I would like to make a DM script acquiring camera images continuously, like VIEW mode. In my plan, the continious camera acquisition is start when START button located at an UIframe dialog is pressed; and a modeless dialog is also shown simultaneously. The continuous acuisition is stopped when the OK button located at the dialog is pressed. For this kind of scripts, I think a background thread is required. However, I don't have enough knowledge about such a background running.
It will be appreciated if you share some wisdom. Thank you very much in advance.
Continuous camera acquisition has actually nothing to do with background-threading in scripting, but rather requires setting the hardware into a continuous readout mode.
This functionality is no supported by the official scripting API up to now.
However, there exits an extended, object-oriented scripting API which gives a more in-depth control over cameras. You will need to be versatile in the object-oriented coding style of DM-scripting to use it, and you will need to get in contact with Gatan to possibly get access to this script API, as it is not officially supported.
You may want to use the support-request form on the Gatan homepage for this. (button on bottom of page)
Picking up on the kachigusa's comment to my other answer....
If you want to run a slow camera acquistion pseudo-continously, that is by performing repeated single readouts on a separate thread, then you may want to use a structure like the following:
class CMyBackgroundAction
{
number isRunning
CMyBackgroundAction(object self) { isRunning = 0; } // Initialisation in constructor
// methods to access flag from outside the object
void StopRunning(object self) { isRunning = 0; }
number GetIsRunning(object self) { return isRunning; }
// The stuff that should run in the background until the flag is set
void RunUntilBreak(object self)
{
Result("\n\n StartRunning")
isRunning = 1
while (isRunning)
{
Result( "\n New line..." )
sleep(0.5)
Result( "....done" )
}
Result("\n\n FINISHED")
}
}
class CmyDLG : UIframe
{
object backGroundRunObj
void OnStartStop( object self )
{
if ( !backGroundRunObj.GetIsRunning() )
{
// Nothing running in the background yet.
backGroundRunObj.StartThread( "RunUntilBreak" )
self.LookUpElement("StartStopButton").DLGTitle("Stop")
}
else
{
// It exists, so it is running. Just set the break-flag
backgroundRunObj.StopRunning();
self.LookUpElement("StartStopButton").DLGTitle("Start")
}
}
TagGroup BuildDLG( object self )
{
TagGroup dlg, dlgitems
dlg = DLGCreateDialog("StartStop",dlgItems)
dlgItems.DLGAddElement( DLGCreatePushButton( "Start", "OnStartStop" ).DLGIdentifier("StartStopButton" ) )
return dlg
}
Object Init(object self)
{
backGroundRunObj = Alloc(CMyBackgroundAction)
self.super.Init( self.BuildDLG() )
return self
}
}
Alloc(CmyDLG).Init().Display("Test")
There are of course several other constructs with a dialog governing a separate thread. This is just an example.
If you are going to start/stop something which performs in very regular intervals, but requires the main thread - another option would be to register/deregister periodic tasks from a dialog. Here is an example:
class CMyBackgroundAction
{
// The stuff that should run periodically
void RunUntilBreak(object self)
{
Result("\n Doing action #" + GetTime(1) )
}
}
class CmyDLG : UIframe
{
object backGroundRunObj
number taskID
void OnStartStop( object self )
{
if ( 0 == taskID )
{
// Start by registering the task ( every 0.5 sec)
taskID = AddMainThreadPeriodicTask(backGroundRunObj,"RunUntilBreak",0.5)
Result("\n STARTED ")
self.LookUpElement("StartStopButton").DLGTitle("Stop")
}
else
{
// Stop by removing the task
RemoveMainThreadTask( taskID )
Result("\n STOPPED ")
taskID = 0
self.LookUpElement("StartStopButton").DLGTitle("Start")
}
}
TagGroup BuildDLG( object self )
{
TagGroup dlg, dlgitems
dlg = DLGCreateDialog("StartStop",dlgItems)
dlgItems.DLGAddElement( DLGCreatePushButton( "Start", "OnStartStop" ).DLGIdentifier("StartStopButton" ) )
return dlg
}
Object Init(object self)
{
backGroundRunObj = Alloc(CMyBackgroundAction)
taskID = 0
return self.super.Init( self.BuildDLG() )
}
}
Alloc(CmyDLG).Init().Display("Test")
Related
To ensure thread-safety, I'm trying to find a generic cross-platform approach to
execute all delegates asynchronously in the main thread or ...
execute delegete in a background thread and pass result to the main one
Considering that console apps do not have synchronization context, I create new context when app is loading and then use one of the following methods.
