OpenRewrite: problem with custom Recipe and doAfterVisit method - openrewrite

I am trying to make a custom recipe with the following logic:
If I find a specific value for a property found in a yaml file, then I add a maven dependency to pom.xml.
In principle it seems to be an easy case, but I can't get it to work...
For this I have created the following recipe:
public class CustomRecipe extends Recipe {
public CustomRecipe() {}
#Override
protected TreeVisitor<?, ExecutionContext> getSingleSourceApplicableTest() {
return new HasSourcePath<>("**/application-*.yml");
}
#Override
public YamlVisitor<ExecutionContext> getVisitor() {
return new YamlIsoVisitor<ExecutionContext>() {
#Override
public Yaml.Mapping.Entry visitMappingEntry(final Yaml.Mapping.Entry entry, final ExecutionContext context) {
final Yaml.Mapping.Entry e = super.visitMappingEntry(entry, context);
if (e.getValue() instanceof Yaml.Scalar && ((Yaml.Scalar) e.getValue()).getValue().contains("ENC(")) {
// if jasypt usage is found in a .yml file then the Jasypt starter is added to pom.xml.
this.doAfterVisit(
new AddDependency("com.google.guava", "guava", "29.X", null, null, null,
null, null, null, null, null));
}
return e;
}
};
}
}
After correctly configuring the plugin in the final project to run the above recipe and running the mvn rewrite:run command in DEBUG mode I notice that the yaml visitor correctly runs through the yaml file properties, including finding the property value, and therefore the maven recipe AddDependency is added in the this.doAfterVisit(...) method.
However, the dependency is never added to the pom.xml of the project...
I'm new to OpenRewrite and maybe I'm not understanding correctly how to do this...
any ideas?
I am using:
maven 3.6.3
Open rewrite plugin: 4.23.0
Open rewrite: 7.22.0

I seem to have found a way to do what I wanted to do:
Use doNext instead of doAfterVisit to glue the Recipe to add the Maven dependency. Also, some parameters were missing to add to the AddDependency recipe like version and ifOnlyUsing:
#Override
public YamlVisitor<ExecutionContext> getVisitor() {
return new YamlIsoVisitor<ExecutionContext>() {
#Override
public Yaml.Mapping.Entry visitMappingEntry(final Yaml.Mapping.Entry entry, final ExecutionContext context) {
final Yaml.Mapping.Entry e = super.visitMappingEntry(entry, context);
if (e.getValue() instanceof Yaml.Scalar && ((Yaml.Scalar) e.getValue()).getValue().startsWith("ENC(")) {
CustomRecipe.this.doNext(new AddDependency("com.google.guava", "guava",
"29.X" , null, null, false, "org.junit.jupiter.api.*", null, null, false, null);
}
return e;
}
};
}

