luckman212

@vitor I pushed a test fix, see https://github.com/vitorgalvao/custom-alfred-iterm-scripts/pull/64 with the new version, setting property iterm_opens_quietly : true works for me. I would love someone else to test & confirm.

I know why this is happening. My iTerm settings are configured to launch iTerm without opening a new window by default. Because I want iTerm always running so I have it in my Login Items so it just launches and waits until I need it. But, the script doesn't account for this, so it launches iTerm and just waits forever for a window to appear (which never does...). I think I can adjust the script myself to account for this, I'll post a PR @vitor once I have it in case you want to incorporate the changes.

I'm not sure where to post this so I'll start here. I'm using @vitor's custom iTerm script with Alfred 5.0.2.2073 and iTerm 3.4.16 on macOS 12.5.1 [21G83] I noticed that if iTerm is not already running, any Terminal command action will fail and hang Alfred with a spinning beachball (I captured 2 spindumps of this just a few minutes ago in case you want me to send them in). I've tried adjusting the delay value as suggested elsewhere but that did not make a difference. I understand this could be an iTerm or AppleScript bug, just wondering if anyone else has encountered it.

Just tried it again with another app that indicated an update was available (Itsycal) Took this screenshot BEFORE updating Then I updated... and captured this ...But Alfred still reported 0.13.2. 🤔 Then I tried quitting and relaunching Alfred, and the info updated. This reminds me a lot of a similar issue I had a while back with stale Contact info being shown.

I've definitely seen it before with other apps. I never bothered to report it because it was a small thing. I didn't know this data was being returned from an API. That's interesting- so I guess the Finder either uses a different API, or retrieves it some other way.

This is minor but- Alfred 5.0.2.2073 I updated (for example) OneDrive.app today, the new version is 22.166.0807. Get info shows this: But Alfred still has the old version in his cache: I tried pressing the Clear Application Cache button, but that had no visible effect.

Thanks @vitor — I actually noticed this by accident a few days ago and was well chuffed about it! I must have missed it in the release notes.

@pankajsz Yes you can control sync direction (bidir, one way etc) see here. Conflict resolution is also configurable, and there is a nice safety net/versioning system to keep old or conflicted copies. Overall very safe. Consuming about 85MB RAM and 0% CPU right now on my system, syncing 65K+ files.

Syncthing has a version that's bundled as a macOS native app. It's open-source, a universal binary so runs well on M1, and hasn't let me down once in over 2 years syncing my Alfred prefs along with 30+GiB of other files between 2 Macs. Highly recommended. https://github.com/syncthing/syncthing-macos

Never tried SyncMate, but I do use Syncthing to sync my Alfred prefs and it has worked beautifully for years.

Or click on his hat in the menu bar of the remote system and choose Toggle Alfred

@erusev I'd like to see this workflow too, sounds useful ! Hope you decide to share it.

It's nothing top-secret—it's an SSH workflow that lets you quickly SSH into known hosts. I'll clean it up a bit and post it to GitHub asap. For now I put the filtering back into the script itself, and used the "Show Alfred" action with a variable saved from argv to re-enter the script filter after actioning. This allows me to remove the dependence on "rerun". The downsides are (1) filtering in the script isn't as fast or powerful as Alfred's, and I can't use extra "{ match: }" keywords and (2) Alfred flickers briefly after actioning. Not huge deal breakers but for sure some kind of extra param in the script filter like "{ rerun: "after-actioning" } would be really useful for this.

Thanks @Andrew for that clarification. I wasn't aware of that. Makes sense, so as soon as filtering is triggered (by entering anything after the Script Filter trigger keyword) the { rerun: N } stops having an effect? Is there a way to alter or override this behavior? If not, it seems then my only option is to move the filtering logic back into my script and go back to argv mode.

Thanks @vitor Okay, I think this might be morphing into a bugreport then. I made a test workflow, you can grab it here if you have a minute to look. Try triggering it using keyword "rrt" (no space after) Watch the clock tick.... Now type a space... then the letter "t" (matches the word "time") Clock stops. (???) Now, try closing Alfred and then re-opening him, then pasting the string "rrd time" directly into the query field. Clock is ticking... and it even picked up the argv... (shown as subtitle) Now hit backspace ... 1 letter... Clock stops. (if it doesn't, try backspacing again...) Something doesn't seem right here.

Alfred 5 I have a Script Filter that generates dynamic results. One of the actions that I can take is "exclude" which will action the selected item and add it to an exclusion list so it's not shown in the results. Currently, I am (ab)using the "rerun" JSON property, combined with the "Don't Close the Alfred Window on Actioning" checkbox to cause the script to re-run every 2 seconds. So after waiting a short moment, the excluded result disappears. This is less than ideal, because normally there is no need whatsoever to rerun this script every 2s. The alternative is to allow Alfred to close, and make the user re-activate him and possibly re-type the query etc. I am wondering: 1) is there a way to programmatically ask Alfred to re-run the script filter just ONE TIME ONLY? and/or 2) is there any way to get at the property of the text typed after the script filter Keyword? At least then I could use the "Show Alfred" action and re-populate exactly what was in the query box prior to actioning to sort of recreate the state. I know I can store argv as a variable and use that, but in this case I am using the "Alfred Filters Results" option on the script filter, so I don't think argv is populated anywhere...

Until the docs are updated, I assume if I want to skip latching it would be { ... skipknowledge: true, ... } correct?

@pbb Same thing happened to me! The fix is easy though. Just uncheck and then re-check Alfred's access in System Preferences > Privacy > Accessibility.

@vitor Just a reminder, now that A5 is out, this new parameter should be added to the documentation at https://www.alfredapp.com/help/workflows/inputs/script-filter/json/

This is cool, but in case someone wants a step that doesn't require installing python3 (which isn't preinstalled on any macOS...) this zsh script should also accomplish the same thing, unless I am missing something?

Count me in for this too! +1

Tested on my end as well, seems to work regardless of the word order... @rob213

Yes it's indeed fixed now, thanks @Andrew !

Ok. It's a minor issue anyway. Only OCD'ers like myself would likely even notice.

No worries. It's actually quaint and I enjoy imagining Alfred asking me "Care to pick from your Favourites?" with a Cockney accent.