import requests
import sys
import subprocess

from datetime import date, datetime
from bs4 import BeautifulSoup

from config import config

def get_old_news():
    print("getting old news")
    year = int(date.today().strftime("%Y")) - 100
    century_ago = date.today().strftime("%d%%2F%m%%2F") + str(year)

    url = f"https://chroniclingamerica.loc.gov/search/pages/results/?state=Illinois&lccn=sn84031489&lccn=sn84031490&lccn=sn84031492&lccn=sn84025828&lccn=sn90053126&lccn=sn91055768&lccn=sn87062321&lccn=sn82014064&lccn=sn86056950&lccn=sn84025819&lccn=sn91055455&dateFilterType=range&date1={century_ago}&date2={century_ago}&language=&ortext=&andtext=&phrasetext=&proxtext=&proxdistance=5&rows=1&searchType=advanced&sequence=1"

    r = requests.get(url)
    soup = BeautifulSoup(r.text, features="lxml")
    try:
        href = soup.select(".highlite a")[0]["href"]
        link = "https://chroniclingamerica.loc.gov" + href
        return f'<div><a href="{link}">Old Chicago News from {year}</a></div>'
    except:
        pass
    return "No old news"

def get_today_wikipedia():
    print("getting today's wikipedia")
    full_url = "https://en.wikipedia.org/wiki/%s" % date.today().strftime("%B_%d")
    return '<div><a href="%s">Today\'s Wikipedia</a></div>' % (full_url)

def get_week_holidays():
    print("getting holidays")
    command = "calendar -f /data/calendar.mark -A 14".split(" ")
    output = subprocess.check_output(command)
    output = output.decode("utf-8").strip().split("\n")
    return "</li>\n<li>".join(output) or "No holidays this fortnight."

def get_crossword():
    date_str = date.today().strftime("%Y-%m-%d")
    url = f"https://simplydailypuzzles.com/daily-cryptic/index.html?puzz=dc1-{date_str}"
    return f'<div><a href="{url}">Cryptic Crossword</a></div>'

def get_today_wikiquote():
    print("getting today's wikiquote")
    full_url = "https://en.wikiquote.org/wiki/%s" % date.today().strftime("%B_%d")
    r = requests.get(full_url)
    soup = BeautifulSoup(r.text, features="lxml")
    table = str(soup.find(text="2020").parent.parent.next_sibling.next_sibling)
    table = table.replace('href="/', 'href="https://en.wikiquote.org/')
    return '<div style="border: 1px solid black">%s</div>' % table

def get_calvin_and_hobbes():
    print("getting calvin and hobbes")
    year = int(date.today().strftime("%Y")) % 9 + 1986
    comic_date = str(year) + date.today().strftime("/%m/%d")
    full_url = "https://www.gocomics.com/calvinandhobbes/%s" % comic_date
    r = requests.get(full_url)
    soup = BeautifulSoup(r.text, features="lxml")
    if not "Today on" in str(soup.title): # gocomics gives you today if 404
        comic_src = soup.select(".item-comic-image")[0].img["src"]
        return '<div><a href="%s">Calvin and Hobbes</a></div>' % (comic_src)
    else:
        return ""

def get_peanuts():
    print("getting peanuts")
    comic_date = date.today().strftime("%Y/%m/%d")
    full_url = "https://www.gocomics.com/peanuts/%s" % comic_date
    r = requests.get(full_url)
    soup = BeautifulSoup(r.text, features="lxml")
    comic_src = soup.select(".item-comic-image")[0].img["src"]
    return '<div><a href="%s">Peanuts</a></div>' % (comic_src)

# Sites without feeds/need to be checked themselves
def get_homepage_links():
    pages = [
            {"url": "https://aldaily.com/", "title": "Arts and Letters Daily"},
            {"url": "https://www.powerlanguage.co.uk/wordle/", "title": "Wordle"},
    ]
    page_links = [
        f'<div><a href="{item["url"]}">{item["title"]}</a></div>'
        for item in pages
    ] 
    return "</li>\n<li>".join(page_links)


def check_for_starts_with_line(lst, line):
    for md in lst:
        if line.startswith(md):
            return True
    return False

def get_thoreau():
    print("getting thoreau")
    year_int = int(date.today().strftime("%Y")) - 183
    year = str(year_int)
    year_stop = str(year_int+1)
    month_days = [
            date.today().strftime("_%b %-d."), 
            date.today().strftime("_%b. %-d."),
            date.today().strftime("_%B %-d."), 
            date.today().strftime("_%B. %-d.")
        ]
    filename = config["thoreau"]["journal1"]
    with open(filename) as f:
        lines = f.readlines()
        # Find lines that the year lies on
        i= 0
        year_start_idx = -1
        for i in range(len(lines)):
            if lines[i].startswith(year):
                year_start_idx = i+1
                break
        year_stop_idx = -1
        for i in range(year_start_idx, len(lines)):
            if lines[i].startswith(year_stop):
                year_stop_idx = i - 2
                break
        entry_start_idx = -1

        # Find the lines inside the year that the date lies on
        i = year_start_idx
        while i < year_stop_idx:
            if check_for_starts_with_line(month_days, lines[i]):
                entry_start_idx = i - 2
                i += 1
                break
            i += 1
        entry_end_idx = -1
        while i < year_stop_idx:
            if lines[i].startswith("_"):
                entry_end_idx = i - 2
                break
            i += 1

        # If found date, join the strings
        if entry_start_idx != -1 and entry_end_idx != -1:
            return "".join(lines[entry_start_idx:entry_end_idx])
        return "No Thoreau entry on " + month_days[0] + year

def year_progress_bar(width=50):
    day_of_year = float(datetime.now().strftime('%j'))
    count = width * (day_of_year / 365)
    year_string = "["
    for i in range(width):
        if i < count:
            year_string += "#"
        else:
            year_string += "_"
    year_string += "]"
    return f"<pre>Year: {year_string}</pre>"