strip punctuation from plain text dump

diff --git a/chloro/phyll/models.py b/chloro/phyll/models.py
index b72bfc509d33e445b12daa1a42093799cc4f3ea3..80acf9b40da68fd782b4294bdaeec59ce0c1fa08 100644 (file)
--- a/chloro/phyll/models.py
+++ b/chloro/phyll/models.py
@@ -15,6 +15,7 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 import html
+import re
 
 from django.db import models, transaction
 from django.utils.html import strip_tags
@@ -47,5 +48,7 @@ class Note(models.Model):
 
     def save(self, *args, **kwargs):
         if kwargs.get('update_fields') is None or 'plain_text' in kwargs.get('update_fields'):
-            self.plain_text = html.unescape(strip_tags(self.text))
+            self.plain_text = re.sub(
+                r"[’'\"«»,;:\. ]", ' ', html.unescape(strip_tags(self.text)).replace('’', ' ')
+            )
         return super().save(*args, **kwargs)