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

sandialabs / LoQS / 42178660578
90%

Build:
DEFAULT BRANCH: main
Ran 06 Nov 2026 07:10AM UTC
Jobs 1
Files 85
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

06 Nov 2026 07:06AM UTC coverage: 89.367% (+0.07%) from 89.297%
42178660578

push

github

web-flow
Speed up checkpoint reads and remove `lazy_loading` (it is always lazy when possible now) (#172)

* Seed checkpoint encode cache from the snapshot's encode

A fresh checkpointed run now seeds its encode cache from the encode that
wrote the canonical file, instead of decoding the program straight back,
saving one decode of the noise model per circuit (#153 Part 6).

- `CheckpointStore.write_snapshot` takes `encode_cache` and fills it,
  once the file is in place, with the new file's IDs, no object IDs.
- `ProgramResults` seeds from its own snapshot write, decodes only an
  existing file on resume, and empties the cache on a store rebuild,
  so shots checkpointed into a second directory read back.
- Minor: `Serializable.dump` honours a caller's `encode_cache`; tests
  for the fresh seed and the second directory; shorter
  `write_snapshot` docstring.

Co-authored-by: claude-sonnet-5-5 <noreply@anthropic.com>
Co-authored-by: claude-opus-5-5 <noreply@anthropic.com>

* Cache checkpoint key index; drop decoded shots

Checkpoint reads stop rebuilding the key index on every call and stop
keeping decoded shots in the decode cache, so per-shot reads cost the
same at any shot count and memory stays bounded (#153 Part 2).

- `CheckpointStore` caches its key index per attr, keyed on each
  source's identity plus the canonical commit marker; the nested view
  finds its entry through it instead of listing the parent's keys.
- `get` and `iter_items` drop the decoded values they added to a
  decode cache when the call ends (`iter_items` by type, so a copy's
  source resolved out of key order goes too).
- Minor: file identity adds `st_size`; tests for other writers'
  payloads, consolidation and coarse-timestamp appends, and for the
  cache drops; `get_shot_history` comment gives the after-collect cause.

Co-authored-by: claude-sonnet-5-5 <noreply@anthropic.com>
Co-authored-by: claude-opus-5-5 <noreply@anthropic.com>

* Read only the selected data in checkpointed collec... (continued)

3778 of 4500 branches covered (83.96%)

Branch coverage included in aggregate %.

12292 of 13482 relevant lines covered (91.17%)

0.91 hits per line

Coverage Regressions

Lines Coverage ∆ File
44
94.29
0.44% internal/checkpointstore.py
40
91.7
0.2% internal/serializable.py
17
92.93
-0.06% core/quantumprogram.py
17
83.93
3.08% internal/encoder/baseencoder.py
15
95.8
-0.03% tools/multiprogramrunner.py
11
93.77
0.87% core/programresults.py
Jobs
ID Job ID Ran Files Coverage
1 42178660578.1 06 Nov 2026 07:10AM UTC 85
89.37
GitHub Action Run
Source Files on build 42178660578
  • Tree
  • List 85
  • Changed 6
  • Source Changed 0
  • Coverage Changed 6
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • Github Actions Build #42178660578
  • 5d46ca6b on github
  • Prev Build on main (#42140515486)
  • Next Build on main (#42656922475)
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