John Morris jtmorris
  • Joined on 2026-05-14
jtmorris created branch perf_agent_improvement in jtmorris/dms_tailscalectl 2026-05-21 04:53:22 +00:00
jtmorris commented on issue jtmorris/dms_tailscalectl#28 2026-05-20 07:02:56 +00:00
No accessibility attributes + hardcoded English strings (a11y + i18n gap)

Yeah, you don't just get to say "I'm not doing that because it's hard" and call that a plan to implement resolution to this issue. I'm fine keeping focus narrow now and working on the localization…

jtmorris commented on issue jtmorris/dms_tailscalectl#24 2026-05-20 06:57:13 +00:00
Missing package.json declaring module type (ESM/CJS mismatch between tests and lib)

Okay. No red flags from me. Do it. Let's see the result.

jtmorris commented on issue jtmorris/dms_tailscalectl#15 2026-05-20 06:54:49 +00:00
No guard against concurrent toggleProcess / exitNodeProcess invocations
  1. There is no resetState() at this time. You're referencing another unimplemented issue's proposed code. That is not guaranteed to exist. You should add this caveat if you're going to be that…
jtmorris commented on issue jtmorris/dms_tailscalectl#32 2026-05-20 06:40:39 +00:00
Inverted MouseArea + manual size binding anti-pattern in peer list delegate

I'm not redundant here! I am the one who smelled something fishy in Denmark. Yes, I called in Claude for backup, but I knew something was off. DON'T JUDGE ME!!!!


Anyway. @vybe. When…

jtmorris commented on issue jtmorris/dms_tailscalectl#30 2026-05-19 06:58:42 +00:00
statusCheck triggered from too many sites without deduplication (wasteful concurrent polls)

The general idea look okay. But check your detectClipboard.onExited behavior before implementing your plan. #12's plan is to completely refactor that behavior. #12's plan may be implemented first.

jtmorris commented on issue jtmorris/dms_tailscalectl#12 2026-05-19 06:49:38 +00:00
Shell interpolation of clipboard detection output in TailscaleWidget.qml

Looks okay enough. Give it a shot @vybe.

jtmorris commented on issue jtmorris/dms_tailscalectl#29 2026-05-19 06:43:28 +00:00
Lax / fragmented error handling throughout the widget (catch-all masking, missing exit checks, no stderr)

Not sold on step 3. This is a simple toggle. Its objective should be a dirt simple toggle. You're storing state all over the place. Why even bother telling the user what state the action failed to…

jtmorris commented on issue jtmorris/dms_tailscalectl#12 2026-05-19 06:23:54 +00:00
Shell interpolation of clipboard detection output in TailscaleWidget.qml

I'm not sold on this approach. It's better than what's there, but it feels overengineered and like we are papering over the problem.

The whitelist validation is only necessary because the…

jtmorris commented on issue jtmorris/dms_tailscalectl#21 2026-05-19 05:23:36 +00:00
Missing edge-case tests for parsePeers and findActiveExitNode

Yes. Very much agree that the current tests are minimally useful. Needs dramatic expansion to cover edge cases and drastically widen coverage, or it should be removed as code maintenance overhead,…

jtmorris commented on issue jtmorris/dms_tailscalectl#13 2026-05-19 04:55:05 +00:00
Double error toast spam on statusCheck failure (TailscaleWidget.qml)

See #33's discussion on the 5-second timer disaster design decision.

jtmorris commented on issue jtmorris/dms_tailscalectl#33 2026-05-19 04:54:23 +00:00
Aggressive unconditional 5-second polling even when popout is closed (battery/CPU waste)

Yeah, this is the second worst decision in @vybe's vibe-coded solution in my opinion. We don't need to poll Tailscale very often at all. I'd up this to 5-minute polling even, with triggers on…

jtmorris commented on issue jtmorris/dms_tailscalectl#25 2026-05-19 04:24:38 +00:00
No stderr capture on toggleProcess / exitNodeProcess (lost diagnostic output)

This issue should probably be looked at more generically as well. In the event of unexpected output or command failure, how does the user know what happened? Some guidance is needed. Either the…

jtmorris commented on issue jtmorris/dms_tailscalectl#23 2026-05-19 04:22:06 +00:00
Hardcoded popout dimensions vs dynamic content (magic numbers, no implicit sizing)

Yes! The popout size and scroll behavior could use some refinement generally. For most users, it can probably dynamically size to fit the content. However, there should be some maximum size.…

jtmorris commented on issue jtmorris/dms_tailscalectl#12 2026-05-19 04:02:59 +00:00
Shell interpolation of clipboard detection output in TailscaleWidget.qml

Gemini also flagged a good point about the which command being non-POSIX compliant and not as universal as the POSIX command command -v. While fixing this issue, serious consideration to…

jtmorris closed issue jtmorris/dms_tailscalectl#1 2026-05-19 00:41:17 +00:00
Tailscale Widget Plugin for Dank Bar
jtmorris pushed to master at jtmorris/dms_tailscalectl 2026-05-19 00:16:39 +00:00
54c64f6d5e Added screenshot. Added ignore for working files not appropriate for the repo.
jtmorris commented on issue jtmorris/dms_tailscalectl#9 2026-05-18 19:41:11 +00:00
8. Final packaging & manual test

@vybe: Note, when you get to this one, let's not publish to DMS plugin registry. Yet. Just polish the packaging, include screenshots, et cetera.

jtmorris commented on issue jtmorris/dms_tailscalectl#10 2026-05-16 03:22:57 +00:00
#3.5 Fix popout opening blocked by MouseArea event consumption

The issue was resolved by adding a root-level MouseArea at the PluginComponent level (sibling to BasePill) with acceptedButtons: Qt.RightButton. This ensures right-click events are captured by our…