#!/usr/bin/env python3

import json
import os
import requests
import subprocess
import time
from datetime import datetime
from pathlib import Path

# Start downloading from this comic number
assert 'START_NUMBER' in os.environ

# Stop downloading at this comic number
assert 'END_NUMBER' in os.environ

START_NUMBER = int(os.environ['START_NUMBER'])
END_NUMBER = int(os.environ['END_NUMBER'])

for n in range(START_NUMBER, END_NUMBER + 1):

    # Skip https://xkcd.com/1608/ "Hoverboard"
    # This is not an image/comics but a small video game
    if n == 1608:
        continue

    # Also skip 1663. This is also a comics using JavaScript.
    # TODO how should we add these comics with JavaScript?
    if n == 1663:
        continue

    req = requests.get(url = f"https://xkcd.com/{n}/info.0.json")

    if req.status_code != 200:
        print(f"Could not fetch comic #{n}. Status code {req.status_code}")
        continue

    data = req.json()

    assert n == data['num']

    image_extension = Path(data['img']).suffix.lower()
    assert image_extension in [ '.jpg', '.jpeg', '.png', '.gif' ]

    node_id = f"xkcd_comic_{n}"

    blob_filename = f'{node_id}{image_extension}'

    node = {
        "@context": {
            "comicstrip": "dokk:vocab:comicstrip:",
            "comicstrip:license": { "@type": "@id" },
            "comicstrip:series": { "@type": "@id" },
            "blob": "dokk:vocab:blob:"
        },
        "@id": f"dokk:{node_id}",
        "@type": "comicstrip:ComicStripPanel",
        "comicstrip:series": "dokk:xkcd",
        "comicstrip:title": data['title'],
        "comicstrip:published": f"{data['year']}-{int(data['month']):02d}-{int(data['day']):02d}",
        "comicstrip:license": "dokk:license:CC-BY-NC-2.5",
        "comicstrip:website": f'https://xkcd.com/{n}/',
        # TODO
        # Most xkcd comics have a transcript posted on xkcd. This info covers details
        # like the comics title, the date and link to the image and the title text
        # Check out explainxkcd.com for a transcript of the text in the images.
        # "comicstrip:transcript":
        "comicstrip:xkcd_alt": data['alt'],
        "comicstrip:number": data['num'],
        "blob:at": {
            "@id": f"file:/images/{blob_filename}",
            "blob:primary_source": data['img'],
            "blob:retrieval_date": f"{datetime.now().year}-{datetime.now().month}-{datetime.now().day}"
        }
    }

    # Download image
    wget_ret = subprocess.run(['wget', '--quiet', '--output-document', f'images/{blob_filename}', data['img']])
    assert wget_ret.returncode == 0 # No errors

    # Save node to file
    with open(f'nodes/{node_id}.jsonld', 'w') as file:
        json.dump(node, file, indent=4)

    print(f'[done] {node_id}')

    # Be nice
    time.sleep(1)