json2csv.py

#!/urs/bin/env python

import os
import json
import csv
import re

fields = [
	"AMT",
	"HERRED",
	"SOGN",
	"titleno",
	"oclc_record_number",
	"author",
	"title",
	"film_no",
	"film_title",
	"digital_film_no",
	"digital_film_rights",
	"fs_indexed",
	"url_FS_catalog_entry",
	"url_FS_film_view",
	"url_FS_index_search",
]

outfile = open('output.csv', 'w')
writer = csv.DictWriter(outfile, fields, delimiter="\t")
writer.writeheader()

ahs = re.compile(r'Den Danske Folkekirke\. ([^(]*)(?:\(([^),]+)(?:, (.*))?\))?')

for subdir, dirs, files in os.walk("json"):
	for file in files:
		data = json.load(open("/".join([subdir, file])))
		for note in data["film_note"]:
			out = dict()
			out["titleno"] = data["titleno"][0]
			out["oclc_record_number"] = data["oclc_record_number"][0]
			out["author"] = data["author"][0]["display_text"][0]
			out["title"] = data["title"][0]
			
			result = ahs.match(out["author"])
			if result:
				out["SOGN"] = result.group(1).strip()
				if result.group(3):
					out["HERRED"] = result.group(2)
					out["AMT"] = result.group(3)
				elif result.group(2):
					out["AMT"] = result.group(2)
			else:
				print("Kunne ikke matche: %s" % out["author"])
			
			out["film_no"] = note["filmno"][0]
			out["digital_film_no"] = note["digital_film_no"][0]
			out["digital_film_rights"] = note["digital_film_rights"][0]
			out["film_title"] = note["text"][0]
			if note.has_key("fs_indexed"):
				out["fs_indexed"] = note["fs_indexed"][0]
			else:
				out["fs_indexed"] = "N"
			out["url_FS_catalog_entry"] = "https://www.familysearch.org/search/catalog/" + out["titleno"]
			if note["digital_film_rights"][0] == "UNREST": #unrestricted; der findes ogsaa NO_ACC
				out["url_FS_film_view"] = "https://www.familysearch.org/search/film/%09d?cat=%s" % (int(out["digital_film_no"]), out["titleno"])
			if out["fs_indexed"] == "Y":
				out["url_FS_index_search"] = "https://www.familysearch.org/search/record/results?count=20&query=%2Bfilm_number%3A" + out["film_no"]
			writer.writerow({k:v.encode('utf8') for k,v in out.items()})

outfile.close()