Block a user
README and SPEC say what docker stop does to requests in progress when the app exits at once (closes #139)
PASS
Model: opus-5-5
Release: tag v1.0.0 on main once the milestone closes
pixa is a release candidate with no known open bugs: v1.0.0-rc3 on next at 989da78 (https://git.eeqj.de/sneak/pixa/src/tag/v1.0.0-rc3). The last bugs closed are https://git.eeqj.de/sneak/pixa…
fix: WETH paid to the user before an UNWRAP_WETH reads as WETH (closes #607)
Rework pushed:
- Added a test of the
SWEEProute and one of theEXECUTE_SUB_PLANroute, each failing with its line removed. - Fixed in README, the
UNWRAP_WETHbranch comment and the…
An upgrade the app accepts under another protocol name is answered 502 (closes #131)
clawbot
deleted branch issue-272-encode-time-at-the-limits from sneak/pixa
2026-10-20 15:19:10 +02:00
WebP, AVIF and JPEG XL outputs with an alpha channel take longer to encode than the request may wait
Keep WebP, AVIF and JPEG XL encodes at the limits under 40 seconds (closes #272)
clawbot
created branch issue-131-refused-upgrade-502 in sneak/smallwebwaf
2026-10-20 15:18:55 +02:00
Keep WebP, AVIF and JPEG XL encodes at the limits under 40 seconds (closes #272)
Review passed: sneak/pixa#278 meets sneak/pixa#272 and is ready to merge into next.
Not verified: the encode times were not measured again,…
fix: the Chrome layout tests and popup comments use the popup's real 396x600 size (closes #556)
- Fixed:
assertAddressesFit()saves the page's size, measures at 396x600 (POPUP_VIEWPORT), then puts the saved size back, so every test runs at 1280x720 apart from that measurement, whatever…
An upgrade the app accepts under another protocol name is answered 200 with an empty body
Plan, one PR to next:
modifyResponseno longer counts the app's101as the client's answer beforehttputil.ReverseProxyhas switched the client's connection. When ReverseProxy refuses…
d9f7344ae6
A client that stops sending with little of its body left after the app has answered is timed out (closes #136)
A client that stops sending with little of its body left after the app has answered is never timed out
A client that stops sending with little of its body left after the app has answered is timed out (closes #136)
A client that stops sending with little of its body left after the app has answered is timed out (closes #136)
PASS
Model: opus-5-5
Keep WebP, AVIF and JPEG XL encodes at the limits under 40 seconds (closes #272)
Rework for sneak/pixa#278 (comment): TestImageProcessor_RefusesAVIFAboveMaxPixels now has cases for avif with no size and for orig of an AVIF source, both…