Fixing relative image urls to be absolute urls. Looks like a BeautifulSoup4 upgrade didn't catch the new attrs syntax. Also fixing bookmarklet loading wrong JS/CSS in development.

.vscode/settings.json

@@ -8,7 +8,7 @@
     ],
     "python.pythonPath": "~/.virtualenvs/newsblur3/bin/python",
     "editor.bracketPairColorization.enabled": true,
-    "editor.guides.bracketPairs":"active",
+    "editor.guides.bracketPairs": "active",
     "git.ignoreLimitWarning": true,
     "search.exclude": {
         "clients": true,
@@ -27,7 +27,6 @@
         "docker/volumes": true,
         "requirements.txt": true, // It's just a symlink to config/requirements.txt, which has git history
     },
-    "python.formatting.provider": "black",
     "python.formatting.blackArgs": [
         "--line-length=110",
         "--skip-string-normalization"
@@ -37,4 +36,5 @@
     },
     "nrf-connect.toolchain.path": "${nrf-connect.toolchain:1.9.1}",
     "C_Cpp.default.configurationProvider": "nrf-connect",
+    "editor.formatOnSave": false,
 }

apps/api/views.py

@@ -206,6 +206,11 @@ def check_share_on_site(request, token):
     usersub    = None
     message    = None
     user       = None
+    users = {}
+    your_story = None
+    same_stories = None
+    other_stories = None
+    previous_stories = None
     
     if not story_url:
         code = -1
@@ -239,34 +244,34 @@ def check_share_on_site(request, token):
             usersub = UserSubscription.objects.filter(user=user, feed=feed)
         except UserSubscription.DoesNotExist:
             usersub = None
-    feed_id = feed and feed.pk
+    if user:
-    your_story, same_stories, other_stories = MSharedStory.get_shared_stories_from_site(feed_id,
+        feed_id = feed and feed.pk
-                                              user_id=user_profile.user.pk, story_url=story_url)
+        your_story, same_stories, other_stories = MSharedStory.get_shared_stories_from_site(feed_id,
-    previous_stories = MSharedStory.objects.filter(user_id=user_profile.user.pk).order_by('-shared_date').limit(3)
+                                                user_id=user.pk, story_url=story_url)
-    previous_stories = [{
+        previous_stories = MSharedStory.objects.filter(user_id=user.pk).order_by('-shared_date').limit(3)
-        "user_id": story.user_id,
+        previous_stories = [{
-        "story_title": story.story_title,
+            "user_id": story.user_id,
-        "comments": story.comments,
+            "story_title": story.story_title,
-        "shared_date": story.shared_date,
+            "comments": story.comments,
-        "relative_date": relative_timesince(story.shared_date),
+            "shared_date": story.shared_date,
-        "blurblog_permalink": story.blurblog_permalink(),
+            "relative_date": relative_timesince(story.shared_date),
-    } for story in previous_stories]
+            "blurblog_permalink": story.blurblog_permalink(),
+        } for story in previous_stories]
     
-    user_ids = set([user_profile.user.pk])
+        user_ids = set([user_profile.user.pk])
-    for story in same_stories:
+        for story in same_stories:
-        user_ids.add(story['user_id'])
+            user_ids.add(story['user_id'])
-    for story in other_stories:
+        for story in other_stories:
-        user_ids.add(story['user_id'])
+            user_ids.add(story['user_id'])
 
-    users = {}
+        profiles = MSocialProfile.profiles(user_ids)
-    profiles = MSocialProfile.profiles(user_ids)
+        for profile in profiles:
-    for profile in profiles:
+            users[profile.user_id] = {
-        users[profile.user_id] = {
+                "username": profile.username,
-            "username": profile.username,
+                "photo_url": profile.photo_url,
-            "photo_url": profile.photo_url,
+            }
-        }
             
-    logging.user(user_profile.user, "~BM~FCChecking share from site: ~SB%s" % (story_url),
+    logging.user(user, "~BM~FCChecking share from site: ~SB%s" % (story_url),
                  request=request)
     
     response = HttpResponse(callback + '(' + json.encode({
@@ -285,12 +290,12 @@ def check_share_on_site(request, token):
     
     return response
 
-@required_params('story_url', 'comments', 'title')
+@required_params('story_url')
 def share_story(request, token=None):
     code      = 0
     story_url = request.POST['story_url']
-    comments  = request.POST['comments']
+    comments  = request.POST.get('comments', "")
-    title     = request.POST['title']
+    title     = request.POST.get('title', None)
     content   = request.POST.get('content', None)
     rss_url   = request.POST.get('rss_url', None)
     feed_id   = request.POST.get('feed_id', None) or 0
@@ -333,10 +338,12 @@ def share_story(request, token=None):
         content = lxml.html.fromstring(content)
         content.make_links_absolute(story_url)
         content = lxml.html.tostring(content)
-    else:
+
+    if not content or not title:
         importer = TextImporter(story=None, story_url=story_url, request=request, debug=settings.DEBUG)
         document = importer.fetch(skip_save=True, return_document=True)
-        content = document['content']
+        if not content:
+            content = document['content']
         if not title:
             title = document['title']
     

apps/rss_feeds/text_importer.py

@@ -225,9 +225,9 @@ class TextImporter:
         
         content = str(soup)
         
