I am not a KDE developer, but this works for me, are you using JavaScript or QML for your script.
And did you add the second print command after you ran your script with only the first print?

just javascript, everything works fine when I only use the print() method once. But when I start to use the print() method multiple times in quick succession then it won't print() some of the print() methods. When I use the "plasma" option to execute the script it will print both but when I use the "kwin" option it will skip some print() statements.

It seems to depend on how long kwin is busy with executing the commands between the consecutive print statements.

for example, if I put a long loop between the two print() statements then both will execute, like this example:

print('hi');
var a = 0;
for (var i = 0 ; i < 10000000; i++)
{
  a += i;
}
print('hihi');

but doing only:

print('hi');
print('hihi');

seems to be too fast and kwin will not print the second statement to console.

Interesting, I can reproduce this on my KDE 5.12.5, but this happens only if I switch between the KWin and Plasma "modes".
It works if I leave it at KWin (the default).

That's because of a different bug (which I should probably report as well...). But if you open "WM console" it shows the default state as "kwin" but it's actually on "plasma" because you don't have access to some of the kwin globals. You need to manually switch to "plasma" and then back to "kwin" to actually be able to run the script as a kwin script.

you can test that you are not on "kwin" because you can't use the global `workspace` for example if you try this, when you first open the console:

print(workspace.desktopGridWidth);

but if you switch to "plasma" and back to "kwin" then you access to these globals.

I noticed that workspace was missing after my comment and went to test a few things.
The missing print outputs also seem to be a scripting console bug because all messages are printed in KWin's console output.

print("hi")
print("hihi")

prints "Hi" in the scripting console and "hi\nhihihi" to KWin's stdout if you enable debug output in KDebugSettings.

Ah okay, so it seems to be purely a console thing. Btw I've only ever used the "wm console" to write my kwin scripts, could you explain to me how to use and set this "KDebugSettings" variable? That would make my kwin scripting a lot easier for now as long as this bug exists.

KDebugSettings is an application to configure logging for KDE applications ( https://community.kde.org/KDevelop/Debugging is the only documentation I was able to find).
Increase the level for "KWin Scripting" (I am using "all") to make KWin log the print()s.
I am viewing the log output by restarting KWin from Konsole using "kwin_x11 --replace".

I run into the same bug. By launching kwin under:
```
QDBUS_DEBUG=1 QT_LOGGING_RULES="kwin_scripting.debug=true"
```

I can see that the missing logs are printed by kwin, but was not sent through dbus -- occasionally only some of the messages are sent through dbus.

After adding a sleep statement before `emit print(message);` in `scripting.cpp`, all messages can be sent normally.

Running bustle it seems the kwin side is correct.

We get the print signal emitted twice before run returns.

Indeed, reproduced occasionally.

From what I understand of the bug a port to using an adaptor will fix this as then it won't need to scan our metaObject so to emit the signal from the DBus thread which might be behind our main thread.

Created attachment 130595 [details]
screenshot of Bustle

When I run bustle it seems to show kwin side is incorrect. I put the following code into the scripting console:
```
print(workspace.numScreens);
print(workspace.workspaceWidth);
print(workspace.workspaceWidth+1);
print(workspace.workspaceWidth+2);
print(workspace.workspaceWidth+3);
print(workspace.workspaceWidth+4);
```
and launch kwin under the environment `QDBUS_DEBUG=1 QT_LOGGING_RULES="kwin_scripting.debug=true"`.
When I click "execute" in the scripting console, I saw this in the output of kwin:
```
QDBusConnectionPrivate(0x7f67fc00f8c0) got message (signal): QDBusMessage(type=MethodCall, service=":1.2859", path="/Scripting", interface="", member="loadScript", signature="s", contents=("/home/
USER/.local/share/plasmashell/interactiveconsoleautosave.js") )                                                                                                                                      QDBusConnectionPrivate(0x7f67fc00f8c0) Automatically sending reply: (QVariant(int, 2))
QDBusConnectionPrivate(0x7f67fc00f8c0) sending message (no reply): QDBusMessage(type=MethodReturn, service="", signature="", contents=(2) )                                                         QDBusConnectionPrivate(0x7f67fc00f8c0) got message (signal): QDBusMessage(type=MethodCall, service=":1.2859", path="/2", interface="", member="run", signature="", contents=() )
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "2"
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "13440"
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "13441"
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "13442"
QThread(ptr=0x7f6811979060, name=QDBusConnectionManager) emitting signal at "/2"
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "13443"
kwin_scripting: "/home/USER/.local/share/plasmashell/interactiveconsoleautosave.js" : "13444"
QThread(ptr=0x7f6811979060, name=QDBusConnectionManager) emitting signal at "/2"
QDBusConnectionPrivate(0x7f67fc00f8c0) sending message (no reply): QDBusMessage(type=MethodReturn, service="", signature="", contents=() )
QThread(ptr=0x7f6811979060, name=QDBusConnectionManager) emitting signal at "/2"
```

According to the output, 6 messages are printed, but only 3 signals are sent (this bug occasionally happens, but not always).
In the scripting console, I only observed the first 3 lines of output.

In Bustle, I only observed 3 signals (see attachment).

*** Bug 477069 has been marked as a duplicate of this bug. ***