Files
openclaw/test/scripts/format-docs.test.ts
wings1029 796ee27b3a fix(scripts): keep format-docs output tails UTF-8 safe (#110825)
* fix(scripts): keep format-docs output tails UTF-8 safe

When formatter (oxfmt) error output exceeds the 16 KiB tail cap and the
retained byte window starts inside a multibyte UTF-8 character, the
diagnostic tail would begin with a replacement character (�).

Add a decodeUtf8Tail helper that skips leading UTF-8 continuation bytes
before decoding the byte suffix, matching the existing pattern in
scripts/run-additional-boundary-checks.mjs (#109167).

Keep the existing 16 KiB cap and ASCII behavior unchanged.

* refactor(scripts): share UTF-8-safe output tails

Co-authored-by: 陈志强0668000989 <chen.zhiqiang1@xydigit.com>

* test(scripts): prove real formatter UTF-8 failures

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
2026-07-29 08:29:24 -04:00

199 lines
6.2 KiB
TypeScript
Raw Permalink Blame History

// Format Docs tests cover the docs formatter helper process spawning.
import { spawnSync } from "node:child_process";
import fs from "node:fs";
import path from "node:path";
import { describe, expect, it } from "vitest";
import {
chunkFilesForCommand,
docsFiles,
formatDocs,
resolveOxfmtInvocation,
runOxfmt,
} from "../../scripts/format-docs.mjs";
import { outputTail } from "../../scripts/lib/output-tail.mjs";
import { createScriptTestHarness } from "./test-helpers.js";
const { createTempDir } = createScriptTestHarness();
function writeDocsFixture(root: string): void {
fs.mkdirSync(path.join(root, "docs"), { recursive: true });
fs.writeFileSync(path.join(root, "README.md"), "# OpenClaw\n", "utf8");
fs.writeFileSync(path.join(root, "docs", "guide.mdx"), "# Guide\n", "utf8");
}
describe("format-docs", () => {
it("wraps the Windows oxfmt.cmd shim through cmd.exe", () => {
const invocation = resolveOxfmtInvocation(["--write", "docs\\guide.mdx"], {
comSpec: "C:\\Windows\\System32\\cmd.exe",
existsSync: (candidate: string) => candidate.endsWith("oxfmt.cmd"),
platform: "win32",
repoRoot: "C:\\repo",
});
expect(invocation.command).toBe("C:\\Windows\\System32\\cmd.exe");
expect(invocation.args.slice(0, 3)).toEqual(["/d", "/s", "/c"]);
expect(invocation.args[3]).toContain("oxfmt.cmd");
expect(invocation.args[3]).toContain("--write");
expect(invocation.args[3]).toContain("docs\\guide.mdx");
expect(invocation.shell).toBe(false);
expect(invocation.windowsVerbatimArguments).toBe(true);
});
it("batches oxfmt invocations when docs exceed the command line budget", () => {
const root = createTempDir("openclaw-format-docs-batch-");
const calls: Array<{ args: string[]; command: string }> = [];
runOxfmt(
["docs/one.md", "docs/two.md", "docs/three.md"],
{
maxCommandLineBytes: 1,
repoRoot: root,
},
{
existsSync: () => false,
spawnSync: (command: string, args: string[]) => {
calls.push({ args, command });
return { status: 0, stderr: "", stdout: "" };
},
},
);
expect(calls).toHaveLength(3);
expect(calls.every((call) => call.command === process.execPath)).toBe(true);
expect(calls.map((call) => call.args.at(-1))).toEqual([
"docs/one.md",
"docs/two.md",
"docs/three.md",
]);
});
it("reports git and oxfmt spawn diagnostics", () => {
const root = createTempDir("openclaw-format-docs-failures-");
expect(() =>
docsFiles(root, {
spawnSync: () => ({
status: 128,
stderr: "fatal: not a git repository",
stdout: "",
}),
}),
).toThrow(/git ls-files failed:[\s\S]*exit status: 128[\s\S]*fatal: not a git repository/u);
expect(() =>
runOxfmt(
["README.md"],
{ repoRoot: root },
{
existsSync: () => false,
spawnSync: () => ({
status: 1,
stderr: "formatter stderr",
stdout: "formatter stdout",
}),
},
),
).toThrow(
/oxfmt failed:[\s\S]*command:[\s\S]*exit status: 1[\s\S]*formatter stderr[\s\S]*formatter stdout/u,
);
});
it("keeps real formatter failure tails UTF-8 safe", () => {
const root = createTempDir("openclaw-format-docs-utf8-tail-");
let message = "";
try {
runOxfmt(
["README.md"],
{ repoRoot: root },
{
existsSync: () => false,
spawnSync: () =>
spawnSync(
process.execPath,
["-e", 'process.stderr.write("你好" + "x".repeat(16_380)); process.exitCode = 1'],
{ encoding: "utf8", maxBuffer: 1024 * 1024, shell: false, timeout: 5_000 },
),
},
);
} catch (error) {
message = error instanceof Error ? error.message : String(error);
}
expect(message).toMatch(/oxfmt failed:[\s\S]*exit status: 1[\s\S]*stderr tail:\nx/u);
expect(message).not.toContain("<22>");
});
it("uses repository paths in write mode and temporary paths in check mode", () => {
const root = createTempDir("openclaw-format-docs-mode-");
writeDocsFixture(root);
const oxfmtFileArgs: string[][] = [];
const spawnSync = (command: string, args: string[]) => {
if (command === "git") {
return {
status: 0,
stderr: "",
stdout: "README.md\ndocs/guide.mdx\n",
};
}
oxfmtFileArgs.push(args.slice(-2));
return { status: 0, stderr: "", stdout: "" };
};
expect(
formatDocs(
{
check: false,
repoRoot: root,
root,
},
{
existsSync: fs.existsSync,
spawnSync,
},
),
).toEqual({ changed: [], fileCount: 2 });
expect(
formatDocs(
{
check: true,
repoRoot: root,
root,
},
{
existsSync: fs.existsSync,
spawnSync,
},
),
).toEqual({ changed: [], fileCount: 2 });
expect(oxfmtFileArgs[0]).toEqual(["README.md", "docs/guide.mdx"]);
expect(oxfmtFileArgs[1]?.every((filePath) => path.isAbsolute(filePath))).toBe(true);
expect(oxfmtFileArgs[1]?.every((filePath) => filePath.startsWith(root))).toBe(false);
});
it("keeps single oversized docs in their own command chunk", () => {
expect(chunkFilesForCommand(["docs/very-long-name.md"], ["--write"], 1)).toEqual([
["docs/very-long-name.md"],
]);
});
});
describe("outputTail UTF-8 safety", () => {
it("skips leading continuation bytes when the retained byte window splits a multibyte character", () => {
// "你好" (6 bytes) + 16380 x's = 16386 bytes, exceeding the 16 KiB cap.
// subarray(-16384) starts at byte 2 = 0xA0 (continuation of 你).
const bigText = "你好" + "x".repeat(16380);
const result = outputTail(bigText, 16 * 1024);
expect(result).not.toContain("<22>");
// The continuation byte is skipped; the tail starts with "好".
expect(result).toMatch(/^好x/);
});
it("passes through output that fits within the tail budget", () => {
expect(outputTail("hello world", 16 * 1024)).toBe("hello world");
});
});