openstax · edwoodward · Mar 9, 2023 · Feb 22, 2023 · Feb 22, 2023 · Feb 24, 2023
diff --git a/openstax/middleware.py b/openstax/middleware.py
@@ -2,6 +2,15 @@
 from django.core.handlers.base import BaseHandler
 from django.middleware.common import CommonMiddleware
 from django.conf import settings
+from ua_parser import user_agent_parser
+from wagtail.models import Page
+from django.shortcuts import get_object_or_404
+from django.template.response import TemplateResponse
+from django.http import HttpResponse
+from books.models import Book
+from openstax.functions import build_image_url
+from news.models import NewsArticle
+from pages.models import HomePage, Supporters
 
 
 class HttpSmartRedirectResponse(HttpResponsePermanentRedirect):
@@ -39,3 +48,95 @@ def process_response(self, request, response):
             response = self.handler.get_response(request)
 
         return response
+
+
+class CommonMiddlewareOpenGraphRedirect(CommonMiddleware):
+    OG_USER_AGENTS = [
+        'twitterbot',
+        'facebookbot',
+        'pinterestbot',
+        'slackbot-linkexpanding',
+    ]
+
+    def __init__(self, get_response):
+        self.get_response = get_response
+
+    def __call__(self, request, *args, **kwargs):
+        if 'HTTP_USER_AGENT' in request.META:
+
+            user_agent = user_agent_parser.ParseUserAgent(request.META["HTTP_USER_AGENT"])
+            if user_agent['family'].lower() in self.OG_USER_AGENTS:
+                # url path minus the trailing /
+                url_path = request.get_full_path()[:-1]
+
+                full_url = request.build_absolute_uri()
+
+                # imdex of last / to find slug, except when there isn't a last /
+                if url_path == '':
+                    page_slug = "openstax-homepage"
+                else:
+                    index = url_path.rindex('/')
+                    page_slug = url_path[index+1:]
+
+                if self.redirect_path_found(url_path):
+                    # supporters page has the wrong slug
+                    if page_slug == 'foundation':
+                        page_slug = 'supporters'
+
+                    # look up correct object based on path
+                    if '/details/books/' in url_path:
+                        page = Book.objects.filter(slug = page_slug)
+                    elif '/blog/' in url_path:
+                        page = NewsArticle.objects.filter(slug = page_slug)
+                    else:
+                        page = self.page_by_slug(page_slug)
+
+                    template = self.build_template(page[0], full_url)
+                    return HttpResponse(template)
+                else:
+                    return self.get_response(request)
+        return self.get_response(request)
+
+    def build_template(self, page, page_url):
+        image_url = self.image_url(page.promote_image)
+        template = '<!DOCTYPE html> <html> <head> <meta charset="utf-8">\n'
+        template += '    <title>' + str(page.seo_title) + '</title>\n'
+        template += '    <meta name="description" content="{}" >\n'.format(page.search_description)
+        template += '    <link rel="canonical" href="{}" />\n'.format(page_url)
+        template += '    <meta property="og:url" content="{}" />\n'.format(page_url)
+        template += '    <meta property="og:type" content="article" />\n'
+        template += '    <meta property="og:title" content="{}" />\n'.format(page.seo_title)
+        template += '    <meta property="og:description" content="{}" />\n'.format(page.search_description)
+        template += '    <meta property="og:image" content="{}" />\n'.format(image_url)
+        template += '    <meta property="og:image:alt" content="{}" />\n'.format(page.seo_title)
+        template += '    <meta name="twitter:card" content="summary_large_image">\n'
+        template += '    <meta name="twitter:site" content="@OpenStax">\n'
+        template += '    <meta name="twitter:title" content="{}">\n'.format(page.seo_title)
+        template += '    <meta name="twitter:description" content="{}">\n'.format(page.search_description)
+        template += '    <meta name="twitter:image" content="{}">\n'.format(image_url)
+        template += '    <meta name="twitter:image:alt" content="OpenStax">\n'
+
+        template += '</head><body></body></html>'
+        return template
+
+    def redirect_path_found(self, url_path):
+        if '/blog/' in url_path or '/details/books/' in url_path or '/foundation' in url_path or '' == url_path:
+            return True
+        else:
+            return False
+
+    def image_url(self, image):
+        image_url = build_image_url(image)
+        if not image_url:
+            return ''
+        return image_url
+
+    def page_by_slug(self, page_slug):
+        if page_slug == 'supporters':
+            return Supporters.objects.all()
+        if page_slug == 'openstax-homepage':
+            return HomePage.objects.filter(locale = 1)
+
+
+
+
diff --git a/openstax/settings/base.py b/openstax/settings/base.py
@@ -140,6 +140,7 @@
     'whitenoise.middleware.WhiteNoiseMiddleware',
     'healthcheck.middleware.HealthCheckMiddleware', # has to be before CommonMiddleware
     'openstax.middleware.CommonMiddlewareAppendSlashWithoutRedirect',
+    'openstax.middleware.CommonMiddlewareOpenGraphRedirect',
     'django.middleware.common.CommonMiddleware',
     'django.contrib.sessions.middleware.SessionMiddleware',
     'django.middleware.csrf.CsrfViewMiddleware',