-        images = set([img['src'] for img in soup.findAll('img') if 'src' in img])
+        images = set([img.attrs['src'] for img in soup.findAll('img') if 'src' in img.attrs])
         for image_url in images:
-            abs_image_url = urljoin(self.story.story_permalink, image_url)
+            abs_image_url = urljoin(self.story_url, image_url)
             content = content.replace(image_url, abs_image_url)
         
         return content
@@ -242,9 +242,11 @@ class TextImporter:
             headers["content-type"] = "application/json"
             headers["x-api-key"] = mercury_api_key
             domain = Site.objects.get_current().domain
+            protocol = "https"
             if settings.DOCKERBUILD:
                 domain = 'haproxy'
-            url = f"https://{domain}/rss_feeds/original_text_fetcher?url={url}"
+                protocol = "http"
+            url = f"{protocol}://{domain}/rss_feeds/original_text_fetcher?url={url}"
             
         try:
             r = requests.get(url, headers=headers, timeout=15)

media/css/bookmarklet/bookmarklet.css

@@ -320,16 +320,12 @@
     margin-left: 0;
     width: 220px;
 }
-.NB-bookmarklet .NB-bookmarklet-comment-submit.NB-disabled,
-.NB-bookmarklet .NB-bookmarklet-save.NB-disabled {
-    padding: 14px 0 0;
-}
 .NB-bookmarklet .NB-bookmarklet-comment-submit .NB-bookmarklet-accept,
-.NB-bookmarklet .NB-bookmarklet-save .NB-bookmarklet-accept {
+.NB-bookmarklet .NB-bookmarklet-save-button .NB-bookmarklet-accept {
     padding: 0 12px;
 }
 .NB-bookmarklet .NB-bookmarklet-comment-submit .NB-bookmarklet-accept img,
-.NB-bookmarklet .NB-bookmarklet-save .NB-bookmarklet-accept img {
+.NB-bookmarklet .NB-bookmarklet-save-button .NB-bookmarklet-accept img {
     vertical-align: top;
     margin: 0;
 }

newsblur_web/assets.yml

@@ -106,7 +106,6 @@ javascripts:
         - js/newsblur/payments/stripe_form.js
     bookmarklet:
         - js/vendor/jquery-3.6.0.min.js
-        # - js/vendor/jquery.migrate-*.js
         - js/vendor/jquery.noConflict.js
         - js/vendor/jquery.browser.js
         - js/vendor/jquery.newsblur.js
@@ -121,7 +120,6 @@ javascripts:
         - js/vendor/readability-*.js
     blurblog:
         - js/vendor/jquery-3.6.0.js
-        # - js/vendor/jquery.migrate-*.js
         - js/vendor/jquery.browser.js
         - js/vendor/jquery.color.js
         - js/vendor/underscore-*.js

node/package.json

@@ -4,7 +4,7 @@
   "description": "Servers used in running NewsBlur",
   "main": "favicons.js",
   "dependencies": {
-    "@postlight/mercury-parser": "^2.2.1",
+    "@postlight/mercury-parser": "^2.2.3",
     "@sentry/browser": "^6.12.0",
     "@sentry/node": "^6.12.0",
     "@sentry/tracing": "^6.12.0",

templates/api/share_bookmarklet.js

@@ -588,10 +588,11 @@
                 this.story_content = selected;
                 console.log(["content selected", this.story_title, this.story_content]);
             } else {
-                var $readability = $(window.readability.init());
+                var documentClone = document.cloneNode(true);
+                var article = new Readability(documentClone).parse();
                 
-                this.story_title = $readability.children("h1").text();
+                this.story_title = article.title;
-                this.story_content = $("#readability-content", $readability).html();
+                this.story_content = article.content;
             }
             
             this.find_video_embeds();

utils/templatetags/utils_tags.py

@@ -3,6 +3,7 @@ import re
 import struct
 import datetime
 import random
+from utils import log as logging
 from django.contrib.sites.models import Site
 from django.conf import settings
 from django import template
@@ -299,12 +300,13 @@ def include_javascripts(parser, token):
 class RawJSNode(JavascriptNode):
     def render(self, context):
         output = super(RawJSNode, self).render(context)
-        path = re.search(r"src=\"/(.*?)\"", output)
+        paths = re.findall(r"src=\"/(.*?)\"", output, re.MULTILINE)
-        assert path
+        assert paths
-        filename = path.group(1)
+        output = ""
-        abs_filename = os.path.join(settings.NEWSBLUR_DIR, filename)
+        for filename in paths:
-        f = open(abs_filename, 'r')
+            abs_filename = os.path.join(settings.NEWSBLUR_DIR, filename)
-        output = f.read()
+            f = open(abs_filename, 'r')
+            output += f.read()
         return output
     
 @register.tag
@@ -317,12 +319,13 @@ def include_javascripts_raw(parser, token):
 class RawStylesheetNode(StylesheetNode):
     def render(self, context):
         output = super(RawStylesheetNode, self).render(context)
-        path = re.search(r"href=\"/(.*?)\"", output)
+        paths = re.findall(r"href=\"/(.*?)\"", output, re.MULTILINE)
-        assert path
+        assert paths
-        filename = path.group(1)
+        output = ""
-        abs_filename = os.path.join(settings.NEWSBLUR_DIR, filename)
+        for filename in paths:
-        f = open(abs_filename, 'r')
+            abs_filename = os.path.join(settings.NEWSBLUR_DIR, filename)
-        output = f.read().replace('"', '\\"').replace('\n', '')
+            f = open(abs_filename, 'r')
+            output += f.read().replace('"', '\\"').replace('\n', '')
         return output
     
 @register.tag