I have a method in some class and i would like to see exactly where it is being used. Compared to Java and Eclipse you can simply tell it to show all the references and even a call hierarchy of a method. Is there anything similar in Xcode?
I know that Objective-C does not follow the same ways of identifying method signature as Java does (i.e. there is no method to a class, just a bunch of selectors mapped to an id at runtime), so i'm having a hard time trying to figure out how Xcode could even accomplish this.
Look here:
There is no equivalent menu or context menu.
In addition to callers you can access callees, superclasses, etc.
If you don't use Ctrl-1 for Spaces you can access that menu with it.
Note that you need to use the Standard or Assistant Editor:
Callers and callees are not shown if the Version Editor is selected:
Your best bet is likely string searching, but realize that you can use patterns (click the little magnifying glass in the search bar in the file browser pane).
You've hinted at one of the reasons it's so hard to do otherwise. Objective-C's runtime allows many different ways of finding and executing code.
Related
I need my Visual Studio extension to react to debugging events. I've registered a IDebugEventCallback2 and I'm receiving events, but all I get for each event is an opaque IDebugEvent2 and a Guid, many of which are not only undocumented but don't appear anywhere on the web (or in my registry).
My specific requirement at the moment is to know when the process is Continued - ie. the user has hit Continue, Run to cursor, etc. What Guid should I be looking for?
Or is there some other event family that I should be subscribing to?
(More generally, is there some way I'm missing to find out about the events that are delivered to my IDebugEventCallback2::Event callback, when many of them don't appear in MSDN or anywhere else? Thanks!)
There is no easy way to do this. Actions such as Continue and Run to cursor are abstractions implemented by Visual Studio and do not correspond to any unique event(s) with respect to the debug engine. The debug engine event reporting interface IDebugEventCallback2 will enable you to get notified only on fine-grained events such as when creating a breakpoint or reaching a breakpoint.
While Visual Studio enables you to perform actions such as Continue and Run to cursor programmatically, it does not provide a direct way to get notified when they are taken.
You can use EnvDTE.DebuggerEvents.OnXxx events to get notified when something is about to happen. In particular, the OnEnterBreakMode event enables you to intercept a breakpoint hit and possibly take an action. You can also inspect all the details of the reached breakpoint(s) using the current EnvDTE.Debugger inside the event handler.
Now with some effort, you can use these constructs to implement events that correspond to all Visual Studio debugging actions including Continue and Run to cursor accurately. If you require additional events not provided by EnvDTE.DebuggerEvents (such as when a breakpoint is inserted), you have no choice but use IDebugEventCallback2.Event. In this case if you have specific events in mind, please mention them explicitly and I might be able to tell you the corresponding IDebugEventCallback2.Event GUIDs.
You probably got off on the wrong foot here, the IDebugXxx interfaces were really intended to create your own debugging engine. Still useful perhaps to see what is going on in the existing engine, you are suppose to use QueryInterface() to discover the specific interface that matches the event. Like IDebugEngineCreateEvent2, IDebugProcessCreateEvent2, IDebugProcessDestroyEvent2, etcetera. There are a bunch of them, they are listed in the VSSDK\VisualStudioIntegration\Common\Inc\msdbg.h header and include their IID.
Best thing to do is peek at this sample, it shows how to crack the event notification into the specific interfaces. The AD7Events.cs source file shows the wrappers. Do keep in mind that this sample was made to show how to create an engine, not how to use the one already built into VS.
But yes, you are not going to get a "continue debug" event out of this. A debugging engine does not need that event since it is already responsible for taking care of that by itself. It already knows when it calls IDebugProgram3::Continue().
Surely what you are looking for is IVsDebugger.AdviseDebuggerEvents(). That's the one that tells you what the built-in debugger is doing. You need to pass an object of a class that implements IVsDebuggerEvents, your OnModeChanged() method will be called with a DBGMODE notification. I don't see a lot of fantastic examples that demonstrate usage, this web page might be helpful.
In Windows 7's Windows Explorer list view (what allegedly is not list view at all) in the Details view, you can start selection marquee in the first (Name) column. You just need to start outside the actual name.
The same is true for default-style list view control in Details view.
But if you set the list view control to the Explorer style (using the SetWindowTheme), what should mimic the Windows Explorer, this does not work anymore. You can start selection in the second and later columns only.
SetWindowTheme(listView1.Handle, "explorer", null);
Is there any way to make list view mimic the Explorer selection style?
I suppose there's no settings to enable such behavior and this would have to be coded. Like handling the mouse down and triggering selection. But I have no idea how to do that.
Thanks.
Ntb, I'm using C++Builder, but this should be purely Win32 issue. I've tested this with WinForms too (hence the C# sample above).
Is there any way to make list view it mimic the Explorer selection style?
No, SysListView32 in explorer theme does not behave that way. The control used by the modern Explorer is actually DirectUIHwnd. And you are not able to use one of them.
The only way to get the behaviour of DirectUIHwnd is to code it yourself. I expect that's possible to do but I'd also expect it to be very difficult to achieve.
Actually there is a way to mimic Explorer selection behavior. It requires a lot of additional declarative work, but it is possible.
You need to get undocumented IListView interface via undocumented LVM_QUERYINTERFACE message (note that interface declaration and GUIDs are different for Windows Vista and Windows 7+). Details about constants and declarations can be found here:
IListView at Geoff Chappell - Software Analyst or
Undocumented List View Features at Code Project.
After acquiring the interface all you need is simply a call to SetSelectionFlags(1, 1) method. Voila you are done.
