fix(sync): robust git coupling — non-ASCII paths, config neutralization, runtime git

Address git-integration fragility (output is not parsed for control flow; we rely
on exit codes + plumbing — but porcelain BEHAVIOR is config-sensitive, and the
runtime image lacked git).

- listTrackedFiles: `git -c core.quotepath=false ls-files -z` + NUL split — fixes
  Cyrillic/UTF-8 vault filenames being returned octal-escaped/quoted
- Dockerfile: install git (node:22-slim ships none; the daemon shells out at runtime)
- VaultGit env: LC_ALL=C/LANG=C, GIT_PAGER=cat, GIT_TERMINAL_PROMPT=0; keep
  stripping GIT_DIR/GIT_WORK_TREE (cwd-isolation, §12)
- ensureRepo local config: core.autocrlf=false + core.safecrlf=false (protect §11
  byte-stability from a global autocrlf=true), commit.gpgsign=false, and
  core.attributesFile=/dev/null (neutralize a global clean/smudge filter that
  would rewrite the stored blob); commit uses --no-verify (skip injected hooks)
- assertGitAvailable() preflight: clear error if the git binary is missing
- tests: Cyrillic listTrackedFiles, LF byte-preservation of the stored blob,
  local-config neutralization incl. attributesFile (590+ green)
This commit is contained in:
vvzvlad
2026-06-17 00:15:17 +03:00
parent 531b320776
commit ec0a3d47c7
4 changed files with 203 additions and 7 deletions

View File

@@ -1,5 +1,5 @@
import { execFile } from 'node:child_process';
import { mkdtemp, rm, writeFile } from 'node:fs/promises';
import { mkdir, mkdtemp, rm, writeFile } from 'node:fs/promises';
import { tmpdir } from 'node:os';
import { join } from 'node:path';
import { promisify } from 'node:util';
@@ -106,6 +106,67 @@ describe('VaultGit (integration; temp repo)', () => {
expect(count.trim()).toBe('1');
});
it('ensureRepo neutralizes correctness-affecting LOCAL config', async () => {
if (!available) return;
const vault = await freshDir();
const git = new VaultGit(vault);
await git.ensureRepo();
// These LOCAL values neutralize a hostile GLOBAL/system config that would
// otherwise change porcelain BEHAVIOR and corrupt the vault (SPEC §11 for
// core.autocrlf; gpgsign/safecrlf for the headless daemon).
const localConfig = async (key: string): Promise<string> => {
const { stdout } = await execFileAsync(
'git',
['config', '--local', '--get', key],
{ cwd: vault },
);
return stdout.trim();
};
expect(await localConfig('core.autocrlf')).toBe('false');
expect(await localConfig('commit.gpgsign')).toBe('false');
expect(await localConfig('core.safecrlf')).toBe('false');
expect(await localConfig('core.attributesFile')).toBe('/dev/null');
// Idempotent: a second run leaves the same single values (no duplicates).
await git.ensureRepo();
expect(await localConfig('core.autocrlf')).toBe('false');
expect(await localConfig('commit.gpgsign')).toBe('false');
expect(await localConfig('core.safecrlf')).toBe('false');
});
it('preserves LF bytes verbatim on commit (SPEC §11: autocrlf=false)', async () => {
if (!available) return;
const vault = await freshDir();
const git = new VaultGit(vault);
await git.ensureRepo();
// Write content with explicit LF line endings. With a hostile
// core.autocrlf=true git would translate these to CRLF in the stored blob,
// breaking the byte-stable round-trip invariant. ensureRepo pins
// core.autocrlf=false locally, so the stored bytes must round-trip exactly.
const fileName = 'lf.md';
const content = 'line1\nline2\nline3\n';
await writeFile(join(vault, fileName), content, 'utf8');
await git.stageAll();
const made = await git.commit('add LF file', {
authorName: BOT_AUTHOR_NAME,
authorEmail: BOT_AUTHOR_EMAIL,
});
expect(made).toBe(true);
// Read the STORED blob (not the worktree file) and assert verbatim bytes:
// still LF-only, no CRLF translation.
const { stdout: stored } = await execFileAsync(
'git',
['--no-pager', 'show', `HEAD:${fileName}`],
{ cwd: vault, encoding: 'buffer' },
);
const storedBuf = stored as unknown as Buffer;
expect(storedBuf.includes(Buffer.from('\r\n'))).toBe(false);
expect(storedBuf.toString('utf8')).toBe(content);
});
it('ensureBranch creates the docmost branch from main', async () => {
if (!available) return;
const vault = await freshDir();
@@ -289,4 +350,53 @@ describe('VaultGit (integration; temp repo)', () => {
const all = await git.listTrackedFiles();
expect(new Set(all)).toEqual(new Set(['keep.md', 'note.txt']));
});
it('listTrackedFiles returns RAW UTF-8 Cyrillic paths (not octal-escaped/quoted)', async () => {
if (!available) return;
const vault = await freshDir();
const git = new VaultGit(vault);
await git.ensureRepo();
// The target wiki is Russian, so file names contain Cyrillic. With git's
// DEFAULT core.quotepath=true these come back as `"\320\232..."` from
// ls-files; `listTrackedFiles` must return them verbatim as UTF-8.
const topName = 'Колонка.md';
const nestedDir = 'Раздел';
const nestedName = 'Подстраница.md';
await writeFile(join(vault, topName), 'top\n', 'utf8');
await mkdir(join(vault, nestedDir), { recursive: true });
await writeFile(join(vault, nestedDir, nestedName), 'nested\n', 'utf8');
await git.stageAll();
await git.commit('add cyrillic files', {
authorName: BOT_AUTHOR_NAME,
authorEmail: BOT_AUTHOR_EMAIL,
});
const md = await git.listTrackedFiles('*.md');
// Exact UTF-8 names, forward-slash separated for the nested one — NOT an
// escaped/quoted form like `"\320\232..."`.
expect(new Set(md)).toEqual(
new Set([topName, `${nestedDir}/${nestedName}`]),
);
// Guard explicitly against the quotepath regression: no entry is quoted or
// contains a backslash escape sequence.
for (const p of md) {
expect(p.startsWith('"')).toBe(false);
expect(p.includes('\\')).toBe(false);
}
// No-glob listing also returns the raw Cyrillic names.
const all = await git.listTrackedFiles();
expect(all).toContain(topName);
expect(all).toContain(`${nestedDir}/${nestedName}`);
});
it('assertGitAvailable resolves when git is present', async () => {
if (!available) return;
const vault = await freshDir();
const git = new VaultGit(vault);
// No repo needed: it only probes `git --version` (and the vault dir need
// not even exist yet).
await expect(git.assertGitAvailable()).resolves.toBeUndefined();
});
});