start on RSS output

for #124

README.md

@@ -303,6 +303,7 @@ On the open source side, there are many related projects. [php-mf2-shim](https:/
 Changelog
 ---
 ### 1.15 - unreleased
+* Add RSS 2.0 output! ([#124](https://github.com/snarfed/granary/issues/124))
 * All silos:
   * Switch users' primary URLs from web site to silo profile ([#158](https://github.com/snarfed/granary/issues/158)).
 * GitHub:

granary/rss.py

@@ -0,0 +1,121 @@
+"""Convert between ActivityStreams and RSS 2.0.
+
+RSS 2.0 spec: http://www.rssboard.org/rss-specification
+"""
+from __future__ import absolute_import, unicode_literals
+from builtins import str
+from past.builtins import basestring
+
+import mimetypes
+
+from feedgen.feed import FeedGenerator
+import mf2util
+from oauth_dropins.webutil import util
+
+from . import microformats2
+
+# allowed ActivityStreams objectTypes for media enclosures
+ENCLOSURE_TYPES = {'audio', 'video'}
+
+
+def from_activities(activities, actor=None, title=None, description=None,
+                    feed_url=None, home_page_url=None, image_url=None):
+  """Converts ActivityStreams activities to an RSS 2.0 feed.
+
+  Args:
+    activities: sequence of ActivityStreams activity dicts
+    actor: ActivityStreams actor dict, the author of the feed
+    title: string, the feed title
+    description, the feed description
+    home_page_url: string, the home page URL
+    # feed_url: the URL of this RSS feed, if any
+    image_url: the URL of an image representing this feed
+
+  Returns:
+    unicode string with RSS 2.0 XML
+  """
+  try:
+    iter(activities)
+  except TypeError:
+    raise TypeError('activities must be iterable')
+
+  if isinstance(activities, (dict, basestring)):
+    raise TypeError('activities may not be a dict or string')
+
+  fg = FeedGenerator()
+  fg.id(feed_url)
+  fg.link(href=feed_url, rel='self')
+  fg.link(href=home_page_url, rel='alternate')
+  fg.title(title)
+  fg.description(description)
+  fg.generator('granary', uri='https://granary.io/')
+  if image_url:
+    fg.image(image_url)
+
+  latest = None
+  for activity in activities:
+    obj = activity.get('object') or activity
+    if obj.get('objectType') == 'person':
+      continue
+
+    item = fg.add_entry()
+    url = obj.get('url')
+    item.id(obj.get('id') or url)
+    item.link(href=url)
+    item.guid(url, permalink=True)
+
+    item.title(obj.get('title') or obj.get('displayName'))
+    content = microformats2.render_content(
+      obj, include_location=True, render_attachments=False) or obj.get('summary')
+    if content:
+      item.content(content, type='CDATA')
+
+    item.category([{'term': t.displayName} for t in obj.get('tags', [])
+                  if t.displayName and t.verb not in ('like', 'react', 'share')])
+
+    author = obj.get('author', {})
+    item.author({
+      'name': author.get('displayName') or author.get('username'),
+      'uri': author.get('url'),
+    })
+
+    for prop in 'published', 'updated':
+      val = obj.get(prop)
+      if val:
+        dt = util.parse_iso8601(val)
+        getattr(item, prop)(dt)
+        if not latest or dt > latest:
+          latest = dt
+
+    enclosures = False
+    for att in obj.get('attachments', []):
+      stream = util.get_first(att, 'stream') or att
+      if not stream:
+        continue
+
+      url = stream.get('url')
+      mime = mimetypes.guess_type(url)[0] if url else None
+      if (att.get('objectType') in ENCLOSURE_TYPES or
+          mime and mime.split('/')[0] in ENCLOSURE_TYPES):
+        enclosures = True
+        item.enclosure(url=url, type=mime) # TODO: length (bytes)
+
+        item.load_extension('podcast')
+        duration = stream.get('duration')
+        if duration:
+          item.podcast.itunes_duration(duration)
+
+  if enclosures:
+    fg.load_extension('podcast')
+    if actor:
+      fg.podcast.itunes_author(actor.get('displayName') or actor.get('username'))
+    fg.podcast.itunes_image(image_url)
+    if description:
+      fg.podcast.itunes_subtitle(description)
+    fg.podcast.itunes_explicit('no')
+    fg.podcast.itunes_block(False)
+
+  if latest:
+    fg.lastBuildDate(dt)
+
+  return fg.rss_str(pretty=True)

granary/tests/test_testdata.py

@@ -13,7 +13,7 @@
 from oauth_dropins.webutil import testutil
 from oauth_dropins.webutil import util
 
-from granary import as2, jsonfeed, microformats2
+from granary import as2, jsonfeed, microformats2, rss
 
 
 def filepairs(ext1, ext2s):
@@ -80,6 +80,12 @@ def jsonfeed_to_activity(jf):
 def html_to_activity(html):
   return microformats2.html_to_activities(html)[0]['object']
 
