#coding:utf-8
import piexel
import re

class Tokenizer:
	def __init__(self, conf, api):
		self.conf = conf
		self.api = api
		self.reload_tokens()

	def reload_tokens(self):
		self.tk = self.api.get_tokens()
		self.steps = list(set(t['step'] for t in self.tk))
		self.steps.sort()

	def get_tokens_step(self, step):
		return [t for t in self.tk if t['step'] == step]

	def tokenize(self, filename):
		found = {'lang':[], 'quality':[], 'subtitle':[]}
		for step in self.steps:
			for tok in self.get_tokens_step(step):
				if(not bool(int(tok['case_sensitive']))):
					reg = re.compile(tok['token'], re.IGNORECASE)
				else:
					reg = re.compile(tok['token'])
				if reg.search(filename):
					for tok_lang in tok['languages']:
						found['lang'].append(tok_lang['value'])
					for tok_qual in tok['qualities']:
						found['quality'].append(tok_qual['value'])
					for tok_sub in tok['subtitle_languages']:
						found['subtitle'].append(tok_sub['value'])
					filename = reg.sub(' ', filename)
		return filename, found