From 66f45d9cb78ddecccbab35f386be4dacc5cf3844 Mon Sep 17 00:00:00 2001
From: marwoodandrew <amarwood@aap.com.au>
Date: Mon, 30 Sep 2024 11:37:07 +1000
Subject: [PATCH] SDAAP-122 Handle fancy quotes in ANPA

---
 .../aap/publish/formatters/anpa_formatter.py   | 12 +++++++-----
 .../publish/formatters/anpa_formatter_test.py  | 18 ++++++++++++++++++
 2 files changed, 25 insertions(+), 5 deletions(-)

diff --git a/server/aap/publish/formatters/anpa_formatter.py b/server/aap/publish/formatters/anpa_formatter.py
index ba609a43a..956d910a8 100644
--- a/server/aap/publish/formatters/anpa_formatter.py
+++ b/server/aap/publish/formatters/anpa_formatter.py
@@ -17,7 +17,7 @@
 from .field_mappers.locator_mapper import LocatorMapper
 from .field_mappers.slugline_mapper import SluglineMapper
 from eve.utils import config
-from .unicodetoascii import to_ascii
+from .unicodetoascii import to_ascii, clean_string
 from .category_list_map import get_aap_category_list
 import re
 from superdesk.etree import parse_html, to_string, etree
@@ -104,7 +104,7 @@ def format(self, article, subscriber, codes=None):
                 anpa.append(b'\x0D\x0A')
 
                 if formatted_article.get('ednote', '') != '':
-                    ednote = '{}\r\n'.format(to_ascii(formatted_article.get('ednote')))
+                    ednote = '{}\r\n'.format(to_ascii(clean_string(formatted_article.get('ednote'))))
                     anpa.append(ednote.encode('ascii', 'replace'))
 
                 if formatted_article.get(BYLINE):
@@ -115,7 +115,7 @@ def format(self, article, subscriber, codes=None):
                     anpa.append(get_text(self.append_body_footer(formatted_article),
                                          content='html').encode('ascii', 'replace'))
                 else:
-                    body = to_ascii(formatted_article.get('body_html', ''))
+                    body = to_ascii(clean_string(formatted_article.get('body_html', '')))
                     # we need to inject the dateline
                     if formatted_article.get('dateline', {}).get('text') and not article.get('auto_publish', False):
                         body_html_elem = parse_html(formatted_article.get('body_html'))
@@ -125,7 +125,8 @@ def format(self, article, subscriber, codes=None):
                             body = to_string(body_html_elem)
                     anpa.append(self.get_text_content(body))
                     if formatted_article.get('body_footer'):
-                        anpa.append(self.get_text_content(to_ascii(formatted_article.get('body_footer', ''))))
+                        anpa.append(
+                            self.get_text_content(to_ascii(clean_string(formatted_article.get('body_footer', '')))))
 
                 anpa.append(b'\x0D\x0A')
                 anpa.append(mapped_source.encode('ascii'))
@@ -171,7 +172,8 @@ def get_text_content(self, content):
     def _process_headline(self, anpa, article, category):
         # prepend the locator to the headline if required
         article['headline'] = get_text(article.get('headline', ''))
-        headline = to_ascii(LocatorMapper().get_formatted_headline(article, category.decode('UTF-8').upper()))
+        headline = to_ascii(
+            clean_string(LocatorMapper().get_formatted_headline(article, category.decode('UTF-8').upper())))
 
         # Set the maximum size to 64 including the sequence number if any
         if len(headline) > 64:
diff --git a/server/aap/publish/formatters/anpa_formatter_test.py b/server/aap/publish/formatters/anpa_formatter_test.py
index a9224fca5..e86ba856b 100644
--- a/server/aap/publish/formatters/anpa_formatter_test.py
+++ b/server/aap/publish/formatters/anpa_formatter_test.py
@@ -388,6 +388,24 @@ def test_preformated(self):
         self.assertTrue(lines.getvalue().split('\r')[3].lstrip(), 'Test line 1')
         self.assertTrue(lines.getvalue().split('\r')[4], 'Test line 2')
 
+    def test_fancy_quotes(self):
+        f = AAPAnpaFormatter()
+        subscriber = self.app.data.find('subscribers', None, None)[0][0]
+        item = self.article.copy()
+        item.update({
+            'body_html': "<p>\"quoted”</p>"
+                         "<p>“In“ IBAC’s</p>"
+                         "<p>Short hyphen­not handled fix one day!</p>"
+                         "<p>“Then ‘You can’t have it’,\" </p>",
+            'format': 'html'})
+        resp = f.format(item, subscriber)[0]
+        out = resp['encoded_item']
+
+        lines = io.StringIO(out.decode())
+        self.assertTrue(lines.getvalue().split('\r')[3].lstrip(), '"quoted"')
+        self.assertTrue(lines.getvalue().split('\r')[4].lstrip(), '"In" IBAC\'s')
+        self.assertTrue(lines.getvalue().split('\r')[6].lstrip(), '"Then \'You can\'t have it\'," ')
+
     def test_embed_in_body_body(self):
         f = AAPAnpaFormatter()
         subscriber = self.app.data.find('subscribers', None, None)[0][0]