From ed4c290ba5bcdcaba4acbca3e2496ef9f2239054 Mon Sep 17 00:00:00 2001
From: Gabriel Augendre <gabriel@augendre.info>
Date: Tue, 28 Dec 2021 10:52:25 +0100
Subject: [PATCH] Fix empty description

This happened with articles starting with an image.
We now fetch the first paragraph containing some text.
---
 src/articles/models.py |  5 ++---
 src/articles/utils.py  | 14 ++++++++++++++
 2 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/src/articles/models.py b/src/articles/models.py
index c148275..a9dc88f 100644
--- a/src/articles/models.py
+++ b/src/articles/models.py
@@ -4,7 +4,6 @@ from functools import cached_property
 
 import rcssmin
 import readtime
-from bs4 import BeautifulSoup
 from django.conf import settings
 from django.contrib.auth.models import AbstractUser
 from django.contrib.contenttypes.models import ContentType
@@ -16,6 +15,7 @@ from django.utils import timezone
 
 from articles.utils import (
     build_full_absolute_url,
+    find_first_paragraph_with_text,
     format_article_content,
     truncate_words_after_char_count,
 )
@@ -92,8 +92,7 @@ class Article(models.Model):
     @cached_property
     def get_description(self):
         html = self.get_formatted_content
-        bs = BeautifulSoup(html, "html.parser")
-        text = bs.find("p", recursive=False).text
+        text = find_first_paragraph_with_text(html)
         return truncate_words_after_char_count(text, 160)
 
     @cached_property
diff --git a/src/articles/utils.py b/src/articles/utils.py
index 0c8e954..224b84c 100644
--- a/src/articles/utils.py
+++ b/src/articles/utils.py
@@ -1,6 +1,7 @@
 import re
 
 import markdown
+from bs4 import BeautifulSoup
 from django.conf import settings
 from markdown.extensions.codehilite import CodeHiliteExtension
 
@@ -36,3 +37,16 @@ def truncate_words_after_char_count(text, char_count):
         text_result.append(word)
         total_length += len(word) + 1
     return " ".join(text_result) + "..."
+
+
+def find_first_paragraph_with_text(html):
+    bs = BeautifulSoup(html, "html.parser")
+    paragraph = bs.find("p", recursive=False)
+    text = paragraph.text.strip()
+    while not text:
+        try:
+            paragraph = paragraph.next_sibling
+            text = paragraph.text.strip()
+        except Exception:
+            break
+    return text