Release updated course archives #7
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
| name: Release updated course archives | |
| on: | |
| push: | |
| branches: | |
| - main | |
| paths: | |
| - "专业课/**" | |
| - "公共必修/**" | |
| - "公共必修课/**" | |
| - "公共选修/**" | |
| - "公共选修课/**" | |
| workflow_dispatch: | |
| inputs: | |
| full: | |
| description: "是否全量重建所有课程压缩包" | |
| required: true | |
| default: "true" | |
| type: choice | |
| options: | |
| - "true" | |
| - "false" | |
| permissions: | |
| contents: write | |
| concurrency: | |
| group: release-updated-courses | |
| cancel-in-progress: false | |
| jobs: | |
| package: | |
| name: Package updated courses | |
| runs-on: ubuntu-latest | |
| env: | |
| RELEASE_TAG: courses-latest | |
| steps: | |
| - name: Checkout repository | |
| uses: actions/checkout@v4 | |
| with: | |
| fetch-depth: 0 | |
| - name: Detect changed courses | |
| id: detect | |
| shell: bash | |
| env: | |
| EVENT_NAME: ${{ github.event_name }} | |
| FULL_REBUILD: ${{ github.event.inputs.full || 'false' }} | |
| BEFORE_SHA: ${{ github.event.before }} | |
| HEAD_SHA: ${{ github.sha }} | |
| run: | | |
| set -euo pipefail | |
| mkdir -p dist | |
| : > changed_courses.txt | |
| echo "event_name=$EVENT_NAME" | |
| echo "full_rebuild=$FULL_REBUILD" | |
| if [ "$EVENT_NAME" = "workflow_dispatch" ] && [ "$FULL_REBUILD" = "true" ]; then | |
| echo "Full rebuild mode." | |
| python3 > changed_courses.txt <<'PY' | |
| from pathlib import Path | |
| categories = [ | |
| "专业课", | |
| "公共必修", | |
| "公共必修课", | |
| "公共选修", | |
| "公共选修课", | |
| ] | |
| courses = [] | |
| for category in categories: | |
| category_path = Path(category) | |
| if not category_path.is_dir(): | |
| continue | |
| for course_path in sorted(category_path.iterdir(), key=lambda p: p.name): | |
| if course_path.is_dir(): | |
| courses.append(str(course_path)) | |
| for course in courses: | |
| print(course) | |
| PY | |
| else | |
| echo "Incremental rebuild mode." | |
| HEAD="$HEAD_SHA" | |
| BEFORE="${BEFORE_SHA:-}" | |
| if [ -z "$BEFORE" ] || [[ "$BEFORE" =~ ^0+$ ]] || ! git cat-file -e "$BEFORE^{commit}" 2>/dev/null; then | |
| BASE="$(git rev-parse "$HEAD^" 2>/dev/null || git rev-list --max-parents=0 "$HEAD")" | |
| else | |
| BASE="$BEFORE" | |
| fi | |
| echo "base=$BASE" | |
| echo "head=$HEAD" | |
| git diff --name-only --diff-filter=ACDMRT "$BASE" "$HEAD" > changed_files.txt | |
| python3 > changed_courses.txt <<'PY' | |
| from pathlib import Path | |
| categories = { | |
| "专业课", | |
| "公共必修", | |
| "公共必修课", | |
| "公共选修", | |
| "公共选修课", | |
| } | |
| courses = set() | |
| with open("changed_files.txt", "r", encoding="utf-8") as f: | |
| for line in f: | |
| path = line.strip() | |
| if not path: | |
| continue | |
| parts = Path(path).parts | |
| if len(parts) < 2: | |
| continue | |
| category = parts[0] | |
| course = parts[1] | |
| if category not in categories: | |
| continue | |
| courses.add(str(Path(category) / course)) | |
| for course in sorted(courses): | |
| print(course) | |
| PY | |
| fi | |
| echo "Changed courses:" | |
| cat changed_courses.txt || true | |
| if [ ! -s changed_courses.txt ]; then | |
| echo "has_changes=false" >> "$GITHUB_OUTPUT" | |
| else | |
| echo "has_changes=true" >> "$GITHUB_OUTPUT" | |
| fi | |
| if [ "$EVENT_NAME" = "workflow_dispatch" ] && [ "$FULL_REBUILD" = "true" ]; then | |
| echo "full_rebuild=true" >> "$GITHUB_OUTPUT" | |
| else | |
| echo "full_rebuild=false" >> "$GITHUB_OUTPUT" | |
| fi | |
| - name: Resolve English slugs | |
| id: slugs | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }} | |
| run: | | |
| set -euo pipefail | |
| # 从 Release 下载已有的 slug 映射（首次不存在则为空）。 | |
| # 映射存为 Release 资产 course-slugs.json，纯 ASCII，不受资产名净化影响。 | |
| rm -f course-slugs.json | |
| if gh release view "$RELEASE_TAG" >/dev/null 2>&1; then | |
| gh release download "$RELEASE_TAG" --pattern course-slugs.json --dir . --clobber 2>/dev/null || true | |
| fi | |
| [ -f course-slugs.json ] || echo '{}' > course-slugs.json | |
| python3 <<'PY' | |
| import hashlib | |
| import json | |
| import os | |
| import re | |
| import sys | |
| import urllib.request | |
| from pathlib import Path | |
| MAP_PATH = Path("course-slugs.json") # 持久映射：课程路径 -> 英文 slug | |
| RUN_MAP_PATH = Path("slug_map.json") # 本次需打包课程的子集 | |
| api_key = os.environ.get("DEEPSEEK_API_KEY", "").strip() | |
| try: | |
| slug_map = json.loads(MAP_PATH.read_text(encoding="utf-8")) or {} | |
| except Exception: | |
| slug_map = {} | |
| used = set(slug_map.values()) | |
| courses = [ | |
| line.strip() | |
| for line in Path("changed_courses.txt").read_text(encoding="utf-8").splitlines() | |
| if line.strip() | |
| ] | |
| def sanitize_slug(text: str) -> str: | |
| # 只保留 GitHub 资产名允许的字符 [A-Za-z0-9._-]，其余折叠成下划线 | |
| text = re.sub(r"[^A-Za-z0-9]+", "_", text.strip()) | |
| text = re.sub(r"_+", "_", text).strip("._") | |
| return text[:80] | |
| def short_hash(text: str, n: int = 6) -> str: | |
| return hashlib.sha1(text.encode("utf-8")).hexdigest()[:n] | |
| def is_ascii_simple(text: str) -> bool: | |
| # 课名本就是 ASCII（如 Matlab）时直接用，省一次 API | |
| return bool(re.fullmatch(r"[A-Za-z0-9 ._\-+()]+", text)) | |
| def translate(name: str) -> str: | |
| if not api_key: | |
| return "" | |
| payload = json.dumps({ | |
| "model": "deepseek-chat", | |
| "messages": [ | |
| {"role": "system", "content": "You translate Chinese university course names into concise English. Reply with ONLY the English course name in Title Case. No quotes, no explanation, no trailing punctuation."}, | |
| {"role": "user", "content": name}, | |
| ], | |
| "temperature": 0, | |
| "stream": False, | |
| }).encode("utf-8") | |
| req = urllib.request.Request( | |
| "https://api.deepseek.com/chat/completions", | |
| data=payload, | |
| headers={ | |
| "Authorization": f"Bearer {api_key}", | |
| "Content-Type": "application/json", | |
| }, | |
| method="POST", | |
| ) | |
| try: | |
| with urllib.request.urlopen(req, timeout=60) as resp: | |
| data = json.loads(resp.read().decode("utf-8")) | |
| return data["choices"][0]["message"]["content"].strip() | |
| except Exception as exc: # 翻译失败不阻断整个发布，回退到哈希名 | |
| print(f"translate failed for {name!r}: {exc}", file=sys.stderr) | |
| return "" | |
| def assign_slug(course_path: str) -> str: | |
| name = course_path.split("/", 1)[1] if "/" in course_path else course_path | |
| if is_ascii_simple(name): | |
| base = sanitize_slug(name) | |
| else: | |
| base = sanitize_slug(translate(name)) | |
| if not base: | |
| base = "course_" + short_hash(course_path, 8) | |
| candidate = base | |
| # 不同课程译名/净化后撞车时，用课程路径哈希保证唯一，绝不再 clobber | |
| while candidate in used: | |
| candidate = f"{base}_{short_hash(course_path + candidate)}" | |
| return candidate | |
| run_map = {} | |
| changed = False | |
| for course in courses: | |
| if course in slug_map: | |
| run_map[course] = slug_map[course] | |
| continue | |
| slug = assign_slug(course) | |
| slug_map[course] = slug | |
| used.add(slug) | |
| run_map[course] = slug | |
| changed = True | |
| print(f"resolved {course} -> {slug}") | |
| RUN_MAP_PATH.write_text( | |
| json.dumps(run_map, ensure_ascii=False, indent=2), | |
| encoding="utf-8", | |
| ) | |
| MAP_PATH.write_text( | |
| json.dumps(slug_map, ensure_ascii=False, indent=2, sort_keys=True) + "\n", | |
| encoding="utf-8", | |
| ) | |
| print(f"this run: {len(run_map)} course(s); map total: {len(slug_map)}; map_changed={changed}") | |
| PY | |
| - name: Package changed courses | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| run: | | |
| set -euo pipefail | |
| python3 <<'PY' | |
| import json | |
| import zipfile | |
| from pathlib import Path | |
| dist = Path("dist") | |
| dist.mkdir(exist_ok=True) | |
| exclude_files = {".DS_Store", "Thumbs.db"} | |
| exclude_dirs = {".git", "__MACOSX"} | |
| run_map = json.loads(Path("slug_map.json").read_text(encoding="utf-8")) | |
| created = 0 | |
| for course, slug in run_map.items(): | |
| course_path = Path(course) | |
| if not course_path.is_dir(): | |
| print(f"skip deleted course: {course}") | |
| continue | |
| zip_path = dist / f"{slug}.zip" | |
| if zip_path.exists(): | |
| zip_path.unlink() | |
| with zipfile.ZipFile( | |
| zip_path, | |
| "w", | |
| compression=zipfile.ZIP_DEFLATED, | |
| compresslevel=6, | |
| allowZip64=True, | |
| ) as zf: | |
| for item in sorted(course_path.rglob("*")): | |
| if not item.is_file(): | |
| continue | |
| if item.name in exclude_files: | |
| continue | |
| if any(part in exclude_dirs for part in item.parts): | |
| continue | |
| # zip 内部保留原始中文路径，仅外部资产名用英文 slug | |
| arcname = item.relative_to(Path(".")) | |
| zf.write(item, arcname.as_posix()) | |
| print(f"created {zip_path}") | |
| created += 1 | |
| print(f"created {created} archive(s)") | |
| PY | |
| echo "Archives:" | |
| python3 <<'PY' | |
| from pathlib import Path | |
| for file in sorted(Path("dist").glob("*.zip")): | |
| print(file) | |
| PY | |
| - name: Build manifest | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| run: | | |
| set -euo pipefail | |
| python3 <<'PY' | |
| import json | |
| from pathlib import Path | |
| dist = Path("dist") | |
| dist.mkdir(exist_ok=True) | |
| slug_map = json.loads(Path("course-slugs.json").read_text(encoding="utf-8")) | |
| courses = [] | |
| for path, slug in sorted(slug_map.items()): | |
| parts = path.split("/") | |
| courses.append({ | |
| "asset": f"{slug}.zip", | |
| "path": path, | |
| "category": parts[0] if parts else "", | |
| "name": parts[1] if len(parts) > 1 else path, | |
| }) | |
| (dist / "manifest.json").write_text( | |
| json.dumps({"courses": courses}, ensure_ascii=False, indent=2), | |
| encoding="utf-8", | |
| ) | |
| print(f"manifest entries: {len(courses)}") | |
| PY | |
| - name: Check archive size | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| run: | | |
| set -euo pipefail | |
| python3 <<'PY' | |
| from pathlib import Path | |
| limit_mib = 1900 | |
| too_large = False | |
| for file in sorted(Path("dist").glob("*.zip")): | |
| size_bytes = file.stat().st_size | |
| size_mib = size_bytes / 1024 / 1024 | |
| print(f"{file}: {size_mib:.2f} MiB") | |
| if size_mib > limit_mib: | |
| print(f"Archive is too large: {file}") | |
| too_large = True | |
| if too_large: | |
| raise SystemExit("Some archives are too large.") | |
| PY | |
| - name: Create release if missing | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| run: | | |
| set -euo pipefail | |
| if gh release view "$RELEASE_TAG" >/dev/null 2>&1; then | |
| echo "Release already exists: $RELEASE_TAG" | |
| else | |
| gh release create "$RELEASE_TAG" \ | |
| --title "课程资料压缩包" \ | |
| --notes "自动生成课程压缩包。资产名为英文（见 manifest.json 对应中文课名）。首次可全量生成，后续只更新发生变化的课程。" | |
| fi | |
| - name: Reset stale assets on full rebuild | |
| if: steps.detect.outputs.has_changes == 'true' && steps.detect.outputs.full_rebuild == 'true' | |
| shell: bash | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| run: | | |
| set -euo pipefail | |
| echo "Full rebuild: clearing existing release assets (keeping map/manifest)." | |
| gh release view "$RELEASE_TAG" --json assets --jq '.assets[].name' 2>/dev/null \ | |
| | while IFS= read -r asset; do | |
| [ -z "$asset" ] && continue | |
| case "$asset" in | |
| course-slugs.json|manifest.json) continue ;; | |
| esac | |
| echo "deleting stale asset: $asset" | |
| gh release delete-asset "$RELEASE_TAG" "$asset" -y || true | |
| done | |
| - name: Upload archives and metadata | |
| if: steps.detect.outputs.has_changes == 'true' | |
| shell: bash | |
| env: | |
| GH_TOKEN: ${{ github.token }} | |
| run: | | |
| set -euo pipefail | |
| python3 <<'PY' | |
| import os | |
| import subprocess | |
| import time | |
| from pathlib import Path | |
| release_tag = os.environ["RELEASE_TAG"] | |
| files = sorted(Path("dist").glob("*.zip")) | |
| files.append(Path("dist/manifest.json")) | |
| files.append(Path("course-slugs.json")) | |
| uploaded = 0 | |
| for file in files: | |
| if not file.exists(): | |
| print(f"skip missing: {file}") | |
| continue | |
| print(f"Uploading: {file}") | |
| subprocess.run( | |
| ["gh", "release", "upload", release_tag, str(file), "--clobber"], | |
| check=True, | |
| ) | |
| uploaded += 1 | |
| time.sleep(1) | |
| print(f"Uploaded {uploaded} file(s).") | |
| PY |