Six stray semicolons, three specs recovered, and two edits withdrawn - #2924
Merged
Conversation
The census's work queue had two rows needing no owner decision: a closing brace
with nothing open (4 specs) and a statement terminated twice (2). Both probed
first -- `struct X { };`, `} else { };` and `print(x);;` each fail, and each
parses with the extra semicolon removed.
The census names only the FIRST occurrence, because that is all the compiler
reports. Fixing that one line moved each error a few lines forward and recovered
nothing. Fixing every occurrence of the same form:
specs/bus/pubsub.t27 2
specs/fpga/testbench/mac_tb.t27 6
specs/fpga/testbench/uart_tb.t27 5
specs that parse 615 -> 618, zero regressions
21 lines changed, every one exactly one character shorter, no declaration
touched
TWO EDITS WITHDRAWN. `hybrid_arithmetic.t27` and `relay_observer.t27` took the
same repair, advanced seven lines each, and still do not parse -- they hit a
different construct. The edits also made their seals stale and turned
`check_seal_coverage.py` red. An edit that buys no measurable change and costs a
re-seal is not a repair, so both are reverted. The gate caught that, not my
judgement.
The corpus ratchet caught the three fixed specs immediately as UNEXPECTED PASS
-- #2913's lesson working. Entries removed, max_entries 150 -> 147, RATCHET
CLEAN. `git grep` over the three names found four more files that mention them;
none is read by any gate, so they are snapshots rather than ledgers.
`tri ledgers audit` refuses to run while a ledger is uncommitted, which it did
mid-iteration. That is the guard, not a failure.
`bootstrap/src/compiler.rs` untouched; FROZEN_HASH does not move.
Refs #2864
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
gHashTag
enabled auto-merge (squash)
August 30, 2026 00:06
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
Contributor
PR DashboardGenerated at: 2026-08-30 00:06:25 UTC
Summary
Seal Status
|
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
Contributor
PR DashboardGenerated at: 2026-08-30 00:07:30 UTC
Summary
Seal Status
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The census's work queue had two rows needing no owner decision: a closing
brace with nothing open (4 specs) and a statement terminated twice (2). Both
probed first —
struct X { };,} else { };andprint(x);;each fail, andeach parses with the extra semicolon removed.
The census names the first occurrence, not the population
Fixing the one line it points at moved each error a few lines forward and
recovered nothing — because the same typo repeats. Fixing every occurrence
of the same form:
specs/bus/pubsub.t27specs/fpga/testbench/mac_tb.t27specs/fpga/testbench/uart_tb.t27Two edits withdrawn
hybrid_arithmetic.t27andrelay_observer.t27took the same repair, advancedseven lines each, and still do not parse — they hit a different construct.
The edits also made their seals stale and turned
check_seal_coverage.pyred.An edit that buys no measurable change and costs a re-seal is not a repair, so
both are reverted. The gate caught that, not my judgement — I would have
shipped them.
The ledger caught it immediately
The corpus ratchet reported the three fixed specs as
UNEXPECTED PASSon thefirst run — #2913's lesson working.
Entries removed,
max_entries150 → 147,RATCHET: CLEAN.git grepover the three spec names found four more files that mention them(
docs/reports/CORPUS-*.tsv,conformance/*.json,architecture/graph_v2.json).None is read by any gate — they are snapshots, not ledgers.
A note on one red
tri ledgers auditreported red mid-iteration because a ledger had uncommittedchanges and the command refuses to run in that state — it rewrites and restores
what it touches. That is the guard, not a failure. It passes on a clean tree.
Gates, run locally
specs generate 0 · verilog widths 0 · specs parse 0 · conflict markers 0 ·
seal coverage 0 · seals fresh 0 · types ratchet 0 · skill check 0 ·
ledgers audit 0 ·
cargo test -p tri0 · bootstrap test ratchet 0 ·corpus ratchet CLEAN.
bootstrap/src/compiler.rsis untouched, soFROZEN_HASHdoes not move.Refs #2864