Set and restore custom SC as described in Await, SynchronizationContext, and Console Apps article by Stephen Toub
Marshall all delegates to main thread using context.Post call as described in the article ExecutionContext vs SynchronizationContext by Stephen Toub
Using background thread with producer-consumer collection as described in Basic synchronization by Joe Albahari
Question
Ideas #1 and #2 set context correctly only if it's done synchronously. If they're called from inside Parallel.For(0, 100) then synchronization context starts using all threads available in a thread pool. Idea #3 always performs tasks within dedicated thread as expected, unfortunately, not in the main thread. Combining idea #3 with IOCompletionPortTaskScheduler, I can achieve asynchrony and single-threading, unfortunately, this approach will work only in Windows. Is there a way to combine these solutions to achieve requirements at the top of the post, including cross-platform.
Scheduler
public class SomeScheduler
{
public Task<T> RunInTheMainThread<T>(Func<T> action, SynchronizationContext sc)
{
var res = new TaskCompletionSource<T>();
SynchronizationContext.SetSynchronizationContext(sc); // Idea #1
sc.Post(o => res.SetResult(action()), null); // Idea #2
ThreadPool.QueueUserWorkItem(state => res.SetResult(action())); // Idea #3
return res.Task;
}
}
Main
var scheduler = new SomeScheduler();
var sc = SynchronizationContext.Current ?? new SynchronizationContext();
new Thread(async () =>
{
var res = await scheduler.ExecuteAsync(() => 5, sc);
});
You can use the lock/Monitor.Pulse/Monitor.Wait and a Queue
I know the title says lock-free. But my guess is that you want the UI updates to occur outside the locks or worker threads should be able to continue working without having to wait for main thread to update the UI (at least this is how I understand the requirement).
Here the locks are never during the producing of items, or updating the UI. They are held only during the short duration it takes to enqueue/dequeue item in the queue.
using System;
using System.Collections.Generic;
using System.Threading;
using System.Threading.Tasks;
using static System.Threading.Thread;
namespace ConsoleApp1
{
internal static class Program
{
private class WorkItem
{
public string SomeData { get; init; }
}
private static readonly Queue<WorkItem> s_workQueue = new Queue<WorkItem>();
private static void Worker()
{
var random = new Random();
// Simulate some work
Sleep(random.Next(1000));
// Produce work item outside the lock
var workItem = new WorkItem
{
SomeData = $"data produced from thread {CurrentThread.ManagedThreadId}"
};
// Acquire lock only for the short time needed to add the work item to the stack
lock (s_workQueue)
{
s_workQueue.Enqueue(workItem);
// Notify the main thread that a new item is added to the queue causing it to wakeup
Monitor.Pulse(s_workQueue);
}
// work item is now queued, no need to wait for main thread to finish updating the UI
// Continue work here
}
private static WorkItem GetWorkItem()
{
// Acquire lock only for the duration needed to get the item from the queue
lock (s_workQueue)
{
WorkItem result;
// Try to get the item from the queue
while (!s_workQueue.TryDequeue(out result))
{
// Lock is released during Wait call
Monitor.Wait(s_workQueue);
// Lock is acquired again after Wait call
}
return result;
}
}
private static void Main(string[] args)
{
const int totalTasks = 10;
for (var i = 0; i < totalTasks; i++)
{
_ = Task.Run(Worker);
}
var remainingTasks = totalTasks;
// Main loop (similar to message loop)
while (remainingTasks > 0)
{
var item = GetWorkItem();
// Update UI
Console.WriteLine("Got {0} and updated UI on thread {1}.", item.SomeData, CurrentThread.ManagedThreadId);
remainingTasks--;
}
Console.WriteLine("Done");
}
}
}
Update
Since you don't want to have the main thread Wait for an event, you can change the code as follows:
private static WorkItem? GetWorkItem()
