I am facing some troubles when processing the NM_CUSTOMDRAW message on 64 bit machines. We have an CListCtrl derived class with a CHeaderCtrl derived header (linked via PreSubclassWindow).
In the CHeader derived class we do some custom painting. This works for a 32bit build. But when I build a 64bit variant the drawstage remains CDDS_PREPAINT.
So I am posting here to get some help on this issue. I tried a lot of combinations of result values, drawstage handling in the OnCustomDraw ... but all of them still receive only the CDDS_PREPAINT drawstage.
Here you have my current testcode for OnCustomDraw:
void CListViewCtrlExHeader::OnCustomDraw(NMHDR* pNMHDR, LRESULT* pResult)
{
LPNMCUSTOMDRAW lpnmcd = (LPNMCUSTOMDRAW )pNMHDR;
*pResult = CDRF_DODEFAULT;
if (lpnmcd)
{
switch (lpnmcd->dwDrawStage)
{
case CDDS_PREERASE:
{
TRACE(_T("CDDS_PREERASE\n"));
*pResult |= CDRF_NOTIFYPOSTERASE;
break;
}
case CDDS_POSTERASE:
{
TRACE(_T("CDDS_POSTERASE\n"));
break;
}
case CDDS_PREPAINT:
{
TRACE(_T("CDDS_PREPAINT\n"));
*pResult |= CDRF_NOTIFYPOSTPAINT;
break;
}
case CDDS_POSTPAINT:
{
TRACE(_T("CDDS_POSTPAINT\n"));
break;
}
default:
{
TRACE(_T("CDDS_OTHER\n"));
break;
}
}
}
}
The only purpose of this header ctrl is painting above the default painting, so there is not much code in there. The painting of the CListCtrl derived class does not not do anything special, it lets CListCtrl handle the OnPaint message. The CListCtrl derived class does contain an OnCustomDraw section. But since it works on 32 bit, I doubt the problems should be searched there, on the other hand I mention it since I am out of options...
I found some posts with similar issues (only64 bit or drawstage remains) but none of them worked for me. One of these solutions was an incorrect definition of NMCUSTOMDRAW struct, but mine is the MFC version and no complaints found for 64bit for it. Another was that result did not get processed because of where the component was placed, but then it should not work on 32 bit too. Other solutions are specific for handling the OnCustomDraw messages and results, but that seems to be fine for my testcode...
Kind Regards,
Kevin
Related
I'm very new to Android and Kotlin so I may be getting something very simple wrong, but as far as I can see when I call BluetoothLeScanner.startScan() none of the possible callback methods of the ScanCallback class which I've created is ever called.
I've understood that at API level 23 & above just putting the location permissions in the manifest may not be enough so I've written code to handle that & am satisfied that my App has both COARSE and FINE location permissions
Here's my override of the OnScanResult method:
override fun onScanResult(callbackType: Int, result: ScanResult?) {
super.onScanResult(callbackType, result)
mScan = true
}
I've put a break point in each of the callback methods and when I hover over these breakpoints while the code is running, I see the message "No executable code found at line..." That's a pretty disturbing message (and I suspect is pointing to where the problem lies) but (a) how can there be no code there when everything builds OK and (b) what do you do about it?
Update on that: I think that message is a red herring. I've now moved the break points to elsewhere within the callback functions and I no longer see the 'no executable code' message. Looks like Android Studio lets you put a break point on a line with no actual code in it!
So we're back to the original question - why are we getting no callbacks?
Looks like this is now solved:
(1) I did find a setting on the phone as distinct from turning on Location. It was enable Bluetooth scanning. However it actually made no difference (2) What looks to have been the real issue is a misunderstanding of the meaning of the string which you pass to the ScanFilter Builder with setDeviceName(). There is a string in our hardware Bluetooth module which we're trying to scan for which is called device name, and I was scanning for that. When I looked instead for the Beacon advertising data, it found it.
Many thanks for suggestions (only 1 I think)
Giving permissions in the manifest is not the same as the app using it.