Related

Liferay create a custom PanelCategory for multiple portlet

I use Liferay 7.2 and Liferay IDE (eclipse). I created two separate Liferay admin portlet to creating a view for the database entries. I added in the first portlet "Teachers" a new panel called school with this generated code in application.list Package.
here is the code of - PanelApp.java
#Component(
immediate = true,
property = {
"panel.app.order:Integer=300",
"panel.category.key=" + TeachersPanelCategoryKeys.CONTROL_PANEL_CATEGORY
},
service = PanelApp.class
)
public class TeachersPanelApp extends BasePanelApp {
#Override
public String getPortletId() {
return TeachersPortletKeys.TEACHERS;
}
#Override
#Reference(
target = "(javax.portlet.name=" + TeachersPortletKeys.TEACHERS+ ")",
unbind = "-"
)
public void setPortlet(Portlet portlet) {
super.setPortlet(portlet);
}
}
.
public class TeachersPanelCategoryKeys {
public static final String CONTROL_PANEL_CATEGORY = "Teachers";
}
And here is the code of - PanelCategory.java
#Component(
immediate = true,
property = {
"panel.category.key=" + PanelCategoryKeys.SITE_ADMINISTRATION,
"panel.category.order:Integer=300"
},
service = PanelCategory.class)
public class TeachersPanelCategory extends BasePanelCategory {
#Override
public String getKey() {
return TeachersPanelCategoryKeys.CONTROL_PANEL_CATEGORY;
}
#Override
public String getLabel(Locale locale) {
return LanguageUtil.get(locale, "School");
}}
And here is the code of portlet.java
#Component(
immediate = true,
property = {
"com.liferay.portlet.add-default-resource=true",
"com.liferay.portlet.display-category=category.hidden",
"com.liferay.portlet.header-portlet-css=/css/main.css",
"com.liferay.portlet.layout-cacheable=true",
"com.liferay.portlet.private-request-attributes=false",
"com.liferay.portlet.private-session-attributes=false",
"com.liferay.portlet.render-weight=50",
"com.liferay.portlet.use-default-template=true",
"javax.portlet.display-name=Teachers",
"javax.portlet.expiration-cache=0",
"javax.portlet.init-param.template-path=/",
"javax.portlet.init-param.view-template=/view.jsp",
"javax.portlet.name=" + TeachersPortletKeys.TEACHERS,
"javax.portlet.resource-bundle=content.Language",
"javax.portlet.security-role-ref=power-user,user",
},
service = Portlet.class
)
public class TeachersPortlet extends MVCPortlet {
// some code to get entries from db
#Override
public void doView(final RenderRequest renderRequest, final RenderResponse renderResponse)
throws IOException, PortletException {
// some code
Now I want to add the second created portlet "Students" under the same Panel "School". I created it in the same way as "Teachers" but now I have two school panel. As it is shown in the image below.
I just want to display one panel category called school that contain both Teachers and Students in the list.
I do not know how I can think to do that.
As you're implementing a TeachersPanelCategory, I'm assuming you're also implementing a StudentsPanelCategory. From a naming perspective, I'd have expected a SchoolPanelCategory.
I'm currently at a loss of how the ControlPanel portlets actually declare their associated panel, but that place would be where you pick the common "school" panel and use the same spelling for both.
In other words: If you deploy two panels with the same name, I'd expect exactly what you document here. Make sure you're only deploying one of them
Edit: I'd like to know what TeachersPanelCategoryKeys.CONTROL_PANEL_CATEGORY is defined as, and the corresponding (assumed, not shown) StudentsPanelCategoryKeys.CONTROL_PANEL_CATEGORY. Both categories have the same label, but if they have different keys, they'll be different. I'm not sure what happens when you deploy two components with the same key: You should deploy only one.
Edit2: I've missed the code before: You're producing the key to your first category as "Teachers", and the label as "School". I'm assuming that the key for your other category is "Students". Liferay organizes the categories by key - and if the keys are different, then you'll end up with two different categories. Make their key more similar to their name (e.g. create a single SchoolCategory and associate your portlets/panelApps with that:
#Component(
immediate = true,
property = {
"panel.category.key=" + PanelCategoryKeys.SITE_ADMINISTRATION,
"panel.category.order:Integer=300"
},
service = PanelCategory.class)
public class SchoolPanelCategory extends BasePanelCategory {
#Override
public String getKey() {
return "school"; // this is the category that you want to associate with
}
#Override
public String getLabel(Locale locale) {
return LanguageUtil.get(locale, "School");
}
}
and
#Component(
immediate = true,
property = {
"panel.app.order:Integer=300",
"panel.category.key=school" // referencing the category created above
// (use the same for your StudentsPanelApp)
},
service = PanelApp.class
)
public class TeachersPanelApp extends BasePanelApp {
#Override
public String getPortletId() {
return TeachersPortletKeys.TEACHERS;
}
#Override
#Reference(
target = "(javax.portlet.name=" + TeachersPortletKeys.TEACHERS+ ")",
unbind = "-"
)
public void setPortlet(Portlet portlet) {
super.setPortlet(portlet);
}
}
(See the one-line comments within the code for the critical lines. Replace with proper constants if you like)