{
// Acquire lock only for the duration needed to get the item from the queue
lock (s_workQueue)
{
// Try to get the item from the queue
s_workQueue.TryDequeue(out var result);
return result;
}
}
private static void Main(string[] args)
{
const int totalTasks = 10;
for (var i = 0; i < totalTasks; i++)
{
_ = Task.Run(Worker);
}
var remainingTasks = totalTasks;
// Main look (similar to message loop)
while (remainingTasks > 0)
{
var item = GetWorkItem();
if (item != null)
{
// Update UI
Console.WriteLine("Got {0} and updated UI on thread {1}.", item.SomeData, CurrentThread.ManagedThreadId);
remainingTasks--;
}
else
{
// Queue is empty, so do some other work here then try again after the work is done
// Do some other work here
// Sleep to simulate some work being done by main thread
Thread.Sleep(100);
}
}
Console.WriteLine("Done");
}
The problem in the above solution is that the Main thread should do only part of the work it is supposed to do, then call GetWorkItem to check if the queue has something, before resuming whatever it was doing again. It is doable if you can divide that work into small pieces that don't take too long.
I don't know if my answer here is what you want. What do you imagine the main thread would be doing when there are no work items in the queue?
if you think it should be doing nothing (i.e. waiting) then the Wait solution should be fine.
If you think it should be doing something, then may be that work it should be doing can be queued as a Work item as well.
In tree widget I have following signal connected:
connect(mTreeWidget, SIGNAL(itemClicked(QTreeWidgetItem*, int)),
SLOT(onItemClicked(QTreeWidgetItem*, int)));
where onItemClicked() slot is following:
void WidgetBox::onItemClicked(QTreeWidgetItem *item, int )
{
int index = getPageIndex(item);
setCurrentIndex(index);
}
int WidgetBox::getPageIndex(QTreeWidgetItem *item)
{
if (!item) return -1;
QTreeWidgetItem *parent = item->parent();
if(parent) // Parent is top level item
{
return mTreeWidget->indexOfTopLevelItem(parent);
}
else // Current item is top level
{
return item->treeWidget()->indexOfTopLevelItem(item);
}
}
void WidgetBox::setCurrentIndex(int index)
{
if (index != currentIndex() && checkIndex(index))
{
mTreeWidget->setCurrentItem(mTreeWidget->topLevelItem(index));
emit currentIndexChanged(index);
}
}
However I can't catch itemClicked() signal and onItemClicked() never executed because top level items has push button widget (set with setItemWidget() method) which intercepts mouse event and child items contain container widgets which may have any widget combinations in them.
Is there a good method here to invoke this itemClicked() signal for both top level and child items of tree widget?
installEventFilter() for all widgets found in an item by something like following:QList<QWidget *> widgets = parentWidget.findChildren<QWidget *>();?
Or establish mouse event propagation somehow?
QCoreApplication::postEvent()?
How to organize such process better so all widgets process mouse event as they need to and TreeWidget issue SIGNAL(itemClicked()) as well?
Full sources to reproduce: https://github.com/akontsevich/WidgetBox
So solution is simple and following - just re-send void itemClicked(QTreeWidgetItem *item, int column); signal in PageEventFilter:
PageEventFilter::PageEventFilter(QObject *parent, QTreeWidgetItem *item)
: QObject(parent)
, mItem(item)
{
connect(this, SIGNAL(itemClicked(QTreeWidgetItem*,int)),
mItem->treeWidget(), SIGNAL(itemClicked(QTreeWidgetItem*,int)));
}
bool PageEventFilter::eventFilter(QObject *obj, QEvent *event)
{
if (event->type() == QEvent::MouseButtonPress)
{
// Resend signal to QTreeWidget
emit itemClicked(mItem, 0);
return false; // Send event to the object (do not filter it)
}
else
{
// standard event processing
return QObject::eventFilter(obj, event);
}
}
P.S. Will leave previous answer as well if somebody needs code or idea.
