+ if (child.getNodeType()==Node.ELEMENT_NODE) expand((Element)child);
+ child=next;
+ }
+ }
+
+ /**
+ * Merge the contents of nodes with the same value for the name attribute.
+ * Attributes from later definitions get added (or overlay) prior
+ * definitions. Elements get appended.
+ * @param type Element localname. Typically project or repository.
+ * @param list Hashtable used for recursion. Must initially be empty.
+ * @param Node Starting point for search.
+ */
+ private void merge(String type, Hashtable list, Node parent) {
+ Node child=parent.getFirstChild();
+ while (child != null) {
+ Node next=child.getNextSibling();
+ if (child.getNodeName().equals(type)) {
+ Element element = (Element) child;
+ String name = element.getAttributeNode("name").getValue();
+
+ Element priorDefinition = (Element)list.get(name);
+ if (priorDefinition == null) {
+ list.put(name, element);
+ merge(type, list, element);
+ } else if (priorDefinition != element) {
+ element.getParentNode().removeChild(element);
+ copyChildren(element, priorDefinition);
+ element=priorDefinition;
+ }
+ }
+ child=next;
+ }
+ }
+
+ /**
+ * Unnest all elements of a given type by moving them all to become
+ * direct children of the specified root node. In the process, merge
+ * all matching nodes which contain the same value for the name attribute.
+ * For elements that get "hoisted", an additional "defined-in" attribute
+ * is added indicating where the element was originally defined.
+ * @param type Element localname. Typically project or repository.
+ * @param Node Root (workspace) node
+ */
+ private void flatten(String type, Node root) {
+ Hashtable list = new Hashtable();
+ merge(type, list, root);
+ for (Enumeration e=list.keys(); e.hasMoreElements();) {
+ Element element = (Element)list.get(e.nextElement());
+ Element parent = (Element)element.getParentNode();
+ if (parent != root) {
+ parent.removeChild(element);
+ String definedIn = parent.getAttributeNode("name").getValue();
+ element.setAttribute("defined-in",definedIn);
+ root.appendChild(element);
+ }
+ }
+ }
+
+ /**
+ * serialize a DOM tree to file
+ * @param DOM to be transformed
+ * @param destination file
+ */
+ private void output(Node dom, String dest) throws Exception {
+ DOMSource in = new DOMSource(dom);
+ StreamResult out = new StreamResult(dest);
+ tFactory.newTransformer().transform(in, out);
+ }
+
+ /**
+ * Replace ant "depend" elements with "property" elements. This is
+ * a convenience "syntatic sugar" that makes for simpler project
+ * definitions. Attribute "property" becomes name. Attributes
+ * reference="jarpath" and classpath="true" are added.
+ * @param document to be transformed
+ */
+ private void antDependsToProperties(Document document) throws Exception {
+ NodeIterator nl = XPathAPI.selectNodeIterator(document, "//ant/depend");
+ for (Node depend=nl.nextNode(); depend!=null; depend=nl.nextNode()) {
+
+ // create a new element based on existing element
+ Element property = document.createElement("property");
+ property.setAttribute("reference", "jarpath");
+ property.setAttribute("classpath", "add");
+ copyChildren((Element)depend, property);
+
+ // change property attribute to name attribute
+ if (property.getAttributeNode("name")==null) {
+ Attr pname = property.getAttributeNode("property");
+ if (pname != null) {
+ property.setAttribute("name",pname.getValue());
+ property.removeAttributeNode(pname);
+ }
+ }
+
+ // replace existing element with new one
+ depend.getParentNode().replaceChild(property, depend);
+ }
+ }
+
+ /**
+ * Rename <module> to <project>. This is a transitional convenience
+ * as I move from the nested project approach to a simple declaration
+ * of the projects (or perhaps, instead the targets) contained within
+ * a module.
+ * @param document to be transformed
+ */
+ private void renameModuleToProject(Document document) throws Exception {
+
+ // safely get a list of all modules
+ NodeIterator nl = XPathAPI.selectNodeIterator(document, "//module");
+ Vector list = new Vector();
+ for (Node module=nl.nextNode(); module!=null; module=nl.nextNode()) {
+ list.add(module);
+ }
+
+ // replace all elements in that list with projects
+ for (Enumeration e=list.elements(); e.hasMoreElements();) {
+ Element module = (Element)e.nextElement();
+ Element project = document.createElement("project");
+ copyChildren(module, project);
+ module.getParentNode().replaceChild(project, module);
+ }
+ }
+
+ /**
+ * merge, sort, and insert defaults into a workspace
+ * @param DOM to be transformed
+ * @param sheet to be used
+ * @return Node
+ */
+ private gen(String source) throws Exception {
+ Document workspace = parse(source);
+ expand((Element)workspace.getFirstChild());
+ renameModuleToProject(workspace);
+ flatten("project", workspace.getFirstChild());
+ flatten("repository", workspace.getFirstChild());
+ antDependsToProperties(workspace);
+
+ Node resolved = transform(workspace, "defaults.xsl");
+ output (resolved, "work/merge.xml");
+
+ Node sorted = transform(resolved, "sortdep.xsl");
+ output (sorted, "work/sorted.xml");
+ }
+
+ public static void main(String[] args) {
+ try {
+ if (args.length == 1)
+ new gen(args[0]);
+ else
+ System.out.println("Usage: gen workspace.xml");
+ } catch (Exception e) {
+ e.printStackTrace();
+ System.exit(99);
+ }
+ }
+}
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
