wp2tumblr.py

import pytumblr
import requests
from pprint import pprint
from re import findall
from urlparse import urlparse

tumblr_blog = 'blog.viktorpetersson.com'
wordpress_site = 'http://viktorpetersson.com/'
source_file = '/Users/mvip/Desktop/pages.txt'

client = pytumblr.TumblrRestClient(
    '<consumer_key>',
    '<consumer_secret>',
    '<oauth_token>',
    '<oauth_secret>',
)


def get_tumblr_posts():
    blog_posts = []
    chunk_size = 20
    current_fetch = 0
    post_count = client.posts(tumblr_blog)['total_posts']

    while current_fetch < post_count:
        posts = client.posts(tumblr_blog, offset=current_fetch, limit=chunk_size)

        for p in posts['posts']:
            blog_posts.append(p['post_url'])

        current_fetch += chunk_size

    return blog_posts


def parse_pages():
    pages = []
    f = open(source_file, 'r')
    for line in f:
        url = line.split()[0]
        pages.append(url)
    return pages


def match_src_and_dst(src, dst):
    pairs = []

    for s in src:
        found_match = False
        for d in dst:
            page_url = d.split('/')[-1]
            if page_url in s:
                pairs.append({'src': s, 'dst': d})
                found_match = True
                break
        if not found_match:
            print "Failed to find match for %s" % s

    return pairs


def generate_nginx_redirects(matched):
    for p in matched:
        src_path = urlparse(p['src'])
        print "rewrite ^%s$ %s permanent;" % (src_path.path, p['dst'])


source_pages = parse_pages()
destination_pages = get_tumblr_posts()
matched = match_src_and_dst(source_pages, destination_pages)

print "Src: " + str(len(source_pages))
print "Dst: " + str(len(destination_pages))
print "Matched: " + str(len(matched))

print generate_nginx_redirects(matched)