외부 프레임워크 없는 Python 표준 라이브러리 기반 정적 사이트 제너레이터(SSG) 설계Zero-Dependency Python Static Site Generator (SSG) Architecture

from __future__ import annotations
import html
import json
import re
from pathlib import Path

REPO_ROOT = Path(__file__).resolve().parents[1]
CONTENT_DIR = REPO_ROOT / "_content" / "posts"
OUTPUT_DIR = REPO_ROOT / "blog" / "posts"
BLOG_INDEX = REPO_ROOT / "blog" / "index.html"
POSTS_MARKERS = ("<!-- POSTS:START -->", "<!-- POSTS:END -->")

def inline_format(text: str) -> str:
    """HTML 이스케이프 후 인라인 마크다운(볼드, 코드, 링크)만 정규식으로 변환합니다."""
    escaped = html.escape(str(text), quote=True)
    escaped = re.sub(r"\*\*([^*]+)\*\*", r"<strong>\1</strong>", escaped)
    escaped = re.sub(r"`([^`]+)`", r"<code>\1</code>", escaped)
    escaped = re.sub(
        r"\[([^\]]+)\]\((https?://[^\)]+)\)",
        r'<a href="\2" target="_blank" rel="noopener noreferrer">\1</a>',
        escaped,
    )
    return escaped

def replace_between_markers(source_html: str, markers: tuple[str, str], payload: str) -> str:
    """HTML 내 주석 마커 사이의 구역을 새로 렌더링된 카드 목록으로 교체합니다."""
    start_tag, end_tag = markers
    if start_tag not in source_html or end_tag not in source_html:
        return source_html
    head, _, rest = source_html.partition(start_tag)
    _, _, tail = rest.partition(end_tag)
    return f"{head}{start_tag}\n{payload}\n        {end_tag}{tail}"

def clean_orphan_outputs(valid_posts: list[dict]) -> None:
    """원본 JSON이 삭제되었으나 출력 폴더에 남아 있는 고아 HTML을 제거합니다."""
    expected_files = {f"{p['date']}-{p['slug']}.html" for p in valid_posts}
    for html_path in OUTPUT_DIR.glob("*.html"):
        # 리다이렉트 스텁 파일은 삭제 대상에서 제외
        head_chunk = html_path.read_text(encoding="utf-8", errors="replace")[:1024]
        if 'http-equiv="refresh"' in head_chunk:
            continue
        if html_path.name not in expected_files:
            html_path.unlink()
            print(f"[DEL] 고아 HTML 제거: {html_path.name}")

from __future__ import annotations
import html
import json
import re
from pathlib import Path

REPO_ROOT = Path(__file__).resolve().parents[1]
CONTENT_DIR = REPO_ROOT / "_content" / "posts"
OUTPUT_DIR = REPO_ROOT / "blog" / "posts"
BLOG_INDEX = REPO_ROOT / "blog" / "index.html"
POSTS_MARKERS = ("<!-- POSTS:START -->", "<!-- POSTS:END -->")

def inline_format(text: str) -> str:
    """Escape HTML first, then convert inline markdown (bold, code, links) via regex."""
    escaped = html.escape(str(text), quote=True)
    escaped = re.sub(r"\*\*([^*]+)\*\*", r"<strong>\1</strong>", escaped)
    escaped = re.sub(r"`([^`]+)`", r"<code>\1</code>", escaped)
    escaped = re.sub(
        r"\[([^\]]+)\]\((https?://[^\)]+)\)",
        r'<a href="\2" target="_blank" rel="noopener noreferrer">\1</a>',
        escaped,
    )
    return escaped

def replace_between_markers(source_html: str, markers: tuple[str, str], payload: str) -> str:
    """Replace the section between HTML comment markers with newly rendered cards."""
    start_tag, end_tag = markers
    if start_tag not in source_html or end_tag not in source_html:
        return source_html
    head, _, rest = source_html.partition(start_tag)
    _, _, tail = rest.partition(end_tag)
    return f"{head}{start_tag}\n{payload}\n        {end_tag}{tail}"

def clean_orphan_outputs(valid_posts: list[dict]) -> None:
    """Remove orphaned HTML files in the output directory whose source JSON was deleted."""
    expected_files = {f"{p['date']}-{p['slug']}.html" for p in valid_posts}
    for html_path in OUTPUT_DIR.glob("*.html"):
        # Exclude redirect stub files from deletion
        head_chunk = html_path.read_text(encoding="utf-8", errors="replace")[:1024]
        if 'http-equiv="refresh"' in head_chunk:
            continue
        if html_path.name not in expected_files:
            html_path.unlink()
            print(f"[DEL] Removed orphan HTML: {html_path.name}")