Detecting viewer component failure

[ugradedeveloper]

Occasionally the PDF-XChange Viewer will present the user with a message saying that it has encountered a problem and needs to close. I wonder if there is a way that this situation can be detected via the SDK, so that my application can re-initialize the ActiveX control. It would be nice if an event were generated which my application could monitor, however I realize that this may not be possible due to the error situation.

If I knew the process ID of the instance of PDFXCview.exe that was running, I could check in code to see if that process had exited. However I have not been able to figure out how to do this. There can be multiple instances of the Viewer running, and some of them might be running standalone; there doesn't seem to be any way to determine which one corresponds to a given instance of the ActiveX control.

Anyone have thoughts on this?

Thanks.

[Vasyl - PDF-XChange]

Code: Select all

Occasionally the PDF-XChange Viewer will present the user with a message saying that it has encountered a problem and needs to close.

Please details: it hypothetically or crashes occurs really?

If I knew the process ID of the instance of PDFXCview.exe that was running, I could check in code to see if that process had exited. However I have not been able to figure out how to do this.

We will to add some special features into next build:
1. new parameters:
"General.ProcessID" - will represents server-process ID.
"General.ProcessName" - will represents name of special global synchronization-object(mutex) which unequivocally corresponds to server-process.
You will be able to test existence of server-process:

Code: Select all

string syncObjName;
GetProperty("General.ProcessName", syncObjName);
hMut = OpenMutex(READ_CONTROL | SYNCHRONIZE, FALSE, syncObjName);
if (hMut)
{
    res = WaitForSingleObject(hMut, INFINITE);
    CloseHandle(hMut);
}

2. also we will to prevent displaying of any crash-messages from server-process.

[ugradedeveloper]

Please details: it hypothetically or crashes occurs really?

Sorry I wasn't more specific - I mean it really happens, my users have reported this a number of times. They are understanding about it because Acrobat (which we formerly used in our application) used to crash a lot. But I want to minimize inconvenience for them.

Thanks.

[John - Tracker Supp]

Understood and investigating - thanks for your patience.

[ugradedeveloper]

By the way, the proposed changes you have described sound good. I look forward to the release of build 42.

[Vasyl - PDF-XChange]

About crashes..

...I mean it really happens, my users have reported this a number of times...

If you have a detailed information about it (for example: the problem document, sequence of user's actions which leads to crash) that please give us this information.
We would like to fix it.

Thank you.

[ugradedeveloper]

If you have a detailed information about it (for example: the problem document, sequence of user's actions which leads to crash) that please give us this information.

I would love to provide you with sufficiently detailed information to reproduce the problem. Whether I can actually do so is another question. Our users do not operate in a standard environment - there are different versions of Windows, not to mention different hardware on which they are operating. There is no one particular document that causes the problem - they work with hundreds of documents per day, from many different sources. And in most cases they cannot tell me exactly what they were doing when an error occurred. So there just seem to be too many variables.

The only thing I can tell you is that their work consists exclusively of adding comments to documents - generally using the Sticky Note, Text Box and Typewriter tools, plus the underliner, strike-out and highlight tools.

I can tell you that one user reported several crashes in one day when she was using the stamp tool on a document that had very small text (I believe it had been created from a print document that was scanned into PDF) - she had the zoom level set at 200%, and as a result she had to resize each stamp after placing it on the page. Apparently the act of resizing the stamp was what would trigger the crash. However I have tried reproducing the error using this information, and could not. So evidently there is some other crucial factor which I cannot identify.

Thanks for your efforts.