Eclipse CDT extend AdapterFactory

I try to override the functionality of CDT ResumeAtLine, MoveToLine, RunToLine. For this reason I created a custom SuspendResumeAdapterFactory but it isn't loaded but compiles and runs without error. Do I maybe need a custom adaptableType too?
Here is the content of my plugin.xml.
<extension point="org.eclipse.core.runtime.adapters">
<factory
class="my.package.CustomSuspendResumeAdapterFactory"
adaptableType="org.eclipse.cdt.dsf.ui.viewmodel.IVMContext">
<adapter type="org.eclipse.debug.core.model.ISuspendResume"/>
</factory>
</extension>
And here my CustomSuspendResumeAdapterFactory this class is reconstructed from memory not 100% sure if the syntax is correct, but I think it should be clear to see what I want to do.
package my.package;
import org.eclipse.cdt.dsf.datamodel.DMContexts;
import org.eclipse.cdt.dsf.debug.internal.ui.actions.MoveToLine;
import org.eclipse.cdt.dsf.debug.internal.ui.actions.ResumeAtLine;
import org.eclipse.cdt.dsf.debug.internal.ui.actions.RunToLine;
import org.eclipse.cdt.dsf.debug.service.IRunControl.IContainerDMContext;
import org.eclipse.cdt.dsf.debug.service.IRunControl.IExecutionDMContext;
import org.eclipse.cdt.dsf.ui.viewmodel.datamodel.IDMVMContext;
import org.eclipse.core.runtime.IAdaptable;
import org.eclipse.core.runtime.IAdapterFactory;
import org.eclipse.debug.core.DebugException;
import org.eclipse.debug.core.model.ISuspendResume;
public class CustomSuspendResumeAdapterFactory implements IAdapterFactory {
static class SuspendResume implements ISuspendResume, IAdaptable {
private final CustomRunToLine fRunToLine;
private final CustomMoveToLine fMoveToLine;
private final CustomResumeAtLine fResumeAtLine;
SuspendResume(IExecutionDMContext execCtx) {
fRunToLine = new CustomRunToLine(execCtx);
fMoveToLine = new CustomMoveToLine(execCtx);
fResumeAtLine = new CustomResumeAtLine(execCtx);
}
#SuppressWarnings("unchecked")
#Override
public <T> T getAdapter(Class<T> adapter) {
if (adapter.isInstance(RunToLine.class)) {
System.out.println("CUSTOM RUNTOLINE");
return (T)fRunToLine;
}
if (adapter.isInstance(MoveToLine.class)) {
System.out.println("CUSTOM MOVETOLINE");
return (T)fMoveToLine;
}
if (adapter.isInstance(ResumeToLine.class)) {
System.out.println("CUSTOM RESUMEATLINE");
return (T)fResumeAtLine;
}
return null;
}
#Override
public boolean canResume() { return false; }
#Override
public boolean canSuspend() { return false; }
// This must return true because the platform
// RunToLineActionDelegate will only enable the
// action if we are suspended
#Override
public boolean isSuspended() { return true; }
#Override
public void resume() throws DebugException {}
#Override
public void suspend() throws DebugException {}
}
#SuppressWarnings("unchecked")
#Override
public <T> T getAdapter(Object adaptableObject, Class<T> adapterType) {
if (ISuspendResume.class.equals(adapterType)) {
if (adaptableObject instanceof IDMVMContext) {
IExecutionDMContext execDmc = DMContexts.getAncestorOfType(
((IDMVMContext)adaptableObject).getDMContext(),
IExecutionDMContext.class);
// It only makes sense to RunToLine, MoveToLine or
// ResumeAtLine if we are dealing with a thread, not a container
if (execDmc != null && !(execDmc instanceof IContainerDMContext)) {
return (T)new SuspendResume(execDmc);
}
}
}
return null;
}
#Override
public Class<?>[] getAdapterList() {
return new Class[] { ISuspendResume.class };
}
}
Why your code is not run
You have provided a new adapter factory that converts object types that are already handled. i.e. your plugin.xml says you can convert IVMContext to ISuspendResume. But the DSF plug-in already provides such an adapter factory. If you have a new target type (like IMySpecialRunToLine) you could install a factory for that, it would take IVMContext and convert it to a IMySpecialRunToLine).
Although dated, the Eclipse Corner Article on Adapter Pattern may be useful if this is a new concept.
How to do custom Run To Line implementation
If you want to provide different implementation of Run To Line, you need to provide your own version of org.eclipse.cdt.dsf.debug.service.IRunControl2.runToLine(IExecutionDMContext, String, int, boolean, RequestMonitor). The org.eclipse.cdt.dsf.debug.internal.ui.actions.RunToLine class is simply glue to connect UI features (such as buttons/etc some provided directly, some by the core eclipse debug) to the DSF backend. i.e. if you look at what RunToLine does, all it actually does is get the IRunControl2 service and call runToLine on it.
The way to provider your own implementation of IRunControl2 is override org.eclipse.cdt.dsf.gdb.service.GdbDebugServicesFactory.createRunControlService(DsfSession) and provide your own GdbDebugServicesFactory in your custom launch delegate by overriding org.eclipse.cdt.dsf.gdb.launching.GdbLaunchDelegate.newServiceFactory(ILaunchConfiguration, String)
RunToLine will be triggered when the user select Run To Line from the popup menu in the editor, as per this screenshot:

Eclipse Plugin, How can I know when IResourceDeltaVisitor ends processing tree nodes?

I wrote an Eclipse Plugin that basically allow a programmer to select a Java source from the Project Explorer and by selecting the corresponding DropDown menu option it will creates an interface .java file based on the one selected.
Everything works fine, but now I need to program the update part of the job.
The update requierement is simple, I need to listen for changes and identify that the sources that have the interface generated have been modified and recreate the interface file.
To do this I wrote a class that implements IResourceChangeListener interface.
That class looks like:
public class DTOChangeListener implements IResourceChangeListener {
private List<UpdatedUnit> updatedUnits;
public DTOChangeListener() {
super();
this.updatedUnits=new ArrayList<UpdatedUnit>();
}
#Override
public void resourceChanged(IResourceChangeEvent event) {
try{
if(event.getType() == IResourceChangeEvent.POST_CHANGE){
event.getDelta().accept(this.buildVisitor());
}
}catch(CoreException ex){
ex.printStackTrace();
}
}
protected IResourceDeltaVisitor buildVisitor(){
IResourceDeltaVisitor result=new IResourceDeltaVisitor() {
#Override
public boolean visit(IResourceDelta resDelta) throws CoreException {
String resName=resDelta.getResource().getName();
if(resName==null || resName.equals("")){
return true;
}
String[] splits=resName.split("\\.");
String name = splits[0];
if(name.contains("PropertyAccess")){
return false;
}
String interfaceName=name + "PropertyAccess";
String interfaceFile=interfaceName + ".java";
IResource res=resDelta.getResource();
if((res instanceof IFolder) || (res instanceof IProject)){
// Avoid Folder & Project Nodes
return true;
}
IProject project=res.getProject();
if(project!=null){
if(project.isNatureEnabled("org.eclipse.jdt.core.javanature")){
IJavaElement element=JavaCore.create(res);
if(element instanceof ICompilationUnit){
ICompilationUnit unit=(ICompilationUnit)element;
IPath path=res.getProjectRelativePath().removeLastSegments(1);
IResource propertyAccess=project.findMember(path.append(interfaceFile));
if(propertyAccess!=null){
UpdatedUnit updatedUnit=new UpdatedUnit(project, path, unit);
updatedUnits.add(updatedUnit);
return false;
}
}
}
}
return true;
}
};
return result;
}
public List<UpdatedUnit> getUpdatedUnits() {
return updatedUnits;
}
}
I add the Listener to the Workspace, now the question I have is:
How can I know when the updatedUnits List is completed in order to proccess the list with my own code?
One posible answer to this question would be, don't worry, the:
event.getData().accept(this.buildVisitor());
will block until proccessing of the visitor finish.
but at least is not documented like it would.
Any ideas would be appreciated.
Thanks in Advance.
Daniel
Unless it's documented to not block, it blocks.