I'm in VS2010, in a new Word Add-In project. This is my first attempt at Word development using VSTO. The example I'm trying has this line:
Document vstoDoc = Globals.Factory.GetVstoObject(this.Application.ActiveDocument);
But when I add this line Visual Studio says it can't find "Factory". Indeed, it's not in Intellisense.
I've got references to:
Accessibility
Microsoft.Office.Interop.Word
Microsoft.Office.Tools.Common.v9.0
Microsoft.Office.Tools.v9.0
Microsoft.Office.Tools.Word.v9.0
Microsoft.VisualStudio.Tools.Applications.Runtime.v9.0
Office
and all the usual System references.
Where am I going wrong and why can't I get to "Factory"?
stdole
That example looks a bit weird to me. Never seen that sort of reference before.
Generally, with Vsto, you hook into EVENTS on, say, the main Word App object.
Then, from within the event, you usually are passed a reference to the particular DOC object that the event is occurring for (say, being opened or saved, etc). In that way, there shouldn't be any need for using the "globals" object or the "factory" object, whereever they might be.
What method is that code in? A little more context might help.
I think the recommended way of doing this is:
Globals.ThisAddin.Application.ActiveDocument
I'm working on a little macro record/replay tool which can automate a few very old Visual Basic 6 GUIs we have. To do so, I'm identifying the controls by their name (the value of the name property of a control, that is).
One part of this tool needs to determine the name of a control given its HWND. For newer Visual Basic applications which were done using VB.NET, I can use the WM_GETCONTROLNAME window message. This works nicely.
However, this message is not understood by older windows. Is there any way to do this for controls of Visual Basic 6 applications? A solution which does not require being in the process of the GUI would be preferrable, but if I had a solution which only works inside the GUI process then that would be acceptable as well (since I can do the injection myself).
UPDATE: One thing I just tried, this moderate success: I used the AccessibleObjectFromWindow to check for implementations of the IAccessible interface of the object which shows the given HWND. In case I get an implementation (it seems that many [all?] Visual Basic controls implement this interface), I use the accName property to read out the "accessible name". Sometimes this does yield a useful string, but usually it doesn't.
I believe the only way would be getting inside the process and obtaining a pointer to the Form object, yet I have no idea how to do it from outside.
Is it possible you add support for the WM_GETCONTROLNAME to those older applications?
Or maybe, you could identify the controls by some other, natively-available properties?
Other that that, as Raymond is saying, there isn't much you can do.
Can you modify the vb6 apps? if so in each form load event you could iterate me.controls and use the SetProp(ctrl.hwnd, "MYNAME:" & ctrl.name, 0) api to add the name to the window's own property list, then in your other app you can EnumProps(ctrl_HWND) looking for the one that begins with MYNAME: and parse out the value.
In other words, can I count on a control ID as a reliable identifier?
From some reading I've done, it sounds like .NET controls can have control IDs that change with every run, is this so for Win32 apps as well, or are they something that's hardcoded in the source?
The window/control in question is actually an Internet Explorer dialog if that helps.
In general win32 dialog resource IDs do not change when you run the app. However, they are internal implementation details and as such they are subject to change whenever an update (patch, service pack, major release) to the application is made.
In general all of IE's dialogs use hard-coded control IDs. There may be some that are dynamic. If you give the specific control I might be able to give you a better answer.
The answer is "it depends on the circumstances, but in the majority of programs, these will not change across multiple executions." In general, a control ID or resource ID will be the same across every execution of the same program.
In most implementations, resources are stored in the resource section of the PE executable and are assigned a resource ID within that data structure. Usually, the developer specifies the resource in a .rc file.
The exceptional case is via APIs such as CreateDialogIndirect() which allow IDs specified through the API at runtime. Such dynamic creation is uncommon, however. Consistency of resource IDs and control IDs is the expected condition, so even in the case of the CreateXXXIndirect() API, users of the API would be ill-advised to chose a varying ID.
Microsoft has spent years trying to deal with applications which embed assumptions about internal windows implementation details. One of the biggest causes of compatibility problems for IE8 was caused by applications which made assumptions about the window order of IE controls. When the UI was changed in IE8, the controls moved and a number of browser plugins broke hideously.
In general you should never ever make assumptions about controls in an application that you didn't write - your code WILL break in the future (not might break, will break).
As a general rule, no they don't change between runs. Control IDs are usually specified with a dialog template, and that's a static resource compiled into an .exe or .dll. Controls can also be created using a regular call to CreateWindow or CreateWindowEx. In such cases, the ID is usually a constant, but it could be anything in principal (even a random value).
In any case, if you're planning to muck around with a dialog in another application, then you are asking for trouble. Control IDs can and do change between different versions of a program.
It depends on the control. Controls that are dynamically created could have a different ID every time they are created. Controls based on static dialog resources will have static IDs.
Even if a dialog is backed by a dialog resource template, controls can be added dynamically at runtime and those controls could have dynamically generated IDs.
Depending on your intent, it may be acceptable to store and reuse it for future lookups or traces; but in general it isn't safe.
If the window is based upon a dialog template, items declared in the template generally don't change. It is perfectly safe under typical circumstances to use these as identifiers.
Sometimes the window class name or a portion of it can be used as an identifier instead, depending on the window host.
If the dialog is one of the standard prompts from internet explorer, you may want to use text stored in adjacent controls or the dialog caption as additional verification info (if localized versions of IE are not an issue). If the dialog is a window that embeds an instance of MSHTML/IE; none of these options may be viable- but you can use OLE accessibility to get at the document shown and then browse the DOM from there.