Created event filter like this:
class PageEventFilter : public QObject
{
Q_OBJECT
public:
PageEventFilter(QObject *parent, QTreeWidgetItem *item);
protected:
bool eventFilter(QObject *obj, QEvent *event);
private:
QTreeWidgetItem *mItem;
};
bool PageEventFilter::eventFilter(QObject *obj, QEvent *event)
{
if (event->type() == QEvent::MouseButtonPress)
{
// Emitate mouse event for parent QTreeWidget
QMouseEvent *oldEvent = (QMouseEvent *)event;
QRect itemRect = mItem->treeWidget()->visualItemRect(mItem);
QPointF mousePos(itemRect.x() + 1, itemRect.y() + 1);
QMouseEvent *newEvent = new QMouseEvent(oldEvent->type(),
mousePos,
oldEvent->button(),
oldEvent->buttons(),
oldEvent->modifiers());
QCoreApplication::postEvent(mItem->treeWidget(), newEvent);
return false; // Sent event to the object (do not filter it)
}
else
{
// standard event processing
return QObject::eventFilter(obj, event);
}
}
Install it to a button in tree widget item. It creates and sends (imitates) mouse event to tree widget, however tree widget still does not send itemClicked(QTreeWidgetItem*, int) signal. What is the problem could be? Wrong mouse pos? Tried to send event to viewport() - no luck as well. Any way to solve this?
I really want to know how do I can update the position of the user in the map while the UWP app was running in bakground
Here is my code right now
private async void PinPoints()
{
//Pin point to the map
Windows.Devices.Geolocation.Geopoint position = await Library.Position();
double lat = position.Position.Latitude;
double lon = position.Position.Longitude;
//Geoposition alttest = await Library.Temp();
//alt = alttest.Coordinate.Altitude;
DependencyObject marker = Library.Marker(""
//+ Environment.NewLine + "Altitude " + alt
);
Display.Children.Add(marker);
Windows.UI.Xaml.Controls.Maps.MapControl.SetLocation(marker, position);
Windows.UI.Xaml.Controls.Maps.MapControl.SetNormalizedAnchorPoint(marker, new Point(0.5, 0.5));
Display.LandmarksVisible = true;
Display.ZoomLevel = 16;
Display.Center = position;
}
This function will pinpoint the current location for me but it will do only when user open this page due to I've put it in the public Map() {}
Current : Get the location when open map page and when I walk the map still be the same place
What I want : The position keep changing while I move on and also run on background (If application is close location data still changed)
Is there any code to solve this location problem if I have to add code where should I fix and what should I do?
Additional now I perform the background (Not sure is it work or not) by create the Window Runtime Component (Universal) with class like this
*I already put this project as the reference of the main one
namespace BackgroundRunning
{
public sealed class TaskBG : IBackgroundTask
{
BackgroundTaskDeferral _deferral = null;
Accelerometer _accelerometer = null;
Geolocator _locator = new Geolocator();
public void Run(IBackgroundTaskInstance taskInstance)
{
_deferral = taskInstance.GetDeferral();
try
{
// force gps quality readings
_locator.DesiredAccuracy = PositionAccuracy.High;
taskInstance.Canceled += taskInstance_Canceled;
_accelerometer = Windows.Devices.Sensors.Accelerometer.GetDefault();
_accelerometer.ReportInterval = _accelerometer.MinimumReportInterval > 5000 ? _accelerometer.MinimumReportInterval : 5000;
_accelerometer.ReadingChanged += accelerometer_ReadingChanged;
}
catch (Exception ex)
{
// Add your chosen analytics here
System.Diagnostics.Debug.WriteLine(ex);
}
}
void taskInstance_Canceled(IBackgroundTaskInstance sender, BackgroundTaskCancellationReason reason)
{
_deferral.Complete();
}
async void accelerometer_ReadingChanged(Windows.Devices.Sensors.Accelerometer sender, Windows.Devices.Sensors.AccelerometerReadingChangedEventArgs args)
{
try
{
if (_locator.LocationStatus != PositionStatus.Disabled)
{
try
{
Geoposition pos = await _locator.GetGeopositionAsync();
}
catch (Exception ex)
{
if (ex.HResult != unchecked((int)0x800705b4))
{
System.Diagnostics.Debug.WriteLine(ex);
}
}
}
}
catch (Exception ex)
{
System.Diagnostics.Debug.WriteLine(ex);
}
}
public void Dispose()
{
if (_accelerometer != null)
{
_accelerometer.ReadingChanged -= accelerometer_ReadingChanged;
_accelerometer.ReportInterval = 0;
}
}
}
}
Your Solution :
Make 3 projects in your solution.
