Browse Source

fix #3, fix #1, add exception handling for weird edgecases

master
rra 3 years ago
parent
commit
162ecce865
  1. 111
      rss_aggregator.py

111
rss_aggregator.py

@ -23,10 +23,10 @@ def write_etag(feed_name, feed_data):
etag = ''
modified = ''
if 'etag' in data:
etag = data.etag
if 'modified' in data:
modified = data.modified
if 'etag' in feed_data:
etag = feed_data.etag
if 'modified' in feed_data:
modified = feed_data.modified
if etag or modified:
with open(os.path.join('etags',feed_name +'.txt'),'w') as f:
@ -56,11 +56,16 @@ def create_frontmatter(entry):
published = arrow.get(published)
if 'author' in entry:
author = entry.author
else:
author = ''
frontmatter = {
'title':entry.title,
'date': published.format(),
'summary': '',
'author': entry.author,
'author': author,
'original_link': entry.link
}
@ -75,7 +80,10 @@ def create_post(post_dir, entry):
if not os.path.exists(post_dir):
os.makedirs(post_dir)
if 'content' in entry:
post_content = entry.content[0].value
else:
post_content = entry.summary
parsed_content = parse_posts(post_dir, post_content)
@ -92,14 +100,10 @@ def grab_media(post_directory, url):
image = urlparse(url).path.split('/')[-1]
try:
if not os.path.exists(os.path.join(post_directory, image)):
#TODO: stream is true is a conditional so we could check the headers for things, mimetype etc
response = requests.get(url, stream=True)
except Exception as e:
print(e)
return url
try:
if not os.path.exists(os.path.join(post_directory, image)):
if response.ok:
with open(os.path.join(post_directory, image), 'wb') as img_file:
shutil.copyfileobj(response.raw, img_file)
print('Downloaded cover image', image)
@ -120,14 +124,41 @@ def parse_posts(post_direntry, post_content):
soup = BeautifulSoup(post_content, "html.parser")
media = []
for img in soup(['img','object']):
local_image = grab_media(post_dir, img['src'])
if img['src'] != local_image:
print(img['src'], '->', local_image)
img['src'] = local_image
return soup.decode()
def grab_feed(feed_url):
"""
check whether feed has been updated
download & return it if it has
"""
feed_name = urlparse(feed_url).netloc
etag, modified = get_etag(feed_name)
try:
if modified:
data = feedparser.parse(feed_url, modified=modified)
elif etag:
data = feedparser.parse(feed_url, etag=etag)
else:
data = feedparser.parse(feed_url)
except Exception as e:
print('Error grabbing feed')
print(feed_name)
print(e)
return False
print(data.status, feed_url)
if data.status == 200:
#304 means the feed has not been modified since we last checked
write_etag(feed_name, data)
return data
return False
feed_urls = open('feeds_list.txt','r').read().splitlines()
@ -141,8 +172,8 @@ env = jinja2.Environment(
loader=jinja2.FileSystemLoader(os.path.curdir)
)
output_dir = os.environ.get('OUTPUT_DIR', '/home/r/Programming/lumbung.space/lumbung.space-web/content/posts/')
#output_dir = os.environ.get('OUTPUT_DIR', 'network/')
#output_dir = os.environ.get('OUTPUT_DIR', '/home/r/Programming/lumbung.space/lumbung.space-web/content/posts/')
output_dir = os.environ.get('OUTPUT_DIR', 'network/')
if not os.path.exists(output_dir):
os.makedirs(output_dir)
@ -150,42 +181,40 @@ if not os.path.exists(output_dir):
template = env.get_template('post_template.md')
for feed_url in feed_urls[7:]:
for feed_url in feed_urls:
feed_name = urlparse(feed_url).netloc
etag, modified = get_etag(feed_name)
if modified:
data = feedparser.parse(feed_url, modified=modified)
elif etag:
data = feedparser.parse(feed_url, etag=etag)
else:
data = feedparser.parse(feed_url)
feed_dir = os.path.join(output_dir, feed_name)
print(data.status, feed_url)
if not os.path.exists(feed_dir):
os.makedirs(feed_dir)
if data.status == 200:
existing_posts = os.listdir(feed_dir)
#write_etag(feed_url, data)
data = grab_feed(feed_url)
# if 'title' in data.feed:
# print('#'*10)
# print(data.feed.title)
# print('#'*10)
# print('\n')
if data:
for entry in data.entries:
# if 'tags' in entry:
# print(entry.title, entry.tags)
# print('FEED KEYS')
# print(data.keys())
# print('\n')
post_name = slugify(entry.title)
post_dir = os.path.join(output_dir, feed_name, post_name)
for entry in data.entries:
# print(entry.title)
# print(entry.keys())
# print('\n')
# # if 'tags' in entry:
# # print(entry.title, entry.tags)
if post_name not in existing_posts:
#if there is a blog entry we dont already have, make it
create_post(post_dir, entry)
post_dir = os.path.join(output_dir, feed_name, slugify(entry.title))
elif post_name in existing_posts:
#if we already have it, update it
create_post(post_dir, entry)
existing_posts.remove(post_name) # create list of posts which have not been returned by the feed
for post in existing_posts:
#remove blog posts no longer returned by the RSS feed
print('deleted', post)
shutil.rmtree(os.path.join(feed_dir, slugify(post)))

Loading…
Cancel
Save