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

cameri / nostream / 43904420785
78%
main: 79%

Build:
Build:
LAST BUILD BRANCH: rewrite/main
DEFAULT BRANCH: main
Ran 16 Nov 2026 04:15PM UTC
Jobs 2
Files 321
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

16 Nov 2026 04:09PM UTC coverage: 77.582% (+2.6%) from 75.031%
43904420785

push

github

web-flow
docs(storage): state the count cap cost as a decode per key past it (#1137)

## Description
ADR 0011 and the comments in
`crates/nostream-storage/src/lmdb/events.rs` state the cost of the
count's id-set cap wrongly: they say it is "a body decode per repeated
key past it". The code reads the row's body for every key the set does
not answer past the cap, unless the range is the family's first exact
criterion, so the cost is not confined to repeated keys.

This PR corrects the prose in four places in the ADR and three comments
in `events.rs`. It changes no code and no behaviour:

- "The price of the cap is a body decode per repeated key past it" now
says every key the set does not answer reads the row's body (a later
criterion's key, or a `#g` prefix's key), not only a repeated one, and
gives the measured cost (5.3× to 5.7× the uncapped walk at 20,000 rows),
using the wording the issue proposes.
- "Below the cap the walk reads no body" is qualified to a walk over a
single tag family, since a second family decodes there too.
- The same correction is applied at each copy: the ADR's other two
statements, the `tag` field comment, the `shortcut:` marker and the
test's doc comment.

What the issue says is right is left as it is: the set bound (270 352
B), the exactness of the count and the free cap below 5,000 rows.

## Related Issue
Fixes #1081

## Motivation and Context
The record is wrong in the direction a reader would trust: a reader who
believes the cost is per repeated key would expect a filter with no
repeated key to pay nothing past the cap, while `{"#t":["zz","a"]}` (one
key a row, zero repeats) costs 5.7× its uncapped time at 20,000 rows.

## How Has This Been Tested?
Prose and comment changes only. The Rust validation commands pass
(`cargo fmt --all --check`, both clippy passes, `cargo test --workspace
--all-features`, `cargo build --workspace --all-features --release`),
and the diff touches no code line.

## Screenshots (if appropriate):

## Types... (continued)

4501 of 6342 branches covered (70.97%)

Branch coverage included in aggregate %.

9792 of 12081 relevant lines covered (81.05%)

758.78 hits per line

Jobs
ID Job ID Ran Files Coverage
1 Unit - 43904420785.1 16 Nov 2026 04:17PM UTC 321
75.03
GitHub Action Run
2 Integration - 43904420785.2 16 Nov 2026 04:15PM UTC 321
32.99
GitHub Action Run
Source Files on build 43904420785
  • Tree
  • List 321
  • Changed 236
  • Source Changed 0
  • Coverage Changed 236
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • Github Actions Build #43904420785
  • cc4685e9 on github
  • Prev Build on rewrite/main (#43903844407)
  • Next Build on rewrite/main (#43976483504)
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