ios: give synthesized text entry back-pressure from the field instead of a fixed characters-per-second bound
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 38/100
- Issue type
- Bug
- Clarity
- Mostly clear
- Activity status
- Active
- Tech stack
- ios, typescript
- Domain
- mobile-dev, performance, testing-qa
Research direction
Start with the synthesized-first-responder-replacement route and the --agent-device-text-entry-async-value-write fixture, then inspect TextEntryTiming.synthesizedCommitCeiling and the existing lane test. Done means back-pressure prevents truncation, the test ends with the exact value or a typed collision reason, and representative 11-, 60-, and 240-character timings are recorded in the PR.
Written by the indexing model from the issue text.
Description
Purpose
#2904 bounds the synthesized replacement burst at 12 characters/second so a field whose app owns its value can acknowledge it. A fixed pace is a trade, not a fix: the burst's wall clock is now linear in the text at ~83 ms per character, so fill loses throughput on long fields, and an app slower than one character interval still wins the race.
Required behavior
Text entry gets back-pressure from the field instead of a guessed constant: the runner keeps characters queued while the field keeps making progress and slows or stops when it does not, so
- a long
fillcosts roughly what the app can absorb rather than a fixed 83 ms per character, - an app whose write-back takes longer than one character interval at the current pace cannot silently truncate the value — the command either completes the text or fails with a typed reason naming the collision,
- no
fillreaches the runner's 45 s command budget for text a healthy host could have typed sooner.
Observable completion conditions
- Measured end-to-end
filldurations for representative text lengths (11 / 60 / 240 characters) on the same Simulator, before and after, recorded in the PR. - A lane test on the
--agent-device-text-entry-async-value-writefixture with a write-back slower than the shipped character interval (~150 ms), which #2904 documents as still corrupting a burst, ends with either the exact value or the typed collision reason. TextEntryTiming.synthesizedCommitCeiling's note about the pace's cost upstream of the ceiling is replaced by whatever the new mechanism owns.
Evidence
iPhone 16 Simulator, iOS 26.2, clean app (no write-back fixture), fill through the synthesized-first-responder-replacement route at 12 characters/second: 11 characters 2.2 s, 60 characters 6.7 s, 150 characters 13.6 s, 240 characters 21.1 s. Delivery completes inside the synthesize call, so the 10 s commit ceiling is not the binding limit; the 45 s command budget is, at roughly 500 characters.
Dependencies
Related: #2903 (the verified element route, same app behaviour, no pace bound at all). Supersedes the fixed constant in #2904 when it lands.
- Dominant language
- TypeScript
- Stars
- 4.7k
- Forks
- 304
- Avg merge
- 11h 29m
- Merged PRs (30d)
- 514
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from callstack/agent-device
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
callstack/agent-device#1869 ·
-
needs-triage
Difficulty 5/5 Over a week Newbie friendliness 32/100
callstack/agent-device#2948 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 55/100
callstack/agent-device#2946 ·
-
needs-triage
Difficulty 4/5 3-5 days Newbie friendliness 45/100
callstack/agent-device#2940 ·
-
bug ready-for-agent
Difficulty 5/5 Over a week Newbie friendliness 35/100
callstack/agent-device#2935 ·
All issues in callstack/agent-device
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
bcgov/bc-wallet-mobile#4761 · 1 comment ·
-
external-issue to-triage
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
-
area-deployment area-integrations triage:bot-seen
Difficulty 2/5 Half a day Newbie friendliness 86/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
-
refactor
Difficulty 2/5 1-3 hours Newbie friendliness 84/100