victormurcia

#Convert midi number column to a numpy array
midi_number = catterina_df['midi_number'].to_numpy()

degrees  = list(midi_number) # MIDI note number
track    = 0
channel  = 0
time     = 0   # In beats
duration = 1   # In beats
tempo    = 240  # In BPM
volume   = 100 # 0-127, as per the MIDI standard

victormurcia

# download a file from a URL, returns content of downloaded file
def download_url(urlpath):
    try:
        # open a connection to the server
        with urlopen(urlpath, timeout=3) as connection:
            # read the contents of the url as bytes and return it
            return connection.read()
    except:
        return None

victormurcia

# decode downloaded html and extract all <a href=""> links
def get_urls_from_html(content):
    
    # decode the provided content as ascii text
    html = content.decode('utf-8')
    
    # parse the document as best we can
    soup = BeautifulSoup(html, 'html.parser')
    
    # find all all of the <a href=""> tags in the document

victormurcia

# return all book unique identifiers from a list of raw links
def get_book_identifiers(links):
    # define a url pattern we are looking for
    pattern = re.compile('/ebooks/[0-9]+')
    # process the list of links for those that match the pattern
    books = set()
    for link in links:
        # check of the link matches the pattern
        if not pattern.match(link):
            continue

victormurcia

# download one book from project gutenberg
def download_book(book_id, save_path):
    print(save_path)
    # construct the download url
    url = f'https://www.gutenberg.org/files/{book_id}/{book_id}-0.txt'
    # download the content
    data = download_url(url)
    if data is None:
        #print(f'Failed to download {url}')
        url = f'https://www.gutenberg.org/files/{book_id}/{book_id}.txt'

victormurcia

def download_all_books(url, save_path):
    # download the page that lists top books
    data = download_url(url)
    print(f'.downloaded {url}')
    # extract all links from the page
    links = get_urls_from_html(data)
    print(f'.found {len(links)} links on the page')
    # retrieve all unique book ids
    book_ids = get_book_identifiers(links)
    print(f'.found {len(book_ids)} unique book ids')

victormurcia

book_id = '\\174.txt'          #What book are we processing?
book_name = book_dir + book_id #Location of book

#Open the book
book = open(book_name, "r", encoding="utf8") 
#Assign the book a name as string
book = str(book.read()) 
#Use regex to split the book into chapters by finding instances of the word CHAPTER
chapters = re.split("CHAPTER ", book)
#Remove first 21 CHAPTER instances since they are just fluff

victormurcia

chapter_list = []
for file in os.listdir("."):
    if file.endswith(".txt"):
        chapter_list.append(file)

#Do a natural sort on the texts files 
chapter_list.sort(key=lambda x: '{0:0>8}'.format(x).lower())
chapter_list

victormurcia

with open('2.txt', encoding = 'utf-8') as f:
    contents = f.read().rstrip()
print(contents[0:248])

victormurcia

chap_token = word_tokenize(contents)
del(chap_token[0:2])