For ble you need to give the location and bluetooth permission. Then:
in the app(on the phone) browse your open apps
find your app and click the 3 dots in the top left
Click app info
Permissions
Toggle location to on
Also the following is a handy bit of code:
public void checkPermission() {
if (Build.VERSION.SDK_INT >= 23) {
if (checkSelfPermission(Manifest.permission.ACCESS_FINE_LOCATION) == PackageManager.PERMISSION_GRANTED && checkSelfPermission(android.Manifest.permission.ACCESS_COARSE_LOCATION) == PackageManager.PERMISSION_GRANTED) {
} else {
ActivityCompat.requestPermissions(this, new String[]{
Manifest.permission.ACCESS_FINE_LOCATION,
Manifest.permission.ACCESS_COARSE_LOCATION,}, 1);
}
}
}
#Override
public void onRequestPermissionsResult(int requestCode, #NonNull String[] permissions, #NonNull int[] grantResults) {
if (requestCode == 1 && grantResults[0] == PackageManager.PERMISSION_GRANTED && grantResults[1] == PackageManager.PERMISSION_GRANTED) {
} else {
checkPermission();
}
}
Ps Be safe all
According to the "Book of Geb" I started to map our portal's web pages. I prefer to use variables defined within static content closure block and accessing them afterwards in page methods:
static content = {
buttonSend { $("input", type: "submit", nicetitle: "Senden") }
}
def sendLetter() {
waitFor { buttonSend.isDisplayed() }
buttonSend.click()
}
Unfortunately, sometimes I get an Geb waiting timeout exception (after 60 secs) or even worse I receive the well known "StaleElementReferenceException".
I could avoid the wait timeout when using "isEnabled" instead of "isDisplayed" but for the "StaleElementReferenceException" I could only apply the below solution:
def sendLetter() {
waitFor { buttonSend.isEnabled() }
try {
buttonSend.click()
} catch (StaleElementReferenceException e) {
log.info(e.getMessage())
buttonSend.click()
}
}
I guess, this solution is not really nice but I could not apply an explicitly wait as described in another article. Thus, I have some general questions:
Should I avoid to use static content definitions when pages are dynamically?
At what time or event Geb is refreshing its DOM? How can I trigger the DOM refreshment?
Why I still get a "StaleElementReferenceException" when using CSS selectors?
I would appreciate every hint which helps to understand or to solve this issue. The best would be to have a simple code example since I'm still a beginner. Thank you!
If you defined an at check on your page class the page would first verify that condition and wait for the first n seconds. Which is assigned in your gebConfig file. The default is 30 seconds.
static at = {
waitFor { buttonSend.isDisplayed() }
}
Thus once you call your pages 'to' method with a test or whatever you are using it for the page will wait and then perform your page manipulations.
to MyPage
buttonSend.click()
Should I avoid to use static content definitions when pages are dynamically?
No. Actually, the static definitions are of closures. So what is
actually happening is each time you make use of that Pages static
components you are calling a closure which is run dynamically on the
current page(collection of webElements). Understanding this is key to
using Geb and discovering the problems you will run into.
At what time or event Geb is refreshing its DOM? How can I trigger the DOM refreshment?
When you call: to, go, at, click ,withFrame(frame, page), withWindow
and browser drive methods it will refresh the current set of
WebElements. Geb has a nice collection of utiliities to make switching
between pages and waiting for page manipulations easy. Note: Geb is
actually built on WebDriver WebElements.
Why I still get a "StaleElementReferenceException" when using CSS selectors?
It is possible the page hasn't finished loading, has been manipulated
with ajax calls or has been refreshed in some other way. Sometimes an
'at' PAGE method call can fix these issues. They are for me most
common when using frames as Geb seems to become confused between pages
and frames a little. There are workarounds.
In short if you use the page pattern you can easily switch expected pages using the Page class you have defined with a static content, at, and url closure using the below:
to(Page)
at(Page)
Navigator.click(Page)
withFrame(frame, Page) { }
In addition to twinj's answer, I would like to point out a couple of other workarounds in case you encounter a StaleElementReferenceException.
Often times I find it is better to write out your selector manually rather than rely on the contents as defined in the page. Even though your page contents should not be cached by default, they still manage to slip away from me at times. This is particularly prevalent when dealing with dynamic content or iterations.
Ex: Let's say we want to click an element from a dynamically created dropdown.
