kickstarter_scraping

kickstarter_clawler.py

#coding: utf-8
import urllib
import json
import datetime
import os
import time

# http://www.kbremner.com/2014/03/15/kickstarter-part1.html
# https://github.com/markolson/kickscraper/wiki/Project

keyword = "" # all
category_id = 16 # technology

project_count = 0
total_hits = 0
end_page = 500

try:
    os.mkdir("result")
except:
    pass

for page_id in xrange(1, end_page):
    query = "http://www.kickstarter.com/projects/search.json?term=%s&category_id=%d&page=%d&sort=most_funded" % (keyword, category_id, page_id)
    respons_json = json.loads(urllib.urlopen(query).read())

    if len(respons_json["projects"]) == 0:
        break

    project_count += len(respons_json["projects"])
    total_hits = respons_json["total_hits"]

    print "progress", project_count, "/", total_hits, round(float(project_count)/total_hits * 100, 2), "%"

    for project in respons_json["projects"]:
        #print project["name"]
        #print project["category"]["slug"], "\t\t",project["name"]
        #print project["urls"]["web"]["project"]
        #print str(datetime.datetime.fromtimestamp(project["launched_at"])), str(datetime.datetime.fromtimestamp(project["deadline"]))
        #print project["pledged"], project["goal"], project["pledged"] / project["goal"],  project["backers_count"],  project["state"]

        filepath = "result/%d.json" % project["id"]
        fp = open(filepath, "w")
        fp.write(json.dumps(project, sort_keys=True, indent=2))
        fp.close()

    if project_count >= total_hits:
        break


    time.sleep(1)

    #print json.dumps(respons_json, sort_keys=True, indent=2)
    #data.append(respons_json)

kickstarter_result_to_csv.py

#coding:utf-8

import os
import glob
import json
import datetime
import csv

fp = open("kickstarter_result.csv", "wb")
cw = csv.writer(fp,lineterminator="\n")

cw.writerow([
    "category",
    "pledged","goal",
    "currency",
    "backers_count",
    "state",
    "name",
    "launched_at",
    "deadline",
    "blurb",
    "url"])

for filename in glob.glob("result/*.json"):
    project = json.loads(open(filename).read())

    items = []
    items.append(project["category"]["slug"])
    items.append(project["pledged"])
    items.append(project["goal"])
    items.append(project["currency"])
    items.append(project["backers_count"])
    items.append(project["state"])
    items.append(project["name"])
    items.append(str(datetime.datetime.fromtimestamp(project["launched_at"])))
    items.append(str(datetime.datetime.fromtimestamp(project["deadline"])))
    items.append(project["blurb"].replace("\n", " ").replace("\r", " ").replace("," , " "))
    items.append(project["urls"]["web"]["project"])

    out = []
    for item in items:
        if type(item) is unicode:
            item = item.encode("shift-jis", "ignore")
        else:
            item = str(item)
        out.append(item)
    cw.writerow(out)
fp.close()

csv data is here
https://dl.dropboxusercontent.com/u/26405542/kickstarter_result.csv