Remove feedparser

.gitlab-ci.yml

@@ -3,7 +3,7 @@ image: python:3.9-alpine
 pages:
     stage: deploy
     script:
-        - pip install -r requirements.txt && cd public/sbbs && python3 ../../gensbbsrss.py
+        - cd public/sbbs && python3 ../../gensbbsrss.py
     artifacts:
         paths:
             - public

genbloglist.py

@@ -1,36 +1,85 @@
 import sys
 import html
 import shlex
-from urllib.parse import urlparse, urlunparse
-import feedparser
+from xml.etree import ElementTree
+from urllib.request import urlopen, Request
+from urllib.parse import urlparse, urlunparse, urljoin
+
+def find_ignore_ns(element, tag):
+    for i in element.iter():
+        if i.tag == tag:
+            return i
+        if i.tag in ('item', 'entry') or i.tag.endswith('}item') or i.tag.endswith('}entry'):
+            break
+    for i in element.iter():
+        if i.tag.endswith('}' + tag):
+            return i
+        if i.tag in ('item', 'entry') or i.tag.endswith('}item') or i.tag.endswith('}entry'):
+            break
+
+is_link = lambda i: i.attrib.get('rel') != 'self' and i.attrib.get('type', 'text/html') == 'text/html'
+
+def find_link(element):
+    for i in element.iter():
+        if i.tag == 'link' and is_link(i):
+            return i
+        if i.tag in ('item', 'entry') or i.tag.endswith('}item') or i.tag.endswith('}entry'):
+            break
+    for i in element.iter():
+        if i.tag.endswith('}link') and is_link(i):
+            return i
+        if i.tag in ('item', 'entry') or i.tag.endswith('}item') or i.tag.endswith('}entry'):
+            break
 
 with open(sys.argv[1]) as file, open(sys.argv[2], 'w+') as out:
     for i in file:
         inp = shlex.split(i, comments=True)
         if 'blog' in inp[1:]:
+            feedurl = inp[0]
             try:
-                d = feedparser.parse(inp[0])
-            except Exception as e:
-                print(inp[0], 'raised', e, file=sys.stderr)
+                with urlopen(Request(feedurl, headers={'User-Agent': "stop being so fucking obsessed that i'm using urllib ffs"}), timeout=60) as resp:
+                    if resp.status != 200:
+                        print(feedurl, 'returned', resp.status, file=sys.stderr)
                         continue
-            feedurl = d['href']
-            if not d['entries']:
+                    tree = ElementTree.ElementTree().parse(resp)
+            except Exception as e:
+                print(feedurl, 'raised', e, file=sys.stderr)
+                continue
+            if tree.tag not in ('rss', 'feed') and not tree.tag.endswith('}rss') and not tree.tag.endswith('}feed'):
+                print(f'{feedurl} is not a feed (root tag is {tree.tag})', file=sys.stderr)
+                continue
+            channel = find_ignore_ns(tree, 'channel')
+            if channel is None:
+                channel = tree
+            if find_ignore_ns(channel, 'item') is None and find_ignore_ns(channel, 'entry') is None:
                 print(feedurl, 'has no entries, skipping', file=sys.stderr)
                 continue
-            if 'links' in d['feed']:
-                url = next(filter(lambda i: i['type'] == 'text/html', d['feed']['links']))['href']
-            else:
+            url = find_link(channel)
+            if url is None:
                 url = list(urlparse(feedurl))
                 url[2] = ''
                 url = urlunparse(url)
                 print(f'No mention of main page on {feedurl}, please see {url} or enter main page url: ', file=sys.stderr, end='', flush=True)
                 url = input().strip() or url
-            desc = d['feed'].get('description')
-            text = f'<li><b><a href="{html.escape(url)}">{html.escape(d["feed"]["title"])}</a> (<a href="{html.escape(feedurl)}">feed</a>)'
-            if desc := d['feed'].get('description'):
+            else:
+                url = url.text or url.attrib['href']
+            text = f'<li><b><a href="{html.escape(urljoin(feedurl, url))}">{html.escape(find_ignore_ns(channel, "title").text)}</a> (<a href="{html.escape(feedurl)}">feed</a>)'
+            desc = find_ignore_ns(channel, 'description')
+            if desc is not None and desc.text:
+                if desc := desc.text.strip():
                     text += f':</b> {html.escape(desc)}'
                 else:
                     text += '</b>'
-            text += '</li>\n'
-            print(text, end='')
+            else:
+                desc = find_ignore_ns(channel, 'subtitle')
+                if desc is not None and desc.text:
+                    if desc := desc.text.strip():
+                        text += f':</b> {html.escape(desc)}'
+                    else:
+                        text += '</b>'
+                else:
+                    text += '</b>'
+            print(text, '</li>', sep='')
             out.write(text)
+            out.write('</li>\n')
+    out.flush()

requirements.txt

@@ -1 +0,0 @@
-feedparser