1> Background Task "references App_Code"
2> App_Code "contains calculations,mostly Backend Code"
3> Map(Main Project) "references App_Code"
Register a background Task to your project and specify the time interval after which it should run again
Scenario 1> App Open,User Requests Update
Trigger Your background Task from code behind.
Scenario 2> App Closed,Not Being Used
Run your background task!
So basically keep your backgroundTask simple(make it a class in whose run method you just call the proper App_Code Classes Method) and all calculations that you want to happen in the background keep them in App_Code. Also, if I am no wrong the minimum interval between which a background Task is triggered by itself cannot be set below 15 minutes.
For real-time you could look at SignalR ( can't help any further here)
I'm trying to use an NSTextView to display high frequency updated logs but it's locking up the UI.
All logs are created in a different queue and only the minimal required updates occur on the main queue so I'm not sure how I can make this more efficient. I feel like I'm getting a leak because memory increases slowly but steadily over time.
If I turn off locking the app runs with minimal stable memory usage.
This is the function which updates the UI on the main queue. LogMessages are stored in an array which is kept at a fixed length on a 'background' queue and the most recent item is pushed to this function.
func logsChanged(message: LogMessage) {
let formatter = NSDateFormatter()
formatter.dateFormat = "yyyy-MM-dd HH:mm:ss.SSS"
let dateString = formatter.stringFromDate(message.time)
let log = "[\(dateString)] " + message.text + "\n"
let attrs = [NSFontAttributeName : NSFont(name: "Courier New", size: 12.0)!]
let attributedLog = NSMutableAttributedString(string: log, attributes: attrs)
if let textStorage = self.logging.textStorage {
textStorage.beginEditing()
textStorage.appendAttributedString(attributedLog)
if textStorage.length > LogMaxSize {
textStorage.deleteCharactersInRange(NSMakeRange(0, attributedLog.length))
}
textStorage.endEditing()
self.logging.scrollToEndOfDocument(nil)
}
}
This is the code which calls UI update. I'm even slowing down updates by limiting to 20/sec
// do not let anything else write to the logs until this has completed.
dispatch_barrier_async(logAddQueue) {
let message = LogMessage(type: type, time: NSDate(), text: text)
if self.logMessages.count > self.maxMessages {
self.logMessages.removeLast()
}
self.logMessages.append(message)
if type == LogMessageType.Send || type == LogMessageType.Receive {
if NSDate().timeIntervalSinceDate(self.lastUpdatedUI) >= 0.05 {
// updates interface so run on the main queue
dispatch_async(GlobalMainQueue) {
self.delegate?.logsChanged(message)
}
self.lastUpdatedUI = NSDate()
}
} else {
// updates interface so run on the main queue
dispatch_async(GlobalMainQueue) {
self.delegate?.logsChanged(message)
}
self.lastUpdatedUI = NSDate()
}
}
Any suggestions?
My app may create / delete thousands of managed objects while running. I have used secondary NSManagedObjectContexts(MOCs) with NSPrivateQueueConcurrencyType and NSOperations to make the app more responsive and most parts work well. But when I pressed ⌘Q and if the number of unsaved objects are large, the app hangs for quite a while before the window closes (the beach ball keeps on rotating...).
How to make the window disappear immediately, before the save of the MOC?
I tried to insert window.close() in applicationShouldTerminate in the AppDelegate, but it has no effect.
My code for deletion is nothing special, except the hierachy is really large. Something like
let items = self.items as! Set<Item>
Group.removeItems(items)
for i in items {
self.managedObjectContext?.deleteObject(i)
}
Item is a hierarchic entity. Group has a one-to-many relationship to items.
