import os
import re

WIKI_DIR = r"C:\Program_coding\임도설계 및 견적자동화 프로그램 개발\docs\wiki"
PAGES_DIR = os.path.join(WIKI_DIR, "pages")
CONCEPTS_DIR = os.path.join(WIKI_DIR, "concepts")
INDEX_PATH = os.path.join(WIKI_DIR, "index.md")

errors = []
warnings = []
all_wiki_files = {}  # relative_path -> absolute_path

# Collect all markdown files in wiki/ (ignoring graphify-out and memory directories)
for root, dirs, files in os.walk(WIKI_DIR):
    # skip graphify-out or memory
    if "graphify-out" in root or "memory" in root:
        continue
    for file in files:
        if file.endswith(".md"):
            abs_path = os.path.join(root, file)
            rel_path = os.path.relpath(abs_path, WIKI_DIR).replace("\\", "/")
            all_wiki_files[rel_path] = abs_path
            simple_name = os.path.splitext(rel_path)[0]
            all_wiki_files[simple_name] = abs_path

# Read index.md content
with open(INDEX_PATH, "r", encoding="utf-8") as f:
    index_content = f.read()

# Pattern for wikilinks: [[link]] or [[link|label]]
wikilink_pat = re.compile(r"\[\[([^\]|]+)(?:\|[^\]]+)?\]\]")

for rel_path, abs_path in list(all_wiki_files.items()):
    if "/" not in rel_path and rel_path != "index" and rel_path != "log":
        continue
    if rel_path in ["index", "log", "index.md", "log.md"]:
        continue

    with open(abs_path, "r", encoding="utf-8") as f:
        lines = f.readlines()

    line_count = len(lines)
    content = "".join(lines)

    # Rule 14: Max 100 lines
    if line_count > 100:
        warnings.append(f"[Line Count] `{rel_path}.md` exceeds 100 lines ({line_count} lines).")

    # Check YAML Frontmatter via simple regex
    frontmatter_match = re.match(r"^---\s*\n(.*?)\n---\s*\n", content, re.DOTALL)
    if not frontmatter_match:
        errors.append(f"[Frontmatter] `{rel_path}.md` has no valid YAML frontmatter.")
        continue

    fm_text = frontmatter_match.group(1)

    # Simple regex parsing for status & page_id
    status_match = re.search(r"^status:\s*(\w+)", fm_text, re.MULTILINE)
    page_id_match = re.search(r"^page_id:\s*([^\n\r]+)", fm_text, re.MULTILINE)

    if not status_match:
        errors.append(f"[Status] `{rel_path}.md` has no status field in frontmatter.")
    else:
        status = status_match.group(1).strip()
        if status not in ["draft", "stable", "stale"]:
            errors.append(
                f"[Status] `{rel_path}.md` has invalid status '{status}'. Must be draft, stable, or stale."
            )
        elif status == "stale":
            warnings.append(
                f"[Stale Page] `{rel_path}.md` is marked as stale and needs updates from raw inputs."
            )

    # Rule 5: page_id required for pages/
    if "pages/" in rel_path:
        if not page_id_match:
            errors.append(f"[page_id] `{rel_path}.md` is in pages/ but lacks a 'page_id' field.")

    # Check for broken wikilinks in content
    links = wikilink_pat.findall(content)
    for link in links:
        link_clean = (
            link.strip().replace("\\", "/").split("#")[0]
        )  # ignore anchor for file existence check
        if not link_clean:
            continue
        found = False

        # 1. Absolute link from vault root (e.g. concepts/storage_paths)
        if link_clean in all_wiki_files:
            found = True
        # 2. Simple name match (e.g. storage_paths, B01_frontend)
        elif link_clean.split("/")[-1] in all_wiki_files:
            found = True
        # 3. Handle subpages like [[B01_Dashboard/B01_frontend]]
        elif f"pages/{link_clean}" in all_wiki_files:
            found = True
        elif f"concepts/{link_clean}" in all_wiki_files:
            found = True
        # 4. Handle nested directory patterns
        elif link_clean.startswith("../"):
            # Resolve relative link
            curr_dir = os.path.dirname(rel_path)
            resolved = os.path.normpath(os.path.join(curr_dir, link_clean)).replace("\\", "/")
            if resolved in all_wiki_files or resolved.replace("pages/", "") in all_wiki_files:
                found = True

        if not found:
            errors.append(f"[Broken Link] `{rel_path}.md` contains broken wikilink: [[{link}]]")

    # Check if this file is registered in index.md (avoid orphans)
    file_base = os.path.basename(abs_path)
    file_name_no_ext = os.path.splitext(file_base)[0]
    if file_name_no_ext not in index_content:
        # check if relative path is in index
        rel_path_no_ext = rel_path.replace(".md", "")
        if (
            rel_path_no_ext not in index_content
            and rel_path_no_ext.split("/")[-1] not in index_content
        ):
            warnings.append(
                f"[Orphan Page] `{rel_path}.md` is not linked or mentioned in index.md."
            )

print("=== LINT ERRORS ===")
for e in sorted(list(set(errors))):
    print(e)
print("\n=== LINT WARNINGS ===")
for w in sorted(list(set(warnings))):
    print(w)
print("\nLint completed.")