A quick and dirty data munger/parser for migrating from ghost to raw markdown

parse_ghost_backup.py

#!/usr/bin/env python

import json
import os
import sys
import datetime

with open("old.json", "r") as f:
    raw = f.read()

old = json.loads(raw)

posts = old["db"][0]["data"]["posts"]
postcount = len(posts)

# Data Munging
tags = {}
for item in old["db"][0]["data"]["tags"]:
    tags[item["id"]] = item["slug"]
post_tags = {}
for item in old["db"][0]["data"]["posts_tags"]:
    post_tags.setdefault(item["post_id"], []).append(tags[item["tag_id"]])

# Kickoff the processing
print("Found %s posts" % postcount)
if not os.path.exists('content'):
    print("I don't think we're in the pelican directory. Exiting")
    sys.exit(1)


def find_post_tags(postid):
    if post_tags.has_key(postid):
        return ", ".join(post_tags[postid])
    return "untagged"

for post in posts:
    date = datetime.datetime.fromtimestamp(post["created_at"] / 1e3)
    filename = "{}-{}.md".format(date.strftime("%Y-%m-%d"), post["slug"])
    print "Processing {}".format(filename)
    with open("content/{}".format(filename), 'w+') as f:
        f.write("Title: {}\n".format(post['title']))
        f.write("Date: {}\n".format(date.strftime("%Y-%m-%d %H:%m")))
        f.write("Tags: {}\n".format(find_post_tags(post["id"])))
        f.write("Slug: {}\n".format(post['slug']))
        f.write("---\n")
        f.write("{}".format(post['markdown'].encode('utf8')))