add gimme_probabilities and replace_nones

methods for "converting" a list of ngrams into a list of probabilities
these ngrams occur in a given language, replacing "None" values on the
fly with a very low "probability of occurence" value

da_detector.py

@@ -90,6 +90,32 @@ class da_detector:
 		print(r_ngrams)
 		return r_ngrams
 
+	# return a probability list for a list of ngrams and a given language
+	def gimme_probabilities(self, lang_probs: dict, ngrams: list):
+		if not isinstance(lang_probs, dict):
+			raise TypeError("lang_probs has to be a dict")
+		if not isinstance(ngrams, list):
+			raise TypeError("ngrams has to be a list")
+
+		if len(lang_probs) == 0:
+			raise ValueError("empty lang_probs dict")
+		if len(ngrams) == 0:
+			raise ValueError("empty ngrams list")
+
+		# may contain None values if not found, hence uncleansed
+		uncleansed_probabilities = []
+		for ngram in ngrams:
+			uncleansed_probabilities.append(lang_probs.get(ngram))
+		return self.replace_nones(uncleansed_probabilities)
+
+	def replace_nones(self, probabilities: list):
+		if not isinstance(probabilities, list):
+			raise TypeError("not a list, bailing")
+		if len(probabilities) == 0:
+			raise ValueError("empty list, bailing")
+		# a pretty good bogus probability is a one close™ reasonably to zero
+		return [0.000000123 if n is None else n for n in probabilities]
+
 
 freqs_folder = "./freqs/"
 test_str = "what freaking ever, nobody cares one bit of a heck"