generateText.py

#! /usr/bin/python

import csv

class GetPostCreator(object):

	THEME_IDS = [1, 2, 3]

	def __init__(self, id, row):
		self.id = self.wrap_in_quotes(str(id))
		self.type = self.get_type(row[0])
		self.headline = self.wrap_in_quotes(row[1])
		self.supporting_text = self.wrap_in_quotes_or_null(row[2])
		self.tags = self.get_tags(row[7])
		self.source_attribution = self.wrap_in_quotes(row[4])
		self.source_url = self.wrap_in_quotes(row[3])
		self.author_name = self.wrap_in_quotes_or_null(row[5])
		self.author_url = self.wrap_in_quotes_or_null(row[6])
		self.created_at = "april28MinusDay(" + str(id) + ")"
		self.theme_id = self.wrap_in_quotes(str(GetPostCreator.THEME_IDS[id % len(GetPostCreator.THEME_IDS)]))
		self.image = self.wrap_in_quotes_or_null(row[9])
		self.embedded_posts = None
		self.is_in_story_stream = row[10] == "Yes"

	def get_type(self, type):
		if type == "Original" or type == "Video":
			return "FANDOM_ARTICLE"
		elif type == "Instagram":
			return "INSTAGRAM"
		elif type == "Tweet":
			return "TWEET"
		elif type == "Curated":
			return "OTHER"
		elif type == "Story Stream":
			return "STORY_STREAM"
		else:
			raise Exception("type not found")

	def wrap_in_quotes_or_null(self, val):
		if val:
			return self.wrap_in_quotes(val)
		return "null"

	def get_tags(self, tags):
		tag_dict = {
			"Featured": "featured()",
			"Games": "games()",
			"Movies": "movies()",
			"TV": "tv()"
		}
		tag_calls = []
		for tag in tags.split(","):
			stripped_tag = tag.strip()
			if stripped_tag in tag_dict:
				tag_calls.append(tag_dict[stripped_tag])

		if tag_calls:
			return "ImmutableList.of(" + ",".join(tag_calls) + ")"
		return "Collections.emptyList()"

	def get_embedded_posts(self):
		if not self.embedded_posts:
			return "null"
		return "ImmutableList.of(" + ",\n".join(map(lambda posts: posts.get_string_repr(), self.embedded_posts)) + ")"

	def wrap_in_quotes(self, val):
		return '"' + val + '"'

	def get_string_repr(self):
		return 'getPost(' + ",\n".join( [
			self.id,
			self.type,
			self.headline,
			self.supporting_text,
			self.tags,
			self.source_attribution,
			self.source_url,
			self.author_name,
			self.author_url,
			self.created_at,
			self.theme_id,
			self.image,
			self.get_embedded_posts()
		] ) + ')'

class GenerateText(object):

	def __init__(self):
		self.reader = csv.reader(open("output.csv", "rb"))
		next(self.reader) # skip header line

	def generateText(self):
		postId = 0
		posts = []
		for row in self.reader:
			postId += 1
			posts.append(GetPostCreator(postId, row))

		posts = self.combine_story_stream(posts)
		print ",\n".join(map(lambda method_call: method_call.get_string_repr(), posts))
			
	def combine_story_stream(self, posts):
		all_posts = []
		stories_in_stream = []
		story_stream = None
		for post in posts:
			if post.is_in_story_stream:
				stories_in_stream.append(post)
			elif post.type == "STORY_STREAM":
				story_stream = post
			else:
				all_posts.append(post)
		story_stream.embedded_posts = stories_in_stream
		all_posts.append(story_stream)
		return all_posts

if __name__ == "__main__":
	generateText = GenerateText()
	generateText.generateText()