The removeItems is generated by CoreData with #NSManaged.
Many thanks.
Updates
I tried the following code, the save still blocks the UI.
#IBAction func quit(sender: AnyObject) {
NSRunningApplication.currentApplication().hide()
NSApp.terminate(sender)
}
func applicationShouldTerminate(sender: NSApplication) -> NSApplicationTerminateReply
{
let op = NSBlockOperation { () -> Void in
do {
try self.managedObjectContext.save()
} catch {
print("error")
}
NSOperationQueue.mainQueue().addOperationWithBlock({ () -> Void in
NSApp.replyToApplicationShouldTerminate(true)
})
}
op.start()
return .TerminateLater
}
This doesn't make the window close first, when the amount of created / deleted managed objects is large.
Then I changed to the following, as suggested by #bteapot. Still has no effect. The window still won't close immediately.
#IBAction func quit(sender: AnyObject) {
NSRunningApplication.currentApplication().hide()
NSApp.terminate(sender)
}
func applicationShouldTerminate(sender: NSApplication) -> NSApplicationTerminateReply {
let op = NSBlockOperation { () -> Void in
self.managedObjectContext.performBlock({ () -> Void in
do {
try self.managedObjectContext.save()
} catch {
print("errr")
}
})
NSOperationQueue.mainQueue().addOperationWithBlock({ () -> Void in
NSApp.replyToApplicationShouldTerminate(true)
})
}
dispatch_async ( dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_DEFAULT, 0),
{() -> Void in
op.start()
})
return .TerminateLater
}
Finally I sort of solved the problem, though the UI is still blocked sometimes, even with the same test data.
The approach used can be found here: https://blog.codecentric.de/en/2014/11/concurrency-coredata/ , Core Data background context best practice , https://www.cocoanetics.com/2012/07/multi-context-coredata/
First I made a backgroundMOC with .PrivateQueueConcurrencyType
lazy var backgroundMOC : NSManagedObjectContext = {
let coordinator = self.persistentStoreCoordinator
let moc = NSManagedObjectContext(concurrencyType: .PrivateQueueConcurrencyType)
moc.persistentStoreCoordinator = coordinator
moc.undoManager = nil
return moc
}()
Then made it prent of the original moc.
lazy var managedObjectContext: NSManagedObjectContext = {
var managedObjectContext = NSManagedObjectContext(concurrencyType: .MainQueueConcurrencyType)
// managedObjectContext.persistentStoreCoordinator = coordinator
managedObjectContext.parentContext = self.backgroundMOC
managedObjectContext.undoManager = nil
return managedObjectContext
}()
Two methods for the save.
func saveBackgroundMOC() {
self.backgroundMOC.performBlock { () -> Void in
do {
try self.backgroundMOC.save()
NSApp.replyToApplicationShouldTerminate(true)
} catch {
print("save error: bg")
}
}
}
func saveMainMOC() {
self.managedObjectContext.performBlock { () -> Void in
do {
try self.managedObjectContext.save()
self.saveBackgroundMOC()
} catch {
print("save error")
}
}
}
Change the applicationShouldTerminate() to
func applicationShouldTerminate(sender: NSApplication) -> NSApplicationTerminateReply {
if !managedObjectContext.commitEditing() {
NSLog("\(NSStringFromClass(self.dynamicType)) unable to commit editing to terminate")
return .TerminateCancel
}
if !managedObjectContext.hasChanges {
return .TerminateNow
}
saveMainMOC()
return .TerminateLater
}
The reason it was so slow was I was using NSXMLStoreType instead of NSSQLiteStoreType.
Quitting an application might take a while since it will first empty the processes in queue.
Do you want immediate quit discarding everything in the Parent or children MOCs? But this will result in data loss.
If you have multi window application then, then close the window only but not quit the app.
Also thousands of entry should not take longer than 5 seconds to get processed and saved, if you have managed it properly. There could be some loopholes in your code, try to optimize using Instruments, CoreData profiler tool that would help you to understand the amount of time it is eating up.
To hide the window you can use the below, and in background all the coredata processing will happen, and once everything is done the app will terminate.
[self.window orderOut:nil];