Typically you might want to do something like...
static content = {
dropdown { $("#parentDiv").find("ul") }
}
void clickDesiredElement(String elementName) {
dropdown.click()
def desiredElement = dropdown.find("li", text:elementName)
waitFor { desiredElement.displayed }
desiredElement.click()
}
If this doesn't work, try getting rid of the contents altogether, and writing out the selector manually...
void clickDesiredElement(String elementName) {
$("#parentDiv").find("ul").click()
def desiredElement = $("#parentDiv").find("ul").find("li", text:elementName)
waitFor { desiredElement.displayed }
desiredElement.click()
}
In really nasty cases, you may have to use a manual timer, as pointed out in this answer, and your code may look like this...
void clickDesiredElement(String elementName) {
$("#parentDiv").find("ul").click()
sleepForNSeconds(2)
def desiredElement = $("#parentDiv").find("ul").find("li", text:elementName)
waitFor { desiredElement.displayed }
desiredElement.click()
}
Keep in mind this is a workaround :)
For large iterations and convenient closure methods, such as each{} or collect{}, you may want to add a waitFor{} in each iteration.
Ex: Let's say we want to get all rows of a large table
Typically you might want to do something like...
def rows = $("#table1").find("tr").collect {
[
name: it.find("td",0),
email: it.find("td",1)
]
}
Sometimes I find myself having to do this iteratively, along with a waitFor{} between each iteration in order to avoid a StaleElementReferentException. It might look something like this...
def rows = []
int numRows = $("#table1").find("tr").size()
int i
for(i=0; i < numRows; i++) {
waitFor {
def row = $("#table1").find("tr",i)
rows << [
name: row.find("td",0),
email: row.find("td",1)
]
}
}
I have figured that it is the navigator which get lost when you load dynamically.
I've solve the issue locally by reinit the page or module with below code:
void waitForDynamically(Double timeout = 20, Closure closure) {
closure.resolveStrategy = Closure.DELEGATE_FIRST
switch (this) {
case Module:
init(browser, browser.navigatorFactory)
break
case Page:
init(browser)
break
default:
throw new UnsupportedOperationException()
}
waitFor {
closure()
}
}
Here is my Traverse method code
protected boolean traverse(int dir, int viewportWidth, int viewportHeight,
int[] visRect_inout) {
try {
if (hori && vert) {
// CustomItems items=new CustomItems("Hi");
switch (dir) {
case Canvas.DOWN:
this.a=dir; //Canvas.DOWN
this.b=viewportWidth; //b=2
this.c=viewportHeight; //c=3
this.d=visRect_inout; //d={2,2,250,250}
this.traverse(Canvas.UP, b, c, d);
break;
case Canvas.UP:
this.a=dir;
this.j=viewportWidth;
this.k=viewportHeight;
this.d=visRect_inout;
this.traverse(Canvas.UP, j, k, d);
break;
case Canvas.LEFT:
this.a=dir;
this.j=viewportWidth;
this.k=viewportHeight;
this.d=visRect_inout;
this.traverse(Canvas.LEFT, j, k, d);
break;
case Canvas.RIGHT:
break;
}
}
} catch (Exception e) {
System.out.println("Exception " + e);
}
return false;
}
I am very new to Custom Item.
If I had done any wrong, let me know.
The way how you invoke traverse from within switch statement looks slippery to me:
this.traverse(Canvas.UP, b, c, d); // ...
// ...and similar in cases Canvas.UP, LEFT
The code you posted so far is quite fragmentary but as far as I can tell, above will lead to infinite recursive calls for traverse, eventually ending in stack overflow error.
In your particular case this might be harmless though because you unconditionally return false from your method. Per my understanding this means that device will never attempt to invoke traverse with Canvas UP and other potentially dangerous values. Feel free to check lcdui CustomItem#traverse API documentation for more details if you're interested.
Given your code I think it makes good sense for you to study introductory code examples in tutorials. There are plenty available online - search the web for something like "MIDP tutorial CustomItem". Here's the link to one for example: Adding Custom Items to MIDP 2.0 Forms
Also, if you are using Wireless Toolkit / Java ME SDK, you could check their code examples. Per my recollection, there were nice working examples of CustomItem code there.
