import csv
from bs4 import BeautifulSoup
from tkinter import Tk, filedialog

# --- Step 0: ask for main keyword ---
mkwrd = input("Main keyword? ").strip()

# --- Step 1: ask user to select the HTML file ---
Tk().withdraw()
filename = filedialog.askopenfilename(
    title="Select HTML file",
    filetypes=[("HTML Files", "*.html"), ("All Files", "*.*")]
)
if not filename:
    print("❌ No file selected. Exiting.")
    exit()

# --- Step 2: read HTML file ---
with open(filename, "r", encoding="utf-8") as f:
    soup = BeautifulSoup(f, "html.parser")

# --- Step 3: define CSV schema ---
headers = [
    "note_id", "Type", "Label", "URL", "Content", "Keywords",
    "Media URL", "Media Type", "Media File Name", "Emoji", "Position"
]
for i in range(1, 11):
    headers.append(f"link_{i}")
    headers.append(f"label_{i}")

rows = []

# --- Helper: clean cell values to prevent CSV spills ---
def clean_csv_value(val: str) -> str:
    if not val:
        return ""  # keep empty
    val = str(val)
    # normalize whitespace
    val = val.replace("\r", " ").replace("\n", " ").replace("\t", " ").strip()
    # escape dangerous Excel formulas (avoid CSV injection)
    if val.startswith(("=", "+", "-", "@")):
        val = "'" + val
    return val

# --- Step 4: process <li> elements ---
for pos, li in enumerate(soup.find_all("li"), start=1):
    anchors = li.find_all("a")
    images = li.find_all("img")

    # First anchor = main Label + URL
    label = clean_csv_value(anchors[0].get_text(strip=True)) if anchors else ""
    url = clean_csv_value(anchors[0]["href"]) if anchors and anchors[0].has_attr("href") else ""

    # Other anchors = link_n + label_n
    link_data = []
    for a in anchors[1:11]:  # up to 10 additional
        lbl = clean_csv_value(a.get_text(strip=True))
        href = clean_csv_value(a["href"]) if a.has_attr("href") else ""
        link_data.append((href, lbl))

    # pad missing slots with blanks
    while len(link_data) < 10:
        link_data.append(("", ""))

    # Image handling
    media_url = clean_csv_value(images[0]["src"]) if images and images[0].has_attr("src") else ""
    media_type = "image" if media_url else ""
    media_file_name = clean_csv_value(media_url.split("/")[-1]) if media_url else ""

    row = [
        "",  # note_id blank
        "link",  # default Type
        label,
        url,
        clean_csv_value(li.get_text(" ", strip=True)),  # Content
        clean_csv_value(mkwrd),  # Keywords
        media_url,
        media_type,
        media_file_name,
        "",  # Emoji
        pos
    ]

    # Add link_1,label_1 ... link_10,label_10
    for href, lbl in link_data:
        row.extend([href, lbl])

    rows.append(row)

# --- Step 5: write CSV with BOM for Excel ---
out_file = filename.rsplit(".", 1)[0] + "_converted.csv"
with open(out_file, "w", newline="", encoding="utf-8-sig") as csvfile:
    writer = csv.writer(csvfile, quoting=csv.QUOTE_ALL)  # force quoting all cells
    writer.writerow(headers)
    writer.writerows(rows)

print(f"✅ Conversion complete. CSV saved as: {out_file}")