Cannot update UI with Invoke() and BeginInvoke() in C# - invoke

My main process:
public void quoteStartReceive()
{
Thread thdWrite = new Thread(new ThreadStart(DoParseGUIDisplay));
thdWrite.IsBackground = true;
thdWrite.Start();
}
My thread function:
void DoParseGUIDisplay()
{
for (int i = 0; i < 1024; i++)
{
if (myQueue.Count > 0)
{
string strOut = myQueue.Dequeue().ToString();
Tick tick = new Tick(strOut);
if (tick.m_last != "")
{
string msg = "Update Text";
if (this.textBox1.InvokeRequired)
{
this.textBox1.BeginInvoke((MethosInvoker)delegate () {this.textBox1.Text = msg; };
}
else
{
this.textBox1.Text = msg;
}
}
}
}
}
No matter I tried to use Invoke() or BeginInvoke(), I cannot update the text in textBox1.
I also tried another way:
public delegate void UpdateTextCallback(string text);
It still cannot help me to update my textBox1.
Help me to find out what stuff I missed. Thanks.

put an argument before the thread:
Application.DoEvents();
and it should be put after the m_last was updated.

Related

managed c++ classes crash in create_task

Basically, what is happening is when trying to change a variable of a Managed class (UWP), it crashes. Additionally, it seems to only crash if I try modifying a variable which is created with the app namespace. In other words, if I create a new namspace and managed class, I can modify variables just fine.
void ASynTaskCategories(void)
{
concurrency::task_completion_event<Platform::String^> taskDone;
MainPage^ rootPage = this;
UberSnip::HELPER::ASYNC_RESPONSE^ responseItem = ref new UberSnip::HELPER::ASYNC_RESPONSE();
create_task([taskDone, responseItem, rootPage]() -> Platform::String^
{
//UBERSNIP API 0.4
UberSnip::UBERSNIP_CLIENT* UberSnipAPI = new UberSnip::UBERSNIP_CLIENT();
UberSnipAPI->Http->RequestURL = "http://api.ubersnip.com/categories.php";
UberSnipAPI->Http->request();
taskDone.set(UberSnipAPI->Client->BodyResponse);
responseItem->Body = UberSnipAPI->Client->BodyResponse;
return "(done)";
}).then([taskDone, responseItem, rootPage](Platform::String^ BodyResponse) {
Platform::String^ BR = responseItem->Body;
cJSON* cats = cJSON_Parse(_string(BR));
cats = cats->child;
int *cat_count = new int(cJSON_GetArraySize(cats));
rootPage->Categories->Clear();
GENERIC_ITEM^ all_item = ref new GENERIC_ITEM();
all_item->Title = "All";
rootPage->Categories->Append(all_item);
for (int i = 0; i < *cat_count; i++) {
cJSON* cat = new cJSON();
cat = cJSON_GetArrayItem(cats, i);
string *track_title = new string(cJSON_GetObjectItem(cat, "name")->valuestring);
GENERIC_ITEM gitem;
gitem.Title = "Hi";
GENERIC_ITEM^ ubersnipCategory = ref new GENERIC_ITEM();
ubersnipCategory->Title = _String(*track_title);
rootPage->Categories->Append(ubersnipCategory);
}
});
This does not crash
UberSnipAPI->Http->RequestURL = "http://api.ubersnip.com/categories.php";
but this one does
all_item->Title = "All";
I am almost positive it has something to do with it being in the apps default namespace and it being accessed outside of the main thread ... At least that's what it seems like as that's really the only difference besides the actual class.
This is what GENERIC_ITEM looks like.
[Windows::UI::Xaml::Data::Bindable]
public ref class GENERIC_ITEM sealed {
private:
Platform::String^ _title = "";
Platform::String^ _description;
