minimal-agent 0.9.0 → 0.10.0

package/skills/ui-design/scripts/check_language_policy.py

@@ -0,0 +1,351 @@
+#!/usr/bin/env python3
+"""检查 ui-design 产物是否默认使用简体中文。
+
+该脚本面向 Markdown、HTML、JSON 等文档型产物，尝试找出用户可见内容中
+不必要的长英文片段。它不是自然语言判定器，而是交付前的守卫：发现可疑英文
+后由设计者判断是否属于品牌名、代码标识、路径、协议字段或行业必要缩写。
+"""
+
+from __future__ import annotations
+
+import argparse
+import html
+import json
+import re
+import sys
+from pathlib import Path
+
+
+DEFAULT_ALLOWLIST = {
+    "api",
+    "agent",
+    "aegis",
+    "amber",
+    "app",
+    "app-ui",
+    "apple",
+    "archetype",
+    "artifact",
+    "aesthetic",
+    "aqua",
+    "airy",
+    "babel",
+    "bento",
+    "block",
+    "body",
+    "border-box",
+    "browser",
+    "browser_check",
+    "canvas",
+    "cdn",
+    "chip",
+    "claude",
+    "cloudflare",
+    "console",
+    "code",
+    "component",
+    "components",
+    "chromium",
+    "css",
+    "codex",
+    "dashboard",
+    "data",
+    "data-screen-label",
+    "deck",
+    "demo",
+    "dark",
+    "design",
+    "direction",
+    "dna",
+    "dom",
+    "dji",
+    "e2e",
+    "edit",
+    "edit-mode",
+    "editmode",
+    "edits",
+    "ember",
+    "end",
+    "execution",
+    "export",
+    "flat",
+    "foreground",
+    "frame",
+    "generate",
+    "github",
+    "glass",
+    "grid",
+    "harbor",
+    "hash",
+    "hero",
+    "hidden",
+    "horizon",
+    "html",
+    "hybrid",
+    "icon",
+    "inline",
+    "javascript",
+    "json",
+    "jsx",
+    "iphone",
+    "kind",
+    "layer",
+    "layout",
+    "layout-report",
+    "ledger",
+    "listener",
+    "localstorage",
+    "macbook",
+    "macos",
+    "manifest",
+    "message",
+    "meridian",
+    "mockup",
+    "nebula",
+    "native",
+    "node",
+    "noto",
+    "obsidian",
+    "openai",
+    "optics",
+    "object-fit",
+    "override",
+    "page",
+    "padding",
+    "playbook",
+    "playwright",
+    "ppt",
+    "pptx",
+    "pptxgenjs",
+    "precision",
+    "prism",
+    "prompt",
+    "prototype",
+    "postmessage",
+    "python-pptx",
+    "react",
+    "reference",
+    "rubric",
+    "runtime",
+    "relay",
+    "raw",
+    "fps",
+    "hdr",
+    "iso",
+    "lut",
+    "saas",
+    "scale",
+    "scaling",
+    "script",
+    "scrollintoview",
+    "scrim",
+    "shape",
+    "skill",
+    "slide",
+    "solo",
+    "solar",
+    "starter",
+    "surface",
+    "transform",
+    "token",
+    "tokens",
+    "tweaks",
+    "ui",
+    "url",
+    "utc",
+    "viewport",
+    "vector",
+    "verdant",
+    "vi",
+    "violet",
+    "warp",
+    "web",
+    "workdir",
+    "workflow",
+    "xml",
+    "yaml",
+    "yml",
+    "ppt-style",
+    "co2e",
+    "nereid",
+    "star",
+}
+
+ENGLISH_WORD_RE = re.compile(r"[A-Za-z][A-Za-z0-9_-]*")
+FENCE_RE = re.compile(r"```.*?```", re.DOTALL)
+INLINE_CODE_RE = re.compile(r"`[^`\n]+`")
+URL_RE = re.compile(r"https?://\S+|file://\S+")
+PATH_RE = re.compile(r"(?:[./~\w -]+/)+[\w .@-]+")
+HEX_COLOR_RE = re.compile(r"#[0-9A-Fa-f]{3,8}\b")
+YAML_FRONTMATTER_RE = re.compile(r"\A---\n(.*?)\n---", re.DOTALL)
+YAML_KEY_RE = re.compile(r"^([A-Za-z_][\w-]*):", re.MULTILINE)
+HTML_SCRIPT_STYLE_RE = re.compile(
+    r"<(script|style)\b[^>]*>.*?</\1>", re.IGNORECASE | re.DOTALL
+)
+HTML_TAG_RE = re.compile(r"<[^>]+>")
+JSON_KEY_RE = re.compile(r'"[^"\n]+"\s*:')
+
+
+def mask_keep_newlines(match: re.Match[str]) -> str:
+    text = match.group(0)
+    return "\n" * text.count("\n") or " "
+
+
+def strip_markdown(text: str) -> str:
+    text = YAML_FRONTMATTER_RE.sub(
+        lambda match: YAML_KEY_RE.sub("", match.group(1)), text
+    )
+    text = FENCE_RE.sub(mask_keep_newlines, text)
+    text = INLINE_CODE_RE.sub(mask_keep_newlines, text)
+    text = HEX_COLOR_RE.sub(mask_keep_newlines, text)
+    text = URL_RE.sub(mask_keep_newlines, text)
+    text = PATH_RE.sub(mask_keep_newlines, text)
+    return text
+
+
+def strip_html(text: str) -> str:
+    text = HTML_SCRIPT_STYLE_RE.sub(mask_keep_newlines, text)
+    text = HTML_TAG_RE.sub(mask_keep_newlines, text)
+    text = html.unescape(text)
+    text = HEX_COLOR_RE.sub(mask_keep_newlines, text)
+    text = URL_RE.sub(mask_keep_newlines, text)
+    text = PATH_RE.sub(mask_keep_newlines, text)
+    return text
+
+
+def strip_json_keys(text: str) -> str:
+    return JSON_KEY_RE.sub(mask_keep_newlines, text)
+
+
+def normalize_for_scan(path: Path, text: str) -> str:
+    suffix = path.suffix.lower()
+    if suffix in {".html", ".htm"}:
+        return strip_html(text)
+    if suffix in {".md", ".markdown"}:
+        return strip_markdown(text)
+    if suffix == ".json":
+        return strip_json_keys(strip_markdown(text))
+    return strip_markdown(text)
+
+
+def is_allowed(word: str, allowlist: set[str]) -> bool:
+    lower = word.lower().strip("_-")
+    if not lower:
+        return True
+    if lower in allowlist:
+        return True
+    if len(lower) <= 3:
+        return True
+    if word.isupper() and len(word) <= 8:
+        return True
+    if re.fullmatch(r"[A-Z][a-z]+(?:[A-Z][a-z]+)+", word):
+        return True
+    if re.fullmatch(r"[a-z]+(?:-[a-z0-9]+)+", lower):
+        parts = lower.split("-")
+        return all(part in allowlist or len(part) <= 3 for part in parts)
+    if "_" in lower:
+        parts = lower.split("_")
+        return all(part in allowlist or len(part) <= 3 for part in parts)
+    return False
+
+
+def line_number(text: str, index: int) -> int:
+    return text.count("\n", 0, index) + 1
+
+
+def scan_file(path: Path, allowlist: set[str], min_len: int) -> list[tuple[int, str, str]]:
+    try:
+        raw = path.read_text(encoding="utf-8")
+    except UnicodeDecodeError:
+        return [(1, "无法按 UTF-8 读取", "请确认文件编码或跳过二进制文件")]
+
+    visible = normalize_for_scan(path, raw)
+    findings: list[tuple[int, str, str]] = []
+    seen: set[tuple[int, str]] = set()
+
+    for match in ENGLISH_WORD_RE.finditer(visible):
+        word = match.group(0)
+        if len(word) < min_len or is_allowed(word, allowlist):
+            continue
+        lineno = line_number(visible, match.start())
+        start = max(0, match.start() - 36)
+        end = min(len(visible), match.end() + 36)
+        snippet = " ".join(visible[start:end].split())
+        key = (lineno, word.lower())
+        if key in seen:
+            continue
+        seen.add(key)
+        findings.append((lineno, word, snippet))
+
+    return findings
+
+
+def parse_args(argv: list[str]) -> argparse.Namespace:
+    parser = argparse.ArgumentParser(
+        description="检查 Markdown/HTML/JSON 中是否存在不必要的长英文可见文案。"
+    )
+    parser.add_argument("files", nargs="+", help="要检查的文件路径。")
+    parser.add_argument(
+        "--allow",
+        action="append",
+        default=[],
+        help="额外允许的英文词，可重复传入，多个词用逗号分隔。",
+    )
+    parser.add_argument(
+        "--min-len",
+        type=int,
+        default=4,
+        help="从多少个字符开始检查英文词，默认 4。",
+    )
+    parser.add_argument(
+        "--json",
+        action="store_true",
+        help="以 JSON 格式输出检查结果。",
+    )
+    return parser.parse_args(argv)
+
+
+def main(argv: list[str]) -> int:
+    args = parse_args(argv)
+    allowlist = set(DEFAULT_ALLOWLIST)
+    for item in args.allow:
+        allowlist.update(word.strip().lower() for word in item.split(",") if word.strip())
+
+    report: dict[str, list[dict[str, str | int]]] = {}
+    exit_code = 0
+
+    for file_arg in args.files:
+        path = Path(file_arg)
+        if not path.exists():
+            report[str(path)] = [
+                {"line": 1, "word": "文件不存在", "snippet": "请检查路径"}
+            ]
+            exit_code = 1
+            continue
+
+        findings = scan_file(path, allowlist, args.min_len)
+        if findings:
+            exit_code = 1
+        report[str(path)] = [
+            {"line": line, "word": word, "snippet": snippet}
+            for line, word, snippet in findings
+        ]
+
+    if args.json:
+        print(json.dumps(report, ensure_ascii=False, indent=2))
+    else:
+        for filename, findings in report.items():
+            if not findings:
+                print(f"通过: {filename}")
+                continue
+            print(f"发现可疑英文: {filename}")
+            for item in findings:
+                print(
+                    f"  第 {item['line']} 行: {item['word']} | {item['snippet']}"
+                )
+
+    return exit_code
+
+
+if __name__ == "__main__":
+    raise SystemExit(main(sys.argv[1:]))