From aae88fa07eb1b8336c3443607c37f28bd0ee3fe2 Mon Sep 17 00:00:00 2001
From: unkn0w7n <51942695+unkn0w7n@users.noreply.github.com>
Date: Tue, 27 Feb 2024 23:12:27 +0530
Subject: [PATCH] ...

---
 recipes/new_yorker.recipe | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/recipes/new_yorker.recipe b/recipes/new_yorker.recipe
index a4eff56522..bfe5c68d00 100644
--- a/recipes/new_yorker.recipe
+++ b/recipes/new_yorker.recipe
@@ -34,6 +34,7 @@ class NewYorker(BasicNewsRecipe):
     timefmt = ' [%b %d]'
     encoding = 'utf-8'
     extra_css = '''
+        img { display:block; margin:0 auto; }
         .byline { font-size:smaller; font-weight: bold;}
         h3 { margin-bottom: 6px; }
         .caption { font-size: smaller; font-style: italic; font-weight: normal; }
@@ -72,6 +73,21 @@ class NewYorker(BasicNewsRecipe):
             self.log.warn('Kindle Output profile being used, reducing image quality to keep file size below amazon email threshold')
 
     def preprocess_html(self, soup):
+        w = '/w_320' # use '/w_640' for highres
+        for img in soup.findAll('img'):
+            if img.has_attr('srcset'):
+                for x in img['srcset'].split():
+                    if w in x:
+                        img['src'] = x
+            elif img.find_previous_sibling('source', attrs={'srcset':True}):
+                srcset = img.find_previous_sibling('source', attrs={'srcset':True})
+                for x in srcset['srcset'].split():
+                    if w in x:
+                        img['src'] = x
+                    elif '/w_560' in x:
+                        img['src'] = x
+        for src in soup.findAll('source'):
+            src.decompose()
         for noscript in soup.findAll('noscript'):
             noscript.name = 'div'
         return soup