Issue hygiene: #1682 duplicates half of #1646, and #2212 plus #2198's open boxes are already in #1688
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
调研方向
Start by reading the linked issues #1646, #1682, #1688, #2198, and #2212, especially their Done-when boxes and comments. Compare the audit claims with the referenced locations in lib/ui_w_data.eigs, src/eigenscript.h, src/tape_read.c, src/state.c, and src/vm.h; then record which items are confirmed or refuted here. Done when the duplicate ownership and remaining task_sched_trace references are resolved in the tracker.
由索引模型根据 Issue 内容生成。
描述
Read-only audit of the open tracker (audit area 3, issue hygiene) at main e9e5cd60. Two groups of open issues track the same defect under different numbers, so whoever fixes one may not close the other. Nothing was closed or edited by the audit.
1. #1682 is #1646's second defect
- #1646 (2026-10-06), section "Second defect: selection contrast":
lib/ui_w_data.eigsfills a highlighted ASCII cell with the highlight colour, then draws the character inhex_ascii/text_dim, which is close to it, so "the selected text ("EigenScript hex_view" in the screenshot) is unreadable". Its Done-when box 2: selected ASCII cells draw text in a contrasting colour, with a test. - #1682 (2026-10-08): selecting "EigenScript hex_view" in
examples/ui_hex.eigsshows "a solid highlight bar with no characters". Done-when: selected ASCII characters stay readable; a draw-list or pixel assertion covers a selected ASCII cell.
Same example, same selection, same fix. Still present on main: lib/ui_w_data.eigs:676 fills the cell with the highlight colour hc, and :688 draws the glyph with ascol, with no selected-text colour.
2. #2212 and #2198 (boxes 2 and 3) are inside #1688
- #1688 asks to delete the machinery left unreachable by #1677:
obs_cfg_sync, theO cfg/O winemitters and readers,trace_obs_window_binding,observer_slot_set_window, and thetask.cscheduler-trace helpers, with replay refusing old tapes. Its 2026-10-21 and 2026-10-24 comments list the stale source comments, includingsrc/eigenscript.h"set via set_observer_thresholds" andsrc/tape_read.c:173-174namingbuiltin_set_observer_scale/builtin_set_observer_thresholds. - #2212 (2026-10-30) asks to fix exactly those two comments (
src/eigenscript.h:2461andsrc/tape_read.c:173-174on main). - #2198's open boxes 2 and 3 ask for a decision on whether replay refuses an
O winrecord and removestrace_obs_window_binding/win_override. #1688's Done-when already chooses: delete the readers and refuse old-format tapes. - #1688 was waiting for "the tape-bump release". Its 2026-10-24 comment notes the bump to format 7 shipped without the removal; main is now at format 8 (
src/trace.h:32), and every function it names is still defined (src/trace.c:1572obs_cfg_sync,src/trace.c:1602trace_obs_window_binding,src/eigenscript.c:1269observer_slot_set_window,src/task.c:307/324/473).
Not in any of the three: src/state.c:310 ("task_sched_trace of 1 arms it from a program") and src/vm.h:691 ("builtins.c task_sched_trace") still describe the removed builtin as live. Both are listed in #1688's 2026-10-24 comment.
Expected vs actual
- Expected: one open issue per defect, so the fix's
Closesline retires all of it. - Actual: two numbers for the hex-view contrast bug, and three for the observer-config removal, with #2198 asking a question #1688 already answers.
Done when
- #1682 is closed as a duplicate of #1646 (or #1646's box 2 is moved to #1682 and #1646 says so), so exactly one open issue owns the selected-ASCII contrast fix.
- #2212 and #2198's boxes 2 and 3 are folded into #1688 (closed as duplicates with a pointer, or #1688 links them as sub-issues), and #1688's Done-when no longer says "in the tape-bump release" now that formats 7 and 8 have shipped.
- Whichever issue owns the comments lists
src/state.c:310andsrc/vm.h:691, and its grep box coverstask_sched_trace. - Each item above is marked confirmed or refuted in a comment here.
— Duet (orchestrator)
- 主要语言
- C
- 星标
- 3
- 派生
- 7
- 平均合并
- 6 小时 32 分钟
- 30 天内合并 PR
- 285
环境准备
在浏览器里用你自己的 GitHub 账号启动这个项目的开发容器。
- 提供 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
InauguralSystems/EigenScript 的其他 Issue
-
good first issue needs-triage
难度 1/5 1 小时以内 新手友好度 88/100
InauguralSystems/EigenScript#2347 · 1 条评论 ·
维护者通常 1 天内回复
-
needs-triage
难度 1/5 1 小时以内 新手友好度 82/100
InauguralSystems/EigenScript#2335 · 1 条评论 ·
维护者通常 1 天内回复
-
area:embed kind:bug
难度 2/5 1-3 小时 新手友好度 62/100
InauguralSystems/EigenScript#2311 · 1 条评论 ·
维护者通常 1 天内回复
-
area:gates kind:gate-defect
难度 2/5 1-3 小时 新手友好度 62/100
InauguralSystems/EigenScript#2305 · 1 条评论 ·
维护者通常 1 天内回复
-
area:ci bug found-by:duet-audit
难度 2/5 1-3 小时 新手友好度 62/100
InauguralSystems/EigenScript#2283 ·
维护者通常 1 天内回复
查看 InauguralSystems/EigenScript 的全部 Issue
相似的 Issue
-
bug
难度 2/5 1-3 小时 新手友好度 78/100
FujiNetWIFI/fujinet-config#379 ·
维护者通常 1 天内回复
-
Ship /Local/Applications, the Local domain's Applications folder可能已有人在做 @pkgdemon 今天认领。 未关闭area:base enhancement
难度 1/5 1 小时以内 新手友好度 94/100
nextbsd/nextbsd-userland#612 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100
zephyrproject-rtos/zephyr#125543 ·
维护者通常 2 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100
colbymchenry/codegraph#3277 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100