Add PEP abstract to the RSS feed (#1679)

* Flake8 fixes

* Use first paragraph of abstract as 'description', and PEP author as 'author'

* Check RSS generation runs with no error

.github/workflows/build.yml

@@ -20,7 +20,9 @@ jobs:
           python -m pip install -U docutils
 
       - name: Build
-        run: make -j$(nproc)
+        run: |
+          make rss
+          make -j$(nproc)
 
       - name: Deploy
         if: >

pep2rss.py

@@ -1,23 +1,71 @@
 #!/usr/bin/env python3
 
-# usage: pep-hook.py $REPOS $REV
+# usage: python3 pep2rss.py .
-# (standard post-commit args)
 
-import os, glob, time, datetime, stat, re, sys
+import datetime
+import glob
+import os
+import re
+import sys
+import time
 import PyRSS2Gen as rssgen
+import docutils.frontend
+import docutils.nodes
+import docutils.parsers.rst
+import docutils.utils
 
 RSS_PATH = os.path.join(sys.argv[1], 'peps.rss')
 
+
+def remove_prefix(text: str, prefix: str) -> str:
+    try:
+        # Python 3.9+
+        return text.removeprefix(prefix)
+    except AttributeError:
+        if text.startswith(prefix):
+            return text[len(prefix):]
+        return text
+
+
+def parse_rst(text: str) -> docutils.nodes.document:
+    parser = docutils.parsers.rst.Parser()
+    components = (docutils.parsers.rst.Parser,)
+    settings = docutils.frontend.OptionParser(components=components).get_default_values()
+    document = docutils.utils.new_document('<rst-doc>', settings=settings)
+    parser.parse(text, document)
+    return document
+
+
+def pep_abstract(full_path: str) -> str:
+    """Return the first paragraph of the PEP abstract"""
+    abstract = None
+    with open(full_path, encoding="utf-8") as f:
+        text = f.read()
+        document = parse_rst(text)
+        nodes = list(document)
+        for node in nodes:
+            if "<title>Abstract</title>" in str(node):
+                for child in node:
+                    if child.tagname == "paragraph":
+                        abstract = child.astext()
+                        # Just fetch the first paragraph
+                        break
+    return abstract
+
+
 def firstline_startingwith(full_path, text):
     for line in open(full_path, encoding="utf-8"):
         if line.startswith(text):
             return line[len(text):].strip()
     return None
 
+
 # get list of peps with creation time
 # (from "Created:" string in pep .rst or .txt)
 peps = glob.glob('pep-*.txt')
 peps.extend(glob.glob('pep-*.rst'))
+
+
 def pep_creation_dt(full_path):
     created_str = firstline_startingwith(full_path, 'Created:')
     # bleh, I was hoping to avoid re but some PEPs editorialize
@@ -35,6 +83,8 @@ def pep_creation_dt(full_path):
     except ValueError:
         t = time.strptime(created_str, '%d-%B-%Y')
     return datetime.datetime(*t[:6])
+
+
 peps_with_dt = [(pep_creation_dt(full_path), full_path) for full_path in peps]
 # sort peps by date, newest first
 peps_with_dt.sort(reverse=True)
@@ -48,13 +98,15 @@ for dt, full_path in peps_with_dt[:10]:
         pass
     title = firstline_startingwith(full_path, 'Title:')
     author = firstline_startingwith(full_path, 'Author:')
+    abstract = pep_abstract(full_path)
     url = 'https://www.python.org/dev/peps/pep-%0.4d/' % n
     item = rssgen.RSSItem(
-        title = 'PEP %d: %s' % (n, title),
+        title='PEP %d: %s' % (n, title),
-        link = url,
+        link=url,
-        description = 'Author: %s' % author,
+        description=abstract,
-        guid = rssgen.Guid(url),
+        author=author,
-        pubDate = dt)
+        guid=rssgen.Guid(url),
+        pubDate=dt)
     items.append(item)
 
 # the rss envelope
@@ -64,11 +116,11 @@ language features, and some meta-information like release
 procedure and schedules
 """.strip()
 rss = rssgen.RSS2(
-    title = 'Newest Python PEPs',
+    title='Newest Python PEPs',
     link = 'https://www.python.org/dev/peps/',
-    description = desc,
+    description=desc,
-    lastBuildDate = datetime.datetime.now(),
+    lastBuildDate=datetime.datetime.now(),
-    items = items)
+    items=items)
 
 with open(RSS_PATH, 'w', encoding="utf-8") as fp:
     fp.write(rss.to_xml(encoding="utf-8"))