Search results

So, you can make a simple ASH script to call cli commands to do your stuff, including running other scripts. There are more complex ways to do it, but here's a really basic script that may get you started. It's a list of CLI commands to execute. void main() { cli_execute("login...

That seems like the best place to detect it. I am still trying to get a reproducible use case. Sorry this is being difficult. One of the issues is that the frame size is too small when we insert the menu if there's only one row or there are two rows with the text awesomemenu. If we could...

I had the same problem during rollover. We probably ought to abort on 0 and not save the value.

Sorry, to be clear, I meant "pressing the button labelled "edit" (vertically) I can wrangle CSS pretty well. There's some janky stuff in awesomemenu.php, and what I really want to see is what they're doing with the event listener on the edit, or maybe the onLoad event. Still poking at this...

I tried a lot of things as well, and saw it sometimes. I was in options switching between styles, I reloaded the page. It happened maybe one time in 10. I have two sets of source from the web inspector. One with it in the right place, one without it. In my case, it's not missing, it's...

If you press the "Edit" button in the Awesome menu, it shows the awesome menu editing interface and the script menu. If you look at "Awesomemenu.php", it writes the following CSS #config {font-size: 11px; display: none; } You can try option #9, "press the Edit button in the...

TBH, I generally turn it off while autoscend is starting up, with the caveat that I am basically on my own for support if it's broken. I really need to get back on my project to provide separate logging levels for what's recorded vs. what's in the gCLI. I don't care if a script fills up my...

Would it be worthwhile to just calculate a start time - stop time value on every request and put it in the response object? We don’t have to expose it, but when we want it, it might be useful to the dev team.

It has a harmless flaw during rollover of trying and returning a 0 second ping. But it might put bad values in certain places, so it might be better to abort if we get a rollover message. > ping 10 api verbose Ping #1 of 10... -> 0 msec (0 bytes) Ping #2 of 10... -> 0 msec (0 bytes) Ping #3...

Does --rerun-tasks work for your setup? Michaels-MBP:kolmafia mcroft$ ./gradlew test Starting a Gradle Daemon, 9 stopped Daemons could not be reused, use --status for details > Configure project : [versioning] WARNING - the working copy has unstaged or uncommitted changes. > Task...

I recall some amount of past confusion including my own when the flyeredML was > 10000 but quest could not be completed. Thread here, and I don't think anything ever came of it. It's possible that it was an addition error, although it really looks like 3 different errors all related to ML...

well, it's exhibiting a race condition again, so I may split "the part that works" from "the part I'm still working on" in the PR. IDK, I'll try to fix it before I refactor my changes. I definitely want the "don't erase it all" part into the codebase. I think the new behavior is that it's...

Hmm. I think we'd want the flyeredML to be Max(jamBandFlyeredML, rockBandFlyeredML). Would that break anything? Is there any use to the player to know that they've flyered 1000 things total when they've flyered 250 jamBand and 750 rockBand? I can't think of any, but that's just me. I'd...

Tests pass. I'm not convinced I've killed every possible bad interaction, but I think it's better and no worse and that I killed one problem at least.

I shouldn't be surprised that GitHub allows only limited resources and fails these tests when my laptop passes them. I need to see what I can do about that. Changing the monitor object for reset to match the one for saveToFile, plus another few fixes seems to have gotten rid of the wipeout...

Putting on the testing cap: 1: How does this work for headless users? Do they get a popup, or do they get a dialog, and can they set a default? 2: What happens if the GUI pops up and I leave it long enough to timeout again and then say (yes/no/cancel)? tests: no: fine. connects again and...

Well, it's very interesting, and I hope I found what I was looking to find. The bad news is my sync fix above didn't solve it. I made a draft PR of just the test in case anyone wants to dig into it. method timeinDoesNotCauseRaceCondition sets an unrelated pref starts a timeinThread starts up...

Does the fact that we have the prefs in memory (until we clear them) make the case atypical? I don't think we read from disk unless we're timing in. That seems like a promising approach. I'd have to read from disk after I wrote to make sure it was persisted and nothing was lost.

Did more research into sync, which is a new area to me. What I learned: Any things that use the same monitor object are mutually exclusive (one of anything that is in a synchronized (lock name) block, regardless of the code block) Synchronized methods and class effectively are like a...

I learned the Kent Beck version: "Make It Work, Make It Right, Make It Fast" The Preferences.saveToFile() method syncronizes with a monitor object named lock, which is not used in any of the other methods of Preferences. Can we reuse the lock across methods, or is it just for one method, which...