I am using UICsutomSwitch for my application. When I try to create it, I am getting an exception like,
Terminating app due to uncaught exception 'NSRangeException', reason: '* -[NSArray objectAtIndex:]: index 2 beyond bounds for empty array'
My code is as follows,
UICustomSwitch.h
#import <Foundation/Foundation.h>
#interface UICustomSwitch : UISwitch
{
}
-(void)setLeftLabelText:(NSString *)labelText;
-(void)setRightLabelText:(NSString *)labelText;
#end
UICustomSwich.m
#import "UICustomSwitch.h"
#implementation UICustomSwitch
-(UIView *)slider
{
return [[self subviews ] lastObject];
}
-(UIView *)textHolder
{
return [[[self slider] subviews]objectAtIndex:2];
}
-(UILabel *)leftLabel
{
return [[[self textHolder] subviews]objectAtIndex:0];
}
-(UILabel *)rightLabel
{
return [[[self textHolder] subviews]objectAtIndex:1];
}
-(void)setLeftLabelText:(NSString *)labelText;
{
[[self leftLabel] setText:labelText];
}
-(void)setRightLabelText:(NSString *)labelText
{
[[self rightLabel]setText:labelText];
}
#end
View Controller:
UICustomSwitch* switchView=[[[UICustomSwitch alloc]initWithFrame:CGRectMake(200,5,90,30)]autorelease];
[switchView setLeftLabelText:#"F"];
[switchView setRightLabelText:#"M"];
[switchView addTarget:self action:#selector(genderAction:) forControlEvents:UIControlEventValueChanged];
[elementView addSubview:switchView];
I am getting exception at " return [[[self slider] subviews]objectAtIndex:2];" call. I don't know what is the wrong, Can you guys please suggest me on this.
Thanks in advance,
Sekhar.
I ran across this issue and found the answer here:Custom UISwitch Text
Basically in iOS 4, there were UILabel's in the UISwitch's subviews that represented the "On/Off" labels. In iOS 5, there are no UILabels in the subviews array (hence the array error you're getting). The above link offers an external class you can download and customize. In my opinion, it seems like Apple is discouraging customization of UISwitch. The functionality you're after could be accomplished another way (segmented control, simulated checkbox, etc).
Also, in the given link above, the author proposes that the issue might be with not including armv6. I tried this and it does not fix the problem.
The exception indicates you are attempting to access an element of the array that is out of bounds (in a place that is larger than the actual size of the array).
You can use breakpoints and/or NSLog() calls carefully placed in your code to determine if this array ever is not-empty, and if that is so, you can continue to use these calls to find out just where the array becomes empty.
Clearly if the array is empty then the switch is setup differently than you expect it to be.
VERSION 1
class Doh {
private:
static std::map<const std::string, const Doh*> someMap;
std::string stringValue_;
public:
Doh(std::string str) : stringValue_(str) {
Doh::someMap.insert(
std::make_pair<const std::string,const Doh*>
(this->stringValue_,this)
);
}
}
The above was ok with MSVC 2010 but with MSVC 2008 it fails – and I guess it is because the object is not constructed yet when it is inserted in the map (I got a memory access violation).
So, I tried a delayed insertion, which worked:
VERSION 2
Doh(std::string str) : stringValue_(str) {
boost::thread(&Doh::insertIntoTheStaticMap,this);
}
void insertIntoTheStaticMap() {
boost::this_thread::sleep(boost::posix_time::milliseconds(1000));
Doh::someMap.insert(
std::make_pair<const std::string,const Doh*>
(this->stringValue_,this)
);
}
But as you might be able to guess, my intention is to have the static Doh::someMap as a common lookup dictionary.
VERSION 1 didn’t need any thread-safety because I would create all Doh instances in the same thread – in initialization blocks - which would be called by dynamic initializers before I enter main().
But with VERSION 2, the naïve sleep() is neither graceful nor reliable (not to mention, I might need to lock the map before insertion).
What would be a nice KISS approach?
Only potential issue I see is the initialization of the static member, if there are multiple source files. Try guarding it with a function.
class Doh {
private:
static std::map< std::string, Doh * > &get_map() {
static std::map< std::string, Doh * > someMap;
return someMap; // initialize upon first use
}
std::string stringValue_;
public:
Doh(std::string str) : stringValue_(str) {
get_map().insert(
std::make_pair
(this->stringValue_,this)
);
}
};
In neither version is there any sign of init for stringvalue_ - what does the debugger show you about this key when you hit the map insert in version 1 of the code? How is this field set up, and what is its type?
Running this in the debugger for VS2008 should allow you to narrow down the point of failure into the <map> source, I would have thought.