diff --git a/openstax/tests.py b/openstax/tests.py
@@ -1,6 +1,16 @@
+import datetime
+import json
 from .functions import remove_locked_links_detail, remove_locked_links_listing, build_document_url, build_image_url
 
 from django.test import TestCase, Client
+from django.core.files.uploadedfile import SimpleUploadedFile
+from wagtail.models import Page
+from pages.models import HomePage, WebinarPage
+from books.models import BookIndex, Book
+from news.models import NewsIndex, NewsArticle
+from snippets.models import Subject, BlogContentType, BlogCollection
+from wagtail.documents.models import Document
+
 
 class TestClass(object):
     pass
@@ -79,3 +89,102 @@ def test_build_document_url_none(self):
 
     def test_build_image_url_none(self):
         self.assertEqual(build_image_url(None), None)
+
+
+class TestOpenGraphMiddleware(TestCase):
+    def setUp(self):
+        self.client = Client(HTTP_USER_AGENT='twitterbot')
+        self.root_page = Page.objects.get(title="Root")
+        self.homepage = HomePage(title="Hello World",
+                            slug="openstax-homepage",
+                            seo_title='OpenStax Home',
+                            search_description='Home page for OpenStax'
+                            )
+        self.root_page.add_child(instance=self.homepage)
+
+    def test_home_page_link_preview(self):
+        response = self.client.get('/')
+        self.assertContains(response, 'og:image')
+
+
+    def test_book_link_preview(self):
+        test_image = SimpleUploadedFile(name='openstax.png',
+                                        content=open("oxauth/static/images/openstax.png", 'rb').read())
+        self.test_doc = Document.objects.create(title='Test Doc', file=test_image)
+        book_index = BookIndex(title="Book Index",
+                               page_description="Test",
+                               dev_standard_1_description="Test",
+                               dev_standard_2_description="Test",
+                               dev_standard_3_description="Test",
+                               dev_standard_4_description="Test",
+                               )
+        # add book index to homepage
+        self.homepage.add_child(instance=book_index)
+        book = Book(title="Biology 2e",
+                    slug="biology-2e",
+                    cnx_id='031da8d3-b525-429c-80cf-6c8ed997733a',
+                    description="Test Book",
+                    cover=self.test_doc,
+                    title_image=self.test_doc,
+                    publish_date=datetime.date.today(),
+                    locale=self.root_page.locale,
+                    license_name='Creative Commons Attribution License',
+                    seo_title='Biology 2e',
+                    search_description='2nd edition of Biology'
+                    )
+        book_index.add_child(instance=book)
+        self.client = Client(HTTP_USER_AGENT='Slackbot-LinkExpanding 1.0 (+https://api.slack.com/robots)')
+        response = self.client.get('/details/books/biology-2e')
+        self.assertContains(response, 'og:image')
+
+    def test_blog_link_preview(self):
+        self.news_index = NewsIndex(title="News Index")
+        self.homepage.add_child(instance=self.news_index)
+        self.math = Subject(name="Math", page_content="Math page content.", seo_title="Math SEO Title",
+                            search_description="Math page description.")
+        self.math.save()
+        math_id = self.math.id
+        self.case_study = BlogContentType(content_type='Case Study')
+        self.case_study.save()
+        case_study_id = self.case_study.id
+        self.learning = BlogCollection(name='Teaching and Learning', description='this is a collection')
+        self.learning.save()
+        learning_id = self.learning.id
+        self.article = NewsArticle(title="Article 1",
+                                   slug="article-1",
+                                   date=datetime.date.today(),
+                                   heading="Sample Article",
+                                   subheading="Sample Subheading",
+                                   author="OpenStax",
+                                   seo_title='Test Article 1',
+                                   search_description='Test Article 1 description',
+                                   body=json.dumps(
+                                       [
+                                           {"type": "paragraph",
+                                            "value": "<p>This is the body of the post</p><p>This is the second paragraph</p>"}
+                                       ]
+                                   ),
+                                   article_subjects=json.dumps(
+                                       [
+                                           {'type': 'subject', 'value': [
+                                               {'type': 'item', 'value': {'subject': math_id, 'featured': False}}]}
+                                       ]
+                                   ),
+                                   content_types=json.dumps(
+                                       [
+                                           {'type': 'content_type', 'value': [
+                                               {'type': 'item', 'value': {'content_type': case_study_id}}]}
+                                       ]
+                                   ),
+                                   collections=json.dumps(
+                                       [
+                                           {'type': 'collection', 'value': [
+                                               {'type': 'item', 'value': {'collection': learning_id, 'featured': False,
+                                                                          'popular': False}}]}
+                                       ]
+                                   ))
+        self.news_index.add_child(instance=self.article)
+        self.client = Client(HTTP_USER_AGENT='facebookexternalhit/1.1')
+        response = self.client.get('/blog/article-1')
+        self.assertContains(response, 'og:image')
+
diff --git a/requirements/base.txt b/requirements/base.txt
@@ -23,6 +23,7 @@ sentry-sdk
 setuptools==65.5.1
 simple_salesforce==1.11.6
 social-auth-app-django==5.0.0
+ua_parser==0.16.1
 unicodecsv==0.14.1
 Unidecode==1.3.4
 vcrpy==4.1.1 # for recording test interactions with third-party APIs