I recently added the ability for a FBP runtime that I am building to send the network protocol commands output and error. When using app.flowhub.io for the client, the error message shows up in a message window but I don't see the message that accompanies the output command when it is sent. The websocket frame sent to the noflo ui from my runtime looks like:
{"protocol":"network","payload":{"type":"message","message":"Hello"},"command":"output"}
Where should I expect to see the message text for an output command?
The network output and errors should show in the right hand panel, under the runtime connection info. You may need to press the eye icon to open it.
However I believe only strings are supported.
Related
currently I use Goland as IDE, it happens that sometimes when I have a compilation error then in the terminal of the IDE it's displayed the error as a link, when I click on that link it takes me to the exact line where the error is, however this behavior is not consistent, sometimes it works sometimes it doesn't (the link is not displayed and only the error message is printed). So, what should I do to ensure that the link is always showed up?
You need to raise an issue on the official tracker, https://youtrack.jetbrains.com/issues/Go, and provide samples for this.
The functionality is automatic and there are no flags to configure.
I call the FM MESSAGES GIVE into Function Module MRM_SRM_INVOICE_SIMULATE, how do I find the point where the error is generated?
Thanks
Further data is needed to give you appropriate help.
But have you already tried to add a breakpoint in the MESSAGE statement?
To do this:
Open debugger via /h + action
then in menu Breakpoints -> Breakpoint at -> Breakpoint at statement [Shift+F5]
in the pop-up window grid, write MESSAGE in the first line and then click Ok.
after that, you need to replicate the error and the debugger will stop in EVERY MESSAGE statement reached in runtime.
Note: there may be many MESSAGE statements reached. Yo need to check one by one using F8 key.
Extract the message ID MSGID and message number MSGNO from the list of messages as displayed.
Then try the following:
Try to find the place where the message was triggered by static code analysis: In transaction SE91, navigate to the message given by MSGID and MSGNO, then perform a Where-Used-List for the single message. This will give you the code places where that message is issued. Put a breakpoint at each place and repeat the transaction that leads to the message. The system should stop at the point where the message is issued.
Some developers issue their messages dynamically (with variables for MSGID and MSGNO), so that the points where they are issued cannot be found by static code analysis. This is bad, but there is another way to find it. The breakpoint at statement MESSAGE as described in the answer of #cape_bsas usually stops too often, but you can put a watchpoint on the field SY-MSGNO, instructing the debugger to stop as soon as SY-MSGNO assumes the given message number.
I would like to capture selected messages from another application similar to the way Spy++ 64 does. I don't have a clue where to start. I guess I need something like WndProc only it would fire for external application not my application.
I looked at the code in the post Capture all Windows Messages
and am not sure it will actually do what I want and I would like a VB.Net Solution.
What I am trying to do is to hook the WM_PAINT message from another application. Take Notepad as a simple example.
In Spy++ 64 I would click Spy->Log Messages, target the textbox body area of Notepad, Click the General checkbox on the Messages Tab and every time Notepad is repainted Spy++ 64 shows:
00140A1C P WM_PAINT
Does anyone have any ideas or can you point me in the right direction?
I now basics of VBA and use it to send or get data in AS 400, but I am not so advanced to understand documentation of IBM "Host Access Class Library". I want to know how can I detect error in AS400 for now I am using script below or skip errors by self, outside of running macro.
Sub check_error()
autECLSession.autECLOIA.WaitForAppAvailable
autECLSession.autECLPS.wait 100
autECLSession.autECLPS.SendKeys "[reset]"
End Sub
When error appear I cannot use some commands like autECLOIA.WaitForInputReady, So instead I use 'wait 100' in my 'check_error' but it not always work. So how can I detect error, is there any command to do that? Or a method to handle it?
This is a massive topic that, unfortunately, depends on the application. There are so many ways that errors can be presented to the user on a 5250 display that one method can not fit all circumstances. It might be best to just ask us about your specific use case rather than all use cases.
However, let me give you a high level view of some of the ways errors can be presented.
External Message Queue
This is when the program sends a status message to the *EXT message Queue. These messages are displayed on the last line of the display. They do not appear in the job log, and are typically not used for exceptions, but they also do not end the program.
Program Message Queue
An application can define a message subfile which is a record format that displays on the screen between lines 1 and 24 on an 80x24 screen, or between lines 1 and 27 on a 132x27 screen. Where it appears, and the number of display lines it uses are defined in the display file DDS. The appearance of a message in this message subfile may be the only indication of an error, but it may contain informational messages as well. You are going to have to determine which it is by the text of the message. In conjunction with the message, a field may also be reverse imaged, or colored in a way that indicates an error, but the keyboard is not locked, and there is no audible indication that there is a problem.
DDS Error Messages
DDS error messages can be defined that lock the keyboard, reverse image the field, and display an error message. These messages can be displayed on any line as well, which is configurable in the DDS. In addition to the typical display lines, there is an addition line: 25 on a 80x24 screen, and 28 on a 132x27 screen, that is accessible to these messages.
Custom Error Messages
And sometimes an error is indicated in a completely custom way, for example by populating a text field somewhere on the screen.
The point is that you need to know your application, and how it provides user feedback, and without that knowledge, we can't help you.
My problem is that i wish to capture the text that is displayed in the
console of Xcode when an application is executed and display it in a
text box in my App.
If i override NSLog i can just capture the explicit NSLog commands
that are issued in the course of the program. However many statements
that are just inserted by the compiler are not captured.
Is there a way to read the Xcode Console buffer while the app is
running and display it in the app too ??
What you see in the log window of Xcode is a composite of the messages that would normally go to standard out and standard error file streams and to the system log. If you want to capture those streams, you need to close them and reopen them as pipes or files.
If you do this, the documentation says that if you redirect standard error from the default, NSLog will log to that as well as the console. Thus you don't need to override it.
Redirecting standard error and standard out is a fairly common thing to do in Unix. The basic technique to redirect to a file is to close the file descriptor using close(2) and then reopen it using open(2) or pipe(2).
The Xcode Console is just a window that reads errors and such from the Console logs. Try reading from there.