Test: room for a giant thread's card on a busy runner #305

Merged

littlefox merged 4 commits from fix-giant-list-timeout into main 2026-10-23 09:30:41 +00:00

Conversation 2 Commits 4 Files changed 8 +100 -17

littlefox commented 2026-10-23 01:53:49 +00:00

Owner

Copy link

Fixes main: run #502 (merge of #302) failed once on giant.spec list test on Android (no card within 15 s); the same test passed on the branch. Gives the first card the 60 s the rest of that test allows.

Refs #285

🤖 Generated with Claude Code

https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

Fixes main: run #502 (merge of #302) failed once on giant.spec list test on Android (no card within 15 s); the same test passed on the branch. Gives the first card the 60 s the rest of that test allows. Refs #285 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

littlefox added 1 commit 2026-10-23 01:53:49 +00:00

Test: room for a giant thread's card on a busy runner ... 70adb5a934

main's run of #302 failed on it once (Android, no card within 15 s); the
same test passed on the branch. The rest of that test already allows
60 s for the giant thread.

Refs #285

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

littlefox commented 2026-10-23 07:17:13 +00:00

Author

Owner

Copy link

Not just a slow runner: building a 3 000-long chain's index walked down a level (one text) per round trip, so with CI's (or any real server's) latency the card never came. Added prefetching the texts after a level in its channel (fetch_after), in the index walk and the chat's walks. Locally with 30 ms each way: the giant list test passes (card in ~20 s); devserver integration tests pass.

Not just a slow runner: building a 3 000-long chain's index walked down a level (one text) per round trip, so with CI's (or any real server's) latency the card never came. Added prefetching the texts after a level in its channel (fetch_after), in the index walk and the chat's walks. Locally with 30 ms each way: the giant list test passes (card in ~20 s); devserver integration tests pass.

littlefox added 1 commit 2026-10-23 07:17:15 +00:00

Thread walks down fetch what comes after, not a level per round trip ... e47006c975

Building a thread's index walks down level by level, and a chain has
one text per level: the 3 000-long chain was 3 000 round trips in a
row, three minutes at 60 ms. On CI (and on a real server) the giant
thread's card never came. Now a narrow level fetches the 200 texts
after it in its channel at once (fetch_after), as the walk up already
fetched the texts before it; the chat's own walks down and up do the
same. With 30 ms each way locally, the giant list's card now comes in
about 20 s, the thread in 15.

Refs #285

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

littlefox referenced this pull request 2026-10-23 07:39:15 +00:00

Remember each text's thread root on the device #288

littlefox added 1 commit 2026-10-23 08:17:55 +00:00

Channel chat stays at its end while the page comes in; mute spec waits ... bba0bfc3d2

A channel shown as a chat scrolled to its end when the first post came,
and the rest coming in above could leave the newest out of view on a
busy server or device (failed on CI, Android). It now stays at the end
until the reader scrolls (any scroll but its own) or five seconds pass.
The mute spec opened a post's menu before its thread arrived, and the
menu closed when the post was drawn again in it.

Refs #286

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

littlefox added 1 commit 2026-10-23 08:53:13 +00:00

Tests: time and channel-views retry what a redraw can undo ... 88479e4e30

time.spec's second tap could land on a time drawn anew when the thread
arrived (as its first tap already allowed for); channel-views looked for
the newest post before a busy machine had it in view. Both failed on
CI (Android, desktop).

Refs #286

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XJW2ocwLunyaLiUwCBHEP8

littlefox merged commit a9d7126ea8 into main 2026-10-23 09:30:41 +00:00

littlefox referenced this pull request from a commit 2026-10-23 09:30:43 +00:00

Merge pull request 'Test: room for a giant thread's card on a busy runner' (#305) from fix-giant-list-timeout into main

littlefox referenced this pull request 2026-10-23 10:05:45 +00:00

Change a post's requests afterwards: lock a thread for replies #292

littlefox commented 2026-10-23 10:05:46 +00:00

Author

Owner

Copy link

Merged; main run #533 green. Besides the fetch-ahead for walks down (giant thread card now ~20 s at 30 ms latency, not minutes): the channel chat stays at its end while its page comes in (any scroll but its own stops it), mute.spec waits for the thread before opening a menu, and time/channel-views retry what a redraw can undo.

Merged; main run #533 green. Besides the fetch-ahead for walks down (giant thread card now ~20 s at 30 ms latency, not minutes): the channel chat stays at its end while its page comes in (any scroll but its own stops it), mute.spec waits for the thread before opening a menu, and time/channel-views retry what a redraw can undo.

littlefox referenced this pull request from a commit 2026-10-23 11:55:54 +00:00

Merge main into issue-287-perf-budgets; chain budgets after #305

littlefox referenced this pull request 2026-10-23 11:56:09 +00:00

Threads 1e: a view per channel (Trådar, Chatt, Klassiskt) #286

littlefox referenced this pull request 2026-10-23 14:31:53 +00:00

Thread view: a burst of 400 calls stalls a step on a slow CPU #311

littlefox referenced this pull request 2026-10-23 17:48:09 +00:00

CI: speed and memory budgets from the perf run #287

littlefox referenced this pull request 2026-10-23 17:48:23 +00:00

Channels as lists of threads, threads as chats (Teams-style) #242

littlefox referenced this pull request 2026-10-24 01:54:11 +00:00

Flaky: a post's menu loses its buttons now and then (calm-post, aux specs) #324

littlefox referenced this pull request 2026-10-24 13:23:54 +00:00

Release 0.7.0 #328

littlefox referenced this pull request from a commit 2026-10-24 13:23:56 +00:00

Release 0.7.0

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

area/ci

Forgejo Actions, runner, tests

  

area/desktop

Desktop build

  

area/packaging

Arch, Debian, Fedora, NixOS packages

  

area/protocol

kom-proto, wire format, charset

  

area/reading

Unread tracking, reading order, review

  

area/ui

Dioxus app, layout, interaction

  

area/web

Web build and relay

  

area/writing

Composing and sending texts

  

kind/bug

Something is broken

  

kind/feature

New functionality

  

kind/infra

Tooling, tests, CI, dev environment

  

kind/request

requests for features or changes from users

  

needs/user

Waiting for a decision or action from littlefox

  

parked

Kept for reference; no work planned until the user says so

  

perf

Run the perf workflow and its budgets on this pull request (#287)

  

security

Security-relevant

  

upstream

A proposal for Protocol A or lyskomd, for the LysKOM development group

No labels

area/ci

area/desktop

area/packaging

area/protocol

area/reading

area/ui

area/web

area/writing

kind/bug

kind/feature

kind/infra

kind/request

needs/user

parked

perf

security

upstream

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

littlefox/neokom!305

No description provided.