Added emdash 0x2014 as a word separator

novel_stats/novel_stats.py

@@ -4,18 +4,21 @@
 import argparse
 import collections
 import tempfile
+import re
 
 CHAPTER_MARKER = '## '
 STATUS_MARKER = '[status]: # '
 ACT_MARKER = '[act]: # '
 # Standard markdown comment marker, supported by Pandoc and Calibre's ebook-convert.
 COMMENT_MARKER = '[//]: # '
+TITLE_MARKER = '# '
+WORD_SEPS = [' ','—']
 
 
 def count_words(line):
     count = 0
 
-    for word in line.strip().split(' '):
+    for word in re.split('|'.join(WORD_SEPS), line.strip()):
         if not word.strip() or word == '*' or word.startswith('#'):
             continue
 
@@ -74,6 +77,7 @@ def main():
     word_count_by_act = collections.defaultdict(int)
     status_by_chapter = {}
     current_status = None
+    title = None
 
     for line in mdfile.readlines():
         if line.startswith(CHAPTER_MARKER):
@@ -97,6 +101,10 @@ def main():
         elif line.startswith(ACT_MARKER):
             act_heading = line[len(ACT_MARKER) :].strip('()\n')
             word_count_by_act[act_heading] = count_words(act_heading)
+        elif line.startswith(TITLE_MARKER):
+            title = line[len(TITLE_MARKER):].strip()
+            line_word_count = count_words(line)
+            word_count_by_chapter[chapter_heading] += line_word_count
         elif line.startswith(COMMENT_MARKER):  # Don't count the words in a comment.
             pass
         else:
@@ -113,6 +121,9 @@ def main():
     word_count_by_act[act_heading] += word_count_by_chapter[chapter_heading]
     total_word_count += word_count_by_chapter[chapter_heading]
 
+    if title:
+        print(f'Novel Stats for {title.upper()}')
+
     # -c or --chapter to give a chapter-by-chapter word count summary.
     if arguments.chapter:
         for chapter_heading, chapter_word_count in word_count_by_chapter.items():