Refactor title normalization

amanuensis/parser/helpers.py

@@ -0,0 +1,23 @@
+def normalize_title(title):
+	"""
+	Normalizes strings as titles:
+	- Strips leading and trailing whitespace
+	- Merges internal whitespace into a single space
+	- Capitalizes the first word
+	"""
+	cleaned = re.sub(r'\s+', " ", title.strip())
+	return cleaned[0:1].upper() + cleaned[1:]
+
+def titlesort(title):
+	"""
+	Strips articles off of titles for alphabetical sorting purposes
+	"""
+	lower = title.lower()
+	if lower.startswith("the "):
+		return lower[4:]
+	elif lower.startswith("an "):
+		return lower[3:]
+	elif lower.startswith("a "):
+		return lower[2:]
+	else:
+		return lower

amanuensis/parser/tokenizer.py

@@ -6,6 +6,7 @@ can be rendered by a renderer.
 
 import re
 
+from amanuensis.parser.helpers import normalize_title
 
 class Renderable():
 	def render(self, renderer):
@@ -54,8 +55,8 @@ class CitationSpan(SpanContainer):
 	"""A citation to another article"""
 	def __init__(self, spans, cite_target):
 		super().__init__(spans)
-		# Normalize citation target by eliminating most whitespace
-		self.cite_target = re.sub(r'\s+', " ", cite_target.strip())
+		# Normalize citation target
+		self.cite_target = normalize_title(cite_target)
 	def __str__(self):
 		return f"{{{' '.join([str(span) for span in self.spans])}:{self.cite_target}}}"