Windows::UI::Xaml::Media::ImageSource^ _Image;
event PropertyChangedEventHandler^ _PropertyChanged;
void OnPropertyChanged(Platform::String^ propertyName)
{
PropertyChangedEventArgs^ pcea = ref new PropertyChangedEventArgs(propertyName);
_PropertyChanged(this, pcea);
}
public:
GENERIC_ITEM() {
};
property Platform::String^ Title {
Platform::String^ get() {
return this->_title;
}
void set(Platform::String^ val) {
this->_title = val;
OnPropertyChanged("Title");
}
}
property Platform::String^ Description {
Platform::String^ get() {
return this->_description;
}
void set(Platform::String^ val) {
this->_description = val;
OnPropertyChanged("Description");
}
}
void SetImage(Platform::String^ path)
{
Windows::Foundation::Uri^ uri = ref new Windows::Foundation::Uri(path);
_Image = ref new Windows::UI::Xaml::Media::Imaging::BitmapImage(uri);
}
};
I know there is no issue with the class because it works perfectly fine if I run this exact same code on the initial thread.
Any suggestions? Thanks! :D
Figured it out after several hours! In case someone else is having this issue, all you must do is use the Dispatcher to run code on the UI that is not available outside of the UI thread.
void loadCats(UberSnip::HELPER::ASYNC_RESPONSE^ responseItem, MainPage^ rootPage) {
Platform::String^ BR = responseItem->Body;
cJSON* cats = cJSON_Parse(_string(BR));
cats = cats->child;
int *cat_count = new int(cJSON_GetArraySize(cats));
rootPage->Categories->Clear();
GENERIC_ITEM^ all_item = ref new GENERIC_ITEM();
all_item->Title = "All";
rootPage->Categories->Append(all_item);
for (int i = 0; i < *cat_count; i++) {
cJSON* cat = new cJSON();
cat = cJSON_GetArrayItem(cats, i);
string *track_title = new string(cJSON_GetObjectItem(cat, "name")->valuestring);
GENERIC_ITEM gitem;
gitem.Title = "Hi";
GENERIC_ITEM^ ubersnipCategory = ref new GENERIC_ITEM();
ubersnipCategory->Title = _String(*track_title);
rootPage->Categories->Append(ubersnipCategory);
}
}
void ASyncTaskCategories(void)
{
concurrency::task_completion_event<Platform::String^> taskDone;
MainPage^ rootPage = this;
UberSnip::HELPER::ASYNC_RESPONSE^ responseItem = ref new UberSnip::HELPER::ASYNC_RESPONSE();
auto dispatch = CoreWindow::GetForCurrentThread()->Dispatcher;
auto op2 = create_async([taskDone, responseItem, rootPage, dispatch] {
return create_task([taskDone, responseItem, rootPage, dispatch]() -> Platform::String^
{
//UBERSNIP API 0.4
UberSnip::UBERSNIP_CLIENT* UberSnipAPI = new UberSnip::UBERSNIP_CLIENT();
UberSnipAPI->Http->RequestURL = "http://api.ubersnip.com/categories.php";
try{
UberSnipAPI->Http->request();
}
catch (...) {
printf("Error");
}
int err = UberSnip::UTILS::STRING::StringToAscIIChars(UberSnipAPI->Client->BodyResponse).find("__api_err");
if (err < 0) {
if (UberSnip::UTILS::STRING::StringToAscIIChars(UberSnipAPI->Client->BodyResponse).length() < 3) {
return;
}
}
taskDone.set(UberSnipAPI->Client->BodyResponse);
responseItem->Body = UberSnipAPI->Client->BodyResponse;
dispatch->RunAsync(Windows::UI::Core::CoreDispatcherPriority::High, ref new Windows::UI::Core::DispatchedHandler([=]()
{
rootPage->loadCats(responseItem, rootPage);
}));
for (int i = 0; i < 100;) {
}
return "(done)";
}).then([taskDone, responseItem, rootPage](Platform::String^ BodyResponse) {
});
});
//rootPage->loadCats(responseItem, rootPage);
}

