#This program pulls up the apod website. It uses lxml to iterate over the html
#tags till it finds the image tag. It downloads and saves the image.
#Then it uses lxml to strip the text from the page. Finally it slices the image
#description block from the text and saves it as a text file.
#This programs depends on python3 and the lxml package.
#www.fallenhobbit.com
#fallenhobbit@gmail.com

#module imports
from urllib.request import urlopen
import io
from lxml import etree
from lxml import html
from lxml.html.clean import clean_html

#Saves image file scraped from url
def saveImg(img_url):
    with open("apod.jpg","wb") as f:
        r = urlopen(img_url).read()                         #open file
        f.write(r)                                          #write file
        f.close()                                           #close file

#Strip the explanation text from the web page
def stripText(text_url):
    tree = html.parse(text_url)                             #create parse tree of html file
    tree = clean_html(tree)                                 #strip html tags
    text = tree.getroot().text_content()                    #get all the text
    return " ".join(text.split())                           #fix whitespace problems

#locates and returns image description block within the page
def findBlock(stripped_text):
    x = stripped_text.find("Explanation")                   #start of image explanation block
    y = stripped_text.find("Tomorrow's picture:")           #end of explanation block
    return stripped_text[x:y]

#saves stripped text as a text file
def saveBlock(stripped_text):
    with open("apod.txt","w") as f:                         #open file
        f.write(stripped_text)                              #write file
        f.close()                                           #close file

#Scrapes image file from url
def stripImage():
    apod = "http://apod.nasa.gov/apod/"                     #url to strip
    u = io.TextIOWrapper(urlopen(apod), encoding='latin1')  #open text stream handle to url
    text = u.read()                                         #read the url
    parser = etree.HTMLParser()                             #create html parser
    tree = etree.fromstring(text, parser)                   #create parse tree
    for elt in tree.getiterator():                          #iterate over tree
        if elt.tag == "img":
            if elt.attrib.has_key("src"):                   #once we find the image tag
                img_url = apod + elt.get("src")
                saveImg(img_url)                            #download it
                break

def main():                                                 #main function that calls everything else
    stripImage()
    saveBlock(findBlock(stripText("http://apod.nasa.gov/apod/")))

if __name__ == "__main__":
    main()