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

FlexMeasures / flexmeasures / 30861150682
84%
main: 88%

Build:
Build:
LAST BUILD BRANCH: feat/rerun-failed-jobs
DEFAULT BRANCH: main
Ran 03 Aug 2026 11:21PM UTC
Jobs 1
Files 173
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

03 Aug 2026 11:06PM UTC coverage: 83.464% (-0.002%) from 83.466%
30861150682

Pull #2382

github

Flix6x
Address review: explain the trigger plainly, and test what the docstring claims

Three points from review, all fair.

The comment on why a trigger is used, rather than code in the save path, assumed the
reader already knew why the save path is not enough.
Say it plainly instead: beliefs reach timed_belief by several routes,
code in one route only ever sees the beliefs that took it,
and a trigger on the table sees every insert whatever the route.
The per-statement behaviour is now stated concretely too --
a million beliefs in one statement adds one small insert, not a million.

The listener's docstring did not say when it actually runs.
It now names both routes a database can take to the trigger:
Alembic migrations install it in production,
this listener installs it for schemas built by create_all(),
which is the test suite and some development setups.
It also explains why it listens on the metadata rather than on timed_belief --
that would bet on table creation order -- without leaning on the word plpgsql.

test_data_source_sensors_uses_the_summary claimed to show the answer came from the
summary rather than from timed_belief, and did not:
it saved a belief and checked the accessor found the sensor,
which passes either way.
Replaced with a test that writes a pair into the summary for which no belief exists,
so an accessor reading timed_belief could not return it and one reading the summary
must.
Confirmed it discriminates by temporarily restoring the old belief-scanning
implementation, against which it fails.
The two behavioural tests remain, renamed to say what they do check.

flexmeasures/data: 257 passed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Signed-off-by: F.N. Claessen <claessen@seita.nl>
Pull Request #2382: Summarise which sources recorded for which sensors

19 of 22 new or added lines in 2 files covered. (86.36%)

35 existing lines in 2 files now uncovered.

16611 of 19902 relevant lines covered (83.46%)

0.83 hits per line

Uncovered Changes

Lines Coverage ∆ File
2
87.7
-0.17% flexmeasures/data/models/time_series.py
1
88.99
-0.2% flexmeasures/data/models/data_sources.py

Coverage Regressions

Lines Coverage ∆ File
22
88.99
-0.2% flexmeasures/data/models/data_sources.py
13
87.7
-0.17% flexmeasures/data/models/time_series.py
Jobs
ID Job ID Ran Files Coverage
1 30861150682.1 03 Aug 2026 11:21PM UTC 173
83.46
GitHub Action Run
Source Files on build 30861150682
  • Tree
  • List 173
  • Changed 2
  • Source Changed 2
  • Coverage Changed 2
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Github Actions Build #30861150682
  • Pull Request #2382
  • PR Base - feat/timed-belief-pk-column-order (#30825158467)
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