Cloudflare Pages의 Clean URL(308 Redirect) 처리와 루트 절대 경로 표준화Handling Cloudflare Pages Clean URLs (308 Redirect) and Root-Relative Path Standardization

from __future__ import annotations
import re
from pathlib import Path

REPO_ROOT = Path(__file__).resolve().parents[1]
SITE_ORIGIN = "https://www.lysc.kr"

HREF_PATTERN = re.compile(r'(?:href|src|content)=["\']([^"\']+)["\']')
CANONICAL_PATTERN = re.compile(r'<link\s+rel="canonical"\s+href="([^"]+)"')

def check_html_file(html_path: Path) -> list[str]:
    """단일 HTML 파일 내 상대 경로 사용 여부와 .html 확장자 노출을 검증합니다."""
    errors: list[str] = []
    content = html_path.read_text(encoding="utf-8")

    # 1. Canonical URL에 .html 확장자가 포함되었는지 검사
    canonical_match = CANONICAL_PATTERN.search(content)
    if canonical_match:
        canon_url = canonical_match.group(1)
        if canon_url.endswith(".html"):
            errors.append(f"[CANONICAL] .html 확장자 포함 금지: {canon_url}")
        if not canon_url.startswith(SITE_ORIGIN):
            errors.append(f"[CANONICAL] 도메인 오리진 불일치: {canon_url}")

    # 2. 내부 링크에서 상대 경로(../) 및 .html 접미사 검사
    for match in HREF_PATTERN.finditer(content):
        url = match.group(1)
        if url.startswith(("http://", "https://", "mailto:", "#", "data:")):
            continue
        if url.startswith("./") or url.startswith("../"):
            errors.append(f"[REL-PATH] 상대 경로 대신 루트 절대 경로(/) 사용 필요: {url}")
        if url.startswith("/blog/posts/") and url.endswith(".html"):
            errors.append(f"[308-RISK] 내부 링크에 .html 확장자 노출: {url}")

    return errors

from __future__ import annotations
import re
from pathlib import Path

REPO_ROOT = Path(__file__).resolve().parents[1]
SITE_ORIGIN = "https://www.lysc.kr"

HREF_PATTERN = re.compile(r'(?:href|src|content)=["\']([^"\']+)["\']')
CANONICAL_PATTERN = re.compile(r'<link\s+rel="canonical"\s+href="([^"]+)"')

def check_html_file(html_path: Path) -> list[str]:
    """Verify that a single HTML file avoids relative paths and exposed .html extensions."""
    errors: list[str] = []
    content = html_path.read_text(encoding="utf-8")

    # 1. Check whether the canonical URL contains a .html extension
    canonical_match = CANONICAL_PATTERN.search(content)
    if canonical_match:
        canon_url = canonical_match.group(1)
        if canon_url.endswith(".html"):
            errors.append(f"[CANONICAL] Must not end with .html: {canon_url}")
        if not canon_url.startswith(SITE_ORIGIN):
            errors.append(f"[CANONICAL] Domain origin mismatch: {canon_url}")

    # 2. Check internal links for relative paths (../) and .html suffixes
    for match in HREF_PATTERN.finditer(content):
        url = match.group(1)
        if url.startswith(("http://", "https://", "mailto:", "#", "data:")):
            continue
        if url.startswith("./") or url.startswith("../"):
            errors.append(f"[REL-PATH] Use root-relative path (/) instead of relative path: {url}")
        if url.startswith("/blog/posts/") and url.endswith(".html"):
            errors.append(f"[308-RISK] Exposed .html extension in internal link: {url}")

    return errors