Slideshare Post Retrieval and Processing

get-slides.py

import hashlib
import requests
import sys
import time

def main():
    # Extract the keyword to search as the command-line argument
    keyword = ' '.join(sys.argv[1:])

    # Set up the auth for the request
    api_key = 'UHk20Fg6'
    shared_secret = 'p4lNW7jg'
    timestamp = int(time.time())
    m = hashlib.sha1()
    m.update(shared_secret.encode())
    m.update(str(timestamp).encode())

    # Set the parameters to pass in the request
    items_per_page = 50
    q = keyword
    detailed = 1

    # Send the request
    params = {
        "api_key": api_key,
        "ts": timestamp,
        "hash": m.hexdigest(),
        "items_per_page": items_per_page,
        "q": q,
        "sort": "mostviewed",
        "detailed": detailed,
        "lang": "en",
        "what": "tag"
    }
    r = requests.get("https://www.slideshare.net/api/2/search_slideshows", params=params)
    print(r.text)

if __name__ == '__main__':
    main()

process-xml.py

import codecs
import os
import sys
import datetime
from xml.etree import ElementTree

with codecs.open(sys.argv[1], encoding='utf-8') as f, \
        codecs.open(os.path.splitext(sys.argv[1])[0] + '.csv', 'w', encoding='utf-8') as w:
    tree = ElementTree.parse(f)
    w.write("Title,Author,Description,URL,Date,Views,Downloads,Comments,Favorites,Slides,Tags\n")
    for node in tree.iter('Slideshow'):
        title = node.find('Title').text.strip()
        description = node.find('Description').text.replace("\n", " ") if \
            node.find('Description').text is not None else ""
        url = node.find('URL').text.strip()
        date = datetime.datetime.strptime(node.find('Created').text.strip(), "%Y-%m-%d %H:%M:%S %Z")
        views = int(node.find('NumViews').text)
        downloads = int(node.find('NumDownloads').text)
        comments = int(node.find('NumComments').text)
        favorites = int(node.find('NumFavorites').text)
        slides = int(node.find('NumSlides').text)
        author = node.find('Username').text.strip()
        tags = [tag.text for tag in node.iter('Tag')]
        tag_string = ','.join(tags)
        w.write(u"\"{}\",\"{}\",\"{}\",{},{},{},{},{},{},{},\"{}\"\n".format(
            title, author, description, url, date.strftime("%m/%d/%Y"), views, downloads, comments,
            favorites, slides, tag_string))

requirements.txt

requests

Example usage:

pip install -r requirements.txt
# Retrieve the top machine learning posts directly from the API
python get-slides.py machine learning > machine-learning.xml 
# Put the relevant data from the XML file generated in the previous command into machine-learning.tsv
python process-xml.py machine-learning.xml