+def rss_from_activities(activities):
+  return rss.from_activities(
+    activities, actor=ACTOR, title='Stuff', description='some stuff by meee',
+    feed_url='http://site/feed', home_page_url='http://site/',
+    image_url='http://site/logo.png').decode('utf-8')
+
 # source extension, destination extension, conversion function, exclude prefix
 mappings = (
   ('as.json', ['mf2-from-as.json', 'mf2.json'], microformats2.object_to_json, ()),
@@ -94,6 +100,7 @@ def html_to_activity(html):
   ('feed.json', ['as-from-feed.json', 'as.json'], jsonfeed_to_activity, ()),
   ('as.json', ['as2-from-as.json', 'as2.json'], as2.from_as1, ()),
   ('as2.json', ['as-from-as2.json', 'as.json'], as2.to_as1, ()),
+  ('as.json', ['rss.xml'], rss_from_activities, ()),
 )
 
 test_funcs = {}

granary/tests/testdata/feed_with_audio_video.as.json

@@ -0,0 +1,32 @@
+[{
+  "url": "http://podcast/post",
+  "objectType": "article",
+  "displayName": "i'm ready to speak",
+  "content": "<p>some HTML</p>",
+  "published": "2012-12-05T00:58:26+00:00",
+  "attachments": [{
+    "stream": {
+      "url": "http://a/podcast.mp3",
+      "duration": 328
+    },
+    "objectType": "audio"
+  }]
+},
+{
+  "url": "http://vidjo/post",
+  "objectType": "article",
+  "displayName": "i'm ready to perform",
+  "summary": "other thing",
+  "updated": "2012-12-06T00:58:26+00:00",
+  "attachments": [{
+    "stream": {
+      "url": "http://a/vidjo.mov",
+      "duration": 428
+    },
+    "objectType": "video"
+  }],
+  "stream": [{
+    "url": "http://a/vidjo.mov",
+    "duration": 428
+  }]
+}]

granary/tests/testdata/feed_with_audio_video.mf2.json

@@ -0,0 +1,52 @@
+{
+  "items": [{
+    "type": ["h-feed"],
+    "lang": "en",
+    "properties": {
+      "name": ["Pawd Kaast"],
+      "summary": ["a pawd kaast by meee"],
+      "photo": ["https://cover/"],
+      "author": [{
+        "type": ["h-card"],
+        "properties": {
+            "name": ["Meee"],
+            "photo": ["https://photo/of/meee"],
+            "url": ["https://meee.com"],
+            "bio": [{
+              "html": "my <a>bio</a>",
+              "value": "my bio"
+            }]
+        },
+        "value": "Meeeeee"
+      }]
+    },
+    "children": [{
+      "type": ["h-entry"],
+      "properties": {
+        "url": ["http://a/podcast"],
+        "name": ["i'm ready to speak"],
+        "audio": ["http://a/podcast.mp3"],
+        "duration": ["328"],
+        "size": ["7.77mb"],
+        "summary": [{
+          "html": "something",
+          "value": "something"
+        }]
+      }
+    },
+    {
+      "type": ["h-entry"],
+      "properties": {
+        "name": ["i'm ready to perform"],
+        "url": ["http://a/vidjo"],
+        "video": ["http://a/vidjo.mov"],
+        "duration": ["428"],
+        "size": ["8.88mb"],
+        "summary": [{
+          "html": "other thing",
+          "value": "other thing"
+        }]
+      }
+    }]
+  }]
+}

granary/tests/testdata/feed_with_audio_video.rss.xml

@@ -0,0 +1,42 @@
+<?xml version='1.0' encoding='UTF-8'?>
+<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" version="2.0">
+<channel>
+  <title>Stuff</title>
+  <link>http://site/</link>
+  <description>some stuff by meee</description>
+  <atom:link href="http://site/feed" rel="self"/>
+  <docs>http://www.rssboard.org/rss-specification</docs>
+  <generator>granary</generator>
+  <image>
+    <url>http://site/logo.png</url>
+    <title>Stuff</title>
+    <link>http://site/</link>
+  </image>
+  <lastBuildDate>Thu, 06 Dec 2012 00:58:26 +0000</lastBuildDate>
+  <itunes:author>Martin Smith</itunes:author>
+  <itunes:block>no</itunes:block>
+  <itunes:image href="http://site/logo.png"/>
+  <itunes:explicit>no</itunes:explicit>
+  <itunes:subtitle>some stuff by meee</itunes:subtitle>
+
+  <item>
+    <title>i'm ready to perform</title>
+    <link>http://vidjo/post</link>
+    <description>other thing</description>
+    <guid isPermaLink="true">http://vidjo/post</guid>
+    <enclosure url="http://a/vidjo.mov" length="0" type="video/quicktime"/>
+    <itunes:duration>428</itunes:duration>
+  </item>
+
+  <item>
+    <title>i'm ready to speak</title>
+    <link>http://podcast/post</link>
+    <description>&lt;p&gt;some HTML&lt;/p&gt;</description>
+    <guid isPermaLink="true">http://podcast/post</guid>
+    <enclosure url="http://a/podcast.mp3" length="0" type="audio/mpeg"/>
+    <pubDate>Wed, 05 Dec 2012 00:58:26 +0000</pubDate>
+    <itunes:duration>328</itunes:duration>
+  </item>
+
+</channel>
+</rss>

requirements.freeze.txt

@@ -4,6 +4,7 @@ certifi==2018.4.16
 chardet==3.0.4
 coverage==4.0.3
 -e git+https://github.com/snarfed/gdata-python-client-1.git@1df4e1efea7e5cf2754bc7eec6c1ab48ab09e3b1#egg=gdata
+feedgen==0.7.0
 future==0.16.0
 google-api-python-client==1.7.4
 html2text==2018.1.9

requirements.txt

@@ -1,6 +1,7 @@
 # Keep in sync with setup.py's install_requires!
 beautifulsoup4
 brevity>=0.2.17
+feedgen>=0.7.0
 future
 html2text
 jinja2

setup.py

@@ -46,6 +46,7 @@ def __init__(self, *args, **kwargs):
           # Keep in sync with requirements.txt!
           'beautifulsoup4',
           'brevity>=0.2.17',
+          'feedgen>=0.7.0',
           'future',
           'html2text',
           'jinja2',