• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

umputun / agterm / 37219287184
94%

Build:
DEFAULT BRANCH: master
Ran 04 Oct 2026 05:12PM UTC
Jobs 1
Files 183
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

04 Oct 2026 05:06PM UTC coverage: 93.729% (+0.004%) from 93.725%
37219287184

push

github

web-flow
feat(remote): retry waiting panes and dropped streams on wake and when the network returns (#687)

* feat(remote): retry waiting panes and dropped streams on wake and when the network returns

The backoff alone can leave a Mac that slept all night waiting up to five minutes. RemoteLinkObserver
listens to the display wake SystemWakeObserver already bridges and to NWPathMonitor; either makes every
waiting pane and dropped presentation stream due now and starts its backoff over, so a probe fired before
the network is back ramps from one second again instead of waiting out the cap. The first path report is
the state at start, never a change.

Claude-Session: https://claude.ai/code/session_01GqkEnpmSQ9B16JabFfNGjx

* fix(remote): a retry during a running probe or launch still starts the backoff over

Review notes on #687. A wake or network return that arrived while a probe or a connecting launch was already running returned before resetting the failure count, so that attempt's failure kept the old backoff and could wait five minutes. The count resets before the in-flight guard; the attempt itself is not started twice. The hosted backoff test asserts the book's probing flag after the second tick, which due() sets synchronously, instead of a counter a spawned task changes. The path monitor retries on every path change that leaves the path usable, a hand-off that stayed usable included, and the docs say that a key on a waiting pane starts the backoff over too.

Claude-Session: https://claude.ai/code/session_01GqkEnpmSQ9B16JabFfNGjx

* test(remote): the wake test runs without a path monitor; one probe fake; one-clause doc comments

Review notes on #687. RemoteLinkObserver takes watchPath, so the wake test cannot count a path change that happens on the machine running it. CountingProbe duplicated Probe. The doc comments on the observer and both retryNow point at control-api.md instead of repeating it, and commands.html says what the reference says about retryin... (continued)

10 of 10 new or added lines in 2 files covered. (100.0%)

14768 of 15756 relevant lines covered (93.73%)

2590004.31 hits per line

Jobs
ID Job ID Ran Files Coverage
1 37219287184.1 04 Oct 2026 05:12PM UTC 183
93.73
GitHub Action Run
Source Files on build 37219287184
  • Tree
  • List 183
  • Changed 2
  • Source Changed 2
  • Coverage Changed 2
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Github Actions Build #37219287184
  • 175231f0 on github
  • Prev Build on master (#37192539793)
  • Next Build on master (#37221636167)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc