chore: post-audit cleanup — quieter per-file log, dated timestamps, EN contract, fix TCP javadoc#65
Merged
Merged
Conversation
timestamps, fix stale TCP javadoc
Picks five LOW/INFO items off the v1.0.0 audit follow-up list. No
behaviour change for production logic; doc + ergonomics only.
Per-file log demoted to DEBUG (Puller.pullFile)
- Pre-v0.4.0 every per-file pull emitted an INFO line `pullFile relPath=...`.
On a 1000-file transfer that's 1k+ INFO lines, drowning out the
job-lifecycle events an operator actually wants to skim. Now DEBUG —
visible at ./mvnw -B test or with -Dlogback.level.dev.netcopy=DEBUG,
silent by default.
logback.xml — date in timestamp
- Pattern: %d{HH:mm:ss.SSS} -> %d{yyyy-MM-dd HH:mm:ss.SSS}. A grep across
log archives spanning days no longer mistakes 23:55:01 yesterday for
23:55:01 today. Container log consumers stamp lines themselves but
those stamps don't survive `kubectl logs --previous` / journalctl
exports / GH Actions UI.
TcpConnectionHandler class javadoc rewritten
- The existing javadoc described the v1 two-pass behaviour
("Once the digest is known we write the DataHead, then we re-read
the range and ship it") even though SERVER_PROTO_VER=2 has been the
default since v0.3.0 and the v2 single-pass path is what nearly every
client takes. Rewrote to describe v2 first (default), v1 second
(legacy / version-negotiation backstop). Anyone reading the code now
gets the actual flow.
data-formats.md fully English
- The frozen-contract document was bilingual (Russian + code). Public
v1.0 OSS doc should be English-only. Translated the remaining Russian
sentences (~7 spots: file header, type-enum aside, chunks.bitmap
section, WS direction headers, frame-layout intro, max-payload note,
ERR codes preamble). No content change.
build.log
- Removed the stray 2 MiB build.log from the worktree root. Already
caught by .gitignore via *.log so it never made it into git history;
was just polluting `ls`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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.
Summary
Picks five LOW/INFO items from the v1.0.0 audit follow-up list. No production-logic change; ergonomics + docs only.
Puller.pullFilelog demoted to DEBUGPre-v0.4.0 every per-file pull emitted an INFO line. On a 1000-file transfer that's 1k+ INFO lines drowning out the job-lifecycle events an operator actually wants to scan. Now DEBUG — silent by default, opt-in via
-Dlogback.level.dev.netcopy=DEBUG.logback.xml— date in timestampA
grepacross log archives spanning days no longer confuses23:55:01yesterday with23:55:01today. Container drivers stamp lines themselves but those stamps don't survivekubectl logs --previous/ journalctl exports / GitHub Actions UI.TcpConnectionHandlerclass javadoc rewrittenThe javadoc still described the v1 two-pass behaviour even though
SERVER_PROTO_VER = 2has been default since v0.3.0 and v2 single-pass is what nearly every client takes. Rewrote to describe v2 first (default), v1 second (legacy). Anyone reading the code now gets the actual flow.tasks/contracts/data-formats.mdfully EnglishThe frozen-contract document was bilingual (Russian + code). Public v1.0 OSS doc should be English-only. Translated ~7 remaining Russian spots (file header, type-enum aside, chunks.bitmap section, WS direction headers, frame-layout intro, max-payload note, ERR codes preamble). No semantic change.
build.logremoved from worktreeThe stray 2 MiB
build.logat the repo root was already caught by.gitignorevia*.logso it never made it into git history; just polluting localls. Removed.Test plan
mvn compile+mvn test -Dtest=ArchitectureTest(8/8 pass).🤖 Generated with Claude Code