Using IoC container as a service locator for HttpHandler

This question relates to my other post.
Ok so after a bit more messing around I decided to do it this way. Which seems to work fine when I run it, although I'm getting the following error in NUnit: Could not load file or assembly 'Castle.Core, Version=1.0.3.0, Culture=neutral, PublicKeyToken=407dd0808d44fbdc' or one of its dependencies. The located assembly's manifest definition does not match the assembly reference. (Exception from HRESULT: 0x80131040) So not sure what is happening there???
Just wanted to know what others thought about the design and if there are any obvious 'no no's' or improvements. I.e. Is the constructor of the base handler a good place to instantiate the windsor component or is there a better place to do this? As I said in the original post the idea behind doing things this way was to keep the components nicely decoupled and to make unit testing easy. I should also add I'm new to unit testing, mocking. Thanks!
public abstract class BaseHttpHandler : IHttpHandler
{
private HttpContext _httpContext;
private ILogger _logger;
private IDataRepository _dataRepository;
protected HttpRequest Request { get { return _httpContext.Request; } }
protected HttpResponse Response { get { return _httpContext.Response; } }
protected bool IsRequestFromUAD { get { return Request.UserAgent == null ? false : Request.UserAgent.Equals("UAD"); } }
protected ILogger Logger { get { return _logger; } }
protected IDataRepository DataRepository { get { return _dataRepository; } }
public virtual bool IsReusable { get { return false; } }
public BaseHttpHandler()
{
var container = new WindsorContainer(new XmlInterpreter(new ConfigResource("castle")));
_logger = container.Resolve<ILogger>();
_dataRepository = container.Resolve<IDataRepository>();
}
public void ProcessRequest(HttpContext context)
{
_httpContext = context;
ProcessRequest(new HttpContextWrapper(context));
}
public abstract void ProcessRequest(HttpContextBase context);
}
public class UADRecordHttpHandler : BaseHttpHandler
{
public override void ProcessRequest(HttpContextBase context)
{
if (IsRequestFromUAD)
{
using (var reader = new StreamReader(context.Request.InputStream))
{
string data = reader.ReadToEnd();
if (Logger != null)
Logger.Log(data);
if(DataRepository != null)
DataRepository.Write(data);
context.Response.Write(data);
}
}
else
ReturnResponse(HttpStatusCode.BadRequest);
}
}
That's a very bad thing to do, what you're doing here. You should have one instance of the container per application, while with this code you will have one per each request.
About the error in NUnit: make sure you don't have other versions of Castle assemblies in the GAC. If so, uninstall them.
About your BaseHttpHandler: the problem with this implementation is that you're creating a new container. Instead, use a single container per application, like Krzysztof said. Use a static service locator, e.g. CommonServiceLocator. (I never recommend this but it's one of the few places where it does make sense).

Running External Apps on save in Eclipse

Since we cannot setup Eclipse's RSE to use at the tool for remote editing, I have installed Unison. But how can I get Eclipse to automatically run unison on every file save? Is there an eclipse plugin available for this?
TIA
You can setup it to be run on every build. Any external tool can be run on every build, just open project's preferences, go to Builders page, click “New…”.
Depending on the importance, I would write a simple plugin to handle this.
EDIT:
All you really need to do is this:
1) Create the plugin from the templates with the RCP\PDE Eclipse install
2) Add the following code to your activator...
#Override
public void start( final BundleContext context ) throws Exception {
super.start( context );
plugin = this;
ICommandService commandService = (ICommandService)plugin.getWorkbench().getService( ICommandService.class );
commandService.addExecutionListener( new IExecutionListener() {
public void notHandled( final String commandId, final NotHandledException exception ) {}
public void postExecuteFailure( final String commandId, final ExecutionException exception ) {}
public void postExecuteSuccess( final String commandId, final Object returnValue ) {
if ( commandId.equals( "org.eclipse.ui.file.save" ) ) {
// add in your action here...
// personally, I would use a custom preference page,
// but hard coding would work ok too
}
}
public void preExecute( final String commandId, final ExecutionEvent event ) {}
} );
}