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

Freegle / Iznik / 28987
73%
master: 73%

Build:
Build:
LAST BUILD BRANCH: feature/rippling-analytics-fixed-horizon
DEFAULT BRANCH: master
Ran 29 Jul 2026 07:40PM UTC
Jobs 4
Files 1442
Run time 3min
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

29 Jul 2026 07:09PM UTC coverage: 72.89% (-0.08%) from 72.973%
28987

Pull #1196

circleci

edwh
fix(rippling): fixed-horizon analytics trends so recent days don't draw as decline

The sysadmin analytics trend series counted replies and takes EVER, so
recent days - which simply hadn't finished accruing - always drew as a
steep mechanical decline: taken outcomes are marked a mean of ~19 days
after the post enters rippling, so the taken-ever line fell to near zero
over the trailing fortnight regardless of reality, and reply-ever/mean-
replies sloped down over the trailing days.

- Trend KPIs now use fixed accrual horizons from reach creation:
  replies + mean replies within 36h (ReplyHorizonHours), taken within
  14 days (TakenHorizonDays). Every day is measured the same way, so a
  lower recent point is a real change.
- Each trend row carries replied_mature / taken_mature flags (false
  until the day's whole horizon has elapsed); the component feeds them
  to a Google Charts 'certainty' role so a still-provisional tail draws
  dashed/pale rather than as a decline. Chart titles + section copy name
  the windows.
- The Section 1 replied-36h clock moves from
  COALESCE(MIN(rippled arrival), MIN(origin arrival)) to
  rippling_reach.created_at: autorepost bumps messages_groups.arrival
  forward, which silently handed older posts ever-longer 36h windows
  (and it drops two correlated subqueries per post).

Verified against prod data: with fixed horizons the July "decline" in
reply rate shrinks from -28pts (ever) to a real -5pts, and the taken
cliff (72% -> 18%) disappears entirely.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Pull Request #1196: fix(rippling): fixed-horizon analytics trends so recent days don't draw as decline

13475 of 17702 branches covered (76.12%)

Branch coverage included in aggregate %.

31 of 55 new or added lines in 2 files covered. (56.36%)

439 existing lines in 3 files now uncovered.

141373 of 194738 relevant lines covered (72.6%)

38.31 hits per line

Uncovered Changes

Lines Coverage ∆ File
24
52.48
-38.82% iznik-server-go/rippling/analytics.go

Coverage Regressions

Lines Coverage ∆ File
276
87.34
-0.15% iznik-server-go/message/message.go
161
52.48
-38.82% iznik-server-go/rippling/analytics.go
2
98.19
-0.25% iznik-nuxt3/composables/useFetchRetry.js
Jobs
ID Job ID Ran Files Coverage
28987 playwright - 28987 29 Jul 2026 07:41PM UTC 75
21.95
CircleCI Job 28987
28987 laravel - 28987 29 Jul 2026 07:40PM UTC 508
70.45
CircleCI Job 28987
28987 go - 28987 29 Jul 2026 07:40PM UTC 164
83.48
CircleCI Job 28987
28987 vitest - 28987 29 Jul 2026 07:40PM UTC 769
70.88
CircleCI Job 28987
Source Files on build 28987
  • Tree
  • List 1442
  • Changed 8
  • Source Changed 0
  • Coverage Changed 8
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • CircleCI Build #28987
  • Pull Request #1196
  • PR Base - master (#28975)
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