MPAndroidChart multiple real time line chart - entries not added at correct xIndex

I modified a little bit the Real Time Line Chart example to show two LineChart like the code below.
Problem: the ViewPort is moving incorrectly does not work properly. It is moving much faster than real points (Entry) are added. In other words, the Entry get added in the wrong place and the ViewPort keeps moving right uncontrollably. It should move right only gradually as each Entry is added. Please help me to fix this one.
private int year = 2015;
SimpleDateFormat sdf = new SimpleDateFormat("hh:mm:ss.SSS");
private int[] mColors = {Color.BLUE,Color.YELLOW,Color.CYAN,Color.MAGENTA,Color.GREEN};
private int mCurrentColorIndex = 0;
private synchronized void addEntry(String id, float value) {
LineData data = mChart.getData();
if (data != null) {
ILineDataSet set = data.getDataSetByLabel(id,true);
//ILineDataSet set1 = data.getDataSetByIndex(0);
if (set == null) {
set = createSet(id, mColors[(mCurrentColorIndex++)%mColors.length ]);
data.addDataSet(set);
}
String xValue = sdf.format(new Date());
// add a new x-value first
data.addXValue(xValue);
set.addEntry(new Entry(value, set.getEntryCount(), 0));
// let the chart know it's data has changed
mChart.notifyDataSetChanged();
// limit the number of visible entries
mChart.setVisibleXRangeMaximum(30);
// mChart.setVisibleYRange(30, AxisDependency.LEFT);
// move to the latest entry
mChart.moveViewToX(data.getXValCount() - 31);
// this automatically refreshes the chart (calls invalidate())
// mChart.moveViewTo(data.getXValCount()-7, 55f,
// AxisDependency.LEFT);
}
}
private LineDataSet createSet(String label, int color) {
LineDataSet set = new LineDataSet(null, label);
set.setAxisDependency(AxisDependency.LEFT);
set.setColor(color);
set.setCircleColor(Color.WHITE);
set.setLineWidth(2f);
set.setCircleRadius(4f);
set.setFillAlpha(65);
set.setFillColor(color);
set.setHighLightColor(Color.rgb(244, 117, 117));
set.setValueTextColor(Color.WHITE);
set.setValueTextSize(9f);
set.setDrawValues(false);
return set;
}
private void feedMultiple() {
new Thread(new Runnable() {
#Override
public void run() {
for(int i = 0; i < 50; i++) {
runOnUiThread(new Runnable() {
#Override
public void run() {
addEntry("name1", (float)(Math.random() * 40) + 30f);
}
});
try {
Thread.sleep(35);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
}).start();
new Thread(new Runnable() {
#Override
public void run() {
for(int i = 0; i < 100; i++) {
runOnUiThread(new Runnable() {
#Override
public void run() {
addEntry("name2", (float)(Math.random() * 40) + 30f);
}
});
try {
Thread.sleep(35);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
}).start();
}
Resolved in the comments by #Tony with the following change:
//set.addEntry(new Entry(value, set.getEntryCount(), 0)); //incorrect
set.addEntry(new Entry(value, data.getXValCount(), 0)); //correct
Explanation:
A LineData is composed of multiple ILineDataSet:
set.getEntryCount() returns the number of y-values in one DataSet and is effectively equivalent to yvals.size().
data.getXValCount() returns the total number of x-values the ChartData object represents.
Since we wish to add an entry at the last x-index, we should use data.getXValCount()
NOTE:
In MPAndroidCharts 3.0.1 data.getXValCount() is no longer present. You can instead use data.getXMax() + 1 instead.

Treeviews QueueDraw doesn't render current row?

I'm working with a treeview, which contains several columns, also one displaying a pixbuf, if audio is playing or paused. If the user double clicks on one row, audio playback starts and the row needs to be rerendered in order to display the pixbuf icon. I used QueueDraw for this, but that does only work, if the cursor leaves the current row. How can I update the pixbuf directly?
CODE:
protected void trvMainCuesheetRowActivated (object o, RowActivatedArgs args)
{
log.debug("trvMainCuesheetRowActivated called");
TreeIter ti = TreeIter.Zero;
this.lsCuesheetData.GetIter(out ti,args.Path);
if (this.lsCuesheetData.GetValue(ti,0) != null)
{
Track tCurTrack = (Track)this.lsCuesheetData.GetValue(ti,0);
if (this.objProgram.getAudioManager().getPlayState() == AudioCuesheetEditor.AudioBackend.PlayState.Stopped)
{
this.objProgram.getAudioManager().play(tCurTrack);
this.refresh();
}
else
{
if (this.objProgram.getAudioManager().getPlayState() == AudioCuesheetEditor.AudioBackend.PlayState.Playing)
{
this.objProgram.getAudioManager().seek(tCurTrack);
this.refresh();
}
}
}
}
private void renderPlaying(TreeViewColumn _tvcColumn, CellRenderer _crCell, TreeModel _tmModel, TreeIter _tiIter)
{
Track tCurTrack = (Track)_tmModel.GetValue (_tiIter, 0);
//Just display an icon, if we are playing
if (this.objProgram.getAudioManager().getPlayState() == AudioCuesheetEditor.AudioBackend.PlayState.Playing)
{
if (this.objProgram.getAudioManager().getCurrentlyPlayingTrack() == tCurTrack)
{
Gdk.Pixbuf icon = this.RenderIcon(Stock.MediaPlay, IconSize.SmallToolbar, null);
(_crCell as CellRendererPixbuf).Pixbuf = icon;
}
else
{
(_crCell as CellRendererPixbuf).Pixbuf = null;
}
}
else
{
if (this.objProgram.getAudioManager().getPlayState() == AudioCuesheetEditor.AudioBackend.PlayState.Paused)
{
if (this.objProgram.getAudioManager().getCurrentlyPlayingTrack() == tCurTrack)
{
Gdk.Pixbuf icon = this.RenderIcon(Stock.MediaPause, IconSize.SmallToolbar, null);
(_crCell as CellRendererPixbuf).Pixbuf = icon;
}
else
{
(_crCell as CellRendererPixbuf).Pixbuf = null;
}
}
else
{
(_crCell as CellRendererPixbuf).Pixbuf = null;
}
}
}
//Purpose: Function used to refresh the MainWindow depending on new options set.
public void refresh()
{
//QueueDraw is needed since it fires a signal to cellrenderers to update
this.trvMainCuesheet.QueueDraw();
this.sbMainWindow.Visible = this.objProgram.getObjOption().getBShowStatusbar();
this.mwToolbar.Visible = this.objProgram.getObjOption().getBToolbarVisible();
}
Greetings
Sven
Found the error myself.
this.objProgram.getAudioManager().getCurrentlyPlayingTrack()
didn't always return a track, where I expected one, so the renderer worked right. Bug is fixed, thanks anyway ;).

How to write a custom FindElement routine in Selenium?

I'm trying to figure out how to write a custom FindElement routine in Selenium 2.0 WebDriver. The idea would be something like this:
driver.FindElement(By.Method( (ISearchContext) => {
/* examine search context logic here... */ }));
The anonymous method would examine the ISearchContext and return True if it matches; False otherwise.
I'm digging through the Selenium code, and getting a bit lost. It looks like the actual By.* logic is carried out server-side, not client side. That seems to be complicating matters.
Any suggestions?
I do a multi-staged search. I have a method that performs a try catch and then a method that gets the element. In theory you could do a try catch until instead of this way but I like this way better because of my setup.
public bool CheckUntil(IWebDriver driver, string selectorType, string selectorInfo)
{
int Timer = 160;
bool itemFound = false;
for (int i = 0; i < Timer; i++)
if(itemFound)
{
i = 0
}
else
{
Thread.Sleep(500);
if(selectorType.ToLower() == "id" && TryCatch(driver, selectorType, selectorInfo))
{
if(driver.FindElement(By.Id(selectorInfo).Displayed)
{
itemFound = true;
}
}
else if(selectorType.ToLower() == "tagname" && TryCatch(driver, selectorType, selectorInfo))
{
if(driver.FindElement(By.TagName(selectorInfo).Displayed)
{
itemFound = true;
}
}
}
return itemFound;
}
Here's my try catch method you can add as many different types as you want id, cssselector, xpath, tagname, classname, etc.
public bool TryCatch(IWebDriver driver, string selectorType, string selectorInfo)
{
bool ElementFound = false;
try
{
switch(selectorType)
{
case "id":
driver.FindElement(By.Id(selectorInfo);
break;
case "tagname":
driver.FindElement(By.TagName(selectorInfo);
break;
}
ElementFound = truel
}
catch
{
ElementFound = false;
}
return ElementFound;
}
Ok, I figured out how to do this. I'm leveraging driver.ExecuteScript() to run custom js on the webdriver. It looks a bit like this:
function elementFound(elem) {
var nodeType = navigator.appName == ""Microsoft Internet
Explorer"" ? document.ELEMENT_NODE : Node.ELEMENT_NODE;
if(elem.nodeType == nodeType)
{
/* Element identification logic here */
}
else { return false; }
}
function traverseElement(elem) {
if (elementFound(elem) == true) {
return elem;
}
else {
for (var i = 0; i < elem.childNodes.length; i++) {
var ret = traverseElement(elem.childNodes[i]);
if(ret != null) { return ret; }
}
}
}
return traverseElement(document);

How do I use a variable defined in an 'if' statement?

public class Help
{
public static void main (String[] args)
{
if (index = 1)
{
String greeting = "hello";
}
else
{
String greeting = "goodbye";
}
}
callAMethod(greeting);
}
When I define the String within the if statement I get a 'cannot find symbol' error. How can I get around this and still be able to create a string depending upon an above condition?
How about
public static void main (String[] args){
String greeting;
if( index == 1){
greeting = "hello";
}else{
greeting = "goodbye";
}
}
callAMethod(greeting);
}
Declare it outside the scope -
String greeting = "goodbye";
if( index == 1)
{
greeting = "hello";
}
callAMethod(greeting);
Declare the variable outside of the if block.
Right before the if statement, you can say String greeting = "";
Then, inside the if and else blocks, you say greeting = "hello"; and so on.
So you have separated declaring the variable from assigning the value to it.
You can define the greeting variable before the statement:
String greeting;
if (index == 1) {
greeting = "hello";
} else {
greeting = "bye";
}
System.out.println(greeting);