feat(review): supersede stale items and inspect each file state once

A semantics table or resolver version change no longer leaves two pending
items for one file. A new item supersedes the other items of the same bytes
when every one is pending and undecided (revision 0, no decision row, the
self-heal's safe delete). Beside a decided or advanced item nothing new is
stored, and the rescan's stat shortcut skips those bytes under any version.

The inspection item version now folds a narrower fingerprint: the table
restricted to the graph's node classes plus its class-independent parts, so
an edit about other classes leaves every item alone.

The inspection profile runs once per file state. The scan resolves the
untraced file at batch level, skips files already queued or decided at their
current size and mtime under their version, and remembers files it found
untraced in this process. The capture reuses the scan's result when it holds
the same records.
This commit is contained in:
Vito Sansevero
2026-10-23 16:02:49 -07:00
parent 341757ea8e
commit 143b2a6b44
8 changed files with 705 additions and 126 deletions
+18 -8
View File
@@ -2622,18 +2622,28 @@ the scan (never the monitor) resolves it again with `INSPECTION_PROFILE`, which
knows `TextEncodeQwenImageEdit`, `StringConstantMultiline`, `JoinStringMulti`
and the rest of `web/js/inspection-semantics.json`. With at least two choosable
candidates there, the file is `ambiguous` and is queued from that profile's
evidence, which adds `profile: "inspection"` and `semantics` (the table's
`version`, SHA-256 of its canonical JSON); the item's `resolver_version` is
`inspection_review_version()`, 2^40 plus 40 bits of SHA-256 over the profile,
`RESOLVER_VERSION` and the table version. A table change therefore behaves like
a resolver change (I2): a rescan records the new evidence beside the pending
item, which stays applicable while the bytes still offer its chosen candidate.
evidence, which adds `profile: "inspection"` and `semantics`: the SHA-256 of
`class_projection`, the table restricted to the graph's node classes plus its
class-independent `schema` and `stage_inputs` (the resolver looks the table up
only by a graph node's class). The item's `resolver_version` is
`inspection_review_version(metadata)`, 2^40 plus 40 bits of SHA-256 over the
profile, `RESOLVER_VERSION` and that fingerprint. A table edit about the
graph's classes behaves like a resolver change (I2); an edit about other
classes changes nothing. One file never holds two pending items for the same
bytes: a new version's item supersedes the byte record's other items when all
are pending and undecided (revision 0, no decision row, the self-heal's safe
delete); beside a decided or advanced item nothing new is stored, and the
rescan's stat shortcut then skips those bytes under any version. A pending item
stays applicable while the bytes still offer its chosen candidate.
Apply re-resolves the capture with the profile the item recorded and re-checks
the chosen hash; the link goes through the same compatible-only ingestion. The
inspection profile never links on its own, and ingestion keeps the default
profile (D1). The heavier profile runs only for files already ambiguous without
a choice (twice per new file: once on the scanned metadata, once on the
capture; a rescan's stat shortcut skips the capture). The 16-candidate limited
a choice, and once per file state: the capture reuses the scan's result when
it holds the same records; a rescan skips files already queued or decided at
their current size and `mtime_ns` under their version, and files this process
already found untraced at the same state (an in-memory map of at most 65,536
files, oldest forgotten first). The 16-candidate limited
rule and the 8 MiB bound apply unchanged. The queue heals itself: after each batch, the scan removes the pending
items of every listed file (read or skipped as known) that a saved row now
links, and the monitor does the same for each ambiguous output it links