atom_to_activity: convert link rel=self/alternate to url

README.md

@@ -293,7 +293,7 @@ Changelog
   * Add new `extract_entries` function.
   * `activity_to_atom`: default actor/author name to username.
   * `atom_to_activities`: support top-level `entry` element as well as `feed`.
-  * `atom_to_activity/ies`: add `object.author`, default `objectType` to `article`/`note` and `verb` to `post`.
+  * `atom_to_activity/ies`: add `object.author`, default `objectType` to `article`/`note` and `verb` to `post`, convert `link rel=self`/`alternate` to `url`.
   * If `entry.author` doesn't have id or url, default them to feed author's.
 * `bluesky`:
   * Fully support both `record` and `object` types in `from_as1` and `to_as1`. Use `to_as1`'s `type` kwarg and `from_as1`'s `out_type` kwarg to disambiguate.

granary/atom.py

@@ -281,7 +281,13 @@ def _atom_to_object(elem, feed_author=None):
   Returns:
     dict: ActivityStreams object
   """
-  uri = _text(elem, 'uri') or _text(elem)
+  self_links = [link for link in elem.iterfind('atom:link', NAMESPACES)
+                if link.get('rel') in ('self', 'alternate')
+                and link.get('type', '').split(';')[0] in ('text/html', '')]
+  uri = (_text(elem, 'uri')
+         or (self_links[0].get('href') if self_links else None)
+         or _text(elem))
+
   title = _text(elem, 'title')
   return {
     'objectType': _as1_value(elem, 'object-type') or 'article' if title else 'note',

granary/tests/test_atom.py

@@ -10,7 +10,7 @@
 from . import test_instagram
 from . import test_twitter
 
-INSTAGRAM_ENTRY_BODY = u"""\
+INSTAGRAM_ENTRY_BODY = """\
 <author>
  <activity:object-type>http://activitystrea.ms/schema/1.0/person</activity:object-type>
  <uri>https://www.instagram.com/snarfed/</uri>
@@ -47,7 +47,7 @@
   <georss:featureName>Le Truc</georss:featureName>
   <link rel="self" href="https://www.instagram.com/p/ABC123/" />
 <link rel="enclosure" href="http://attach/image/big" type="" />"""
-INSTAGRAM_ENTRY = u"""\
+INSTAGRAM_ENTRY = """\
 <?xml version="1.0" encoding="UTF-8"?>
 <entry xml:lang="en-US"
        xmlns="http://www.w3.org/2005/Atom"
@@ -59,7 +59,7 @@
 """ + INSTAGRAM_ENTRY_BODY + """
 </entry>
 """
-INSTAGRAM_FEED = u"""\
+INSTAGRAM_FEED = """\
 <?xml version="1.0" encoding="UTF-8"?>
 <feed xml:lang="en-US"
        xmlns="http://www.w3.org/2005/Atom"
@@ -77,14 +77,16 @@
   'objectType': 'activity',
   'verb': 'post',
   'id': 'tag:instagram.com:123_456',
+  'url': 'https://www.instagram.com/p/ABC123/',
   'actor': {
     'displayName': 'Ryan B',
     'objectType': 'person',
     'url': 'https://www.instagram.com/snarfed/',
   },
   'object': {
-    'id': 'tag:instagram.com:123_456',
     'objectType': 'photo',
+    'id': 'tag:instagram.com:123_456',
+    'url': 'https://www.instagram.com/p/ABC123/',
     'author': {
       'displayName': 'Ryan B',
       'objectType': 'person',
@@ -159,9 +161,9 @@ def test_atom_to_activity_like(self):
       }, atom.atom_to_activity(f"""<?xml version="1.0" encoding="UTF-8"?>
 <entry xmlns="http://www.w3.org/2005/Atom"
        xmlns:activity="http://activitystrea.ms/spec/1.0/">
-<uri>like-url</uri>
-<activity:verb>http://activitystrea.ms/schema/1.0/like</activity:verb>
-<activity:object>{atom_obj}</activity:object>
+  <uri>like-url</uri>
+  <activity:verb>http://activitystrea.ms/schema/1.0/like</activity:verb>
+  <activity:object>{atom_obj}</activity:object>
 </entry>
 """))
 
@@ -231,13 +233,13 @@ def test_atom_to_activity_reply(self):
         'inReplyTo': [{'id': 'foo-id', 'url': 'foo-url'}],
       },
     }
-    self.assert_equals(expected, atom.atom_to_activity(u"""\
+    self.assert_equals(expected, atom.atom_to_activity("""\
 <?xml version="1.0" encoding="UTF-8"?>
 <entry xmlns="http://www.w3.org/2005/Atom"
        xmlns:thr="http://purl.org/syndication/thread/1.0">
-<uri>reply-url</uri>
-<thr:in-reply-to ref="foo-id" href="foo-url" />
-<content>I hereby ☕ reply.</content>
+  <uri>reply-url</uri>
+  <thr:in-reply-to ref="foo-id" href="foo-url" />
+  <content>I hereby ☕ reply.</content>
 </entry>
 """))
 
@@ -251,10 +253,10 @@ def test_atom_to_activity_in_reply_to_text(self):
         'inReplyTo': [{'id': 'my-inreplyto', 'url': 'my-inreplyto'}],
       },
     }
-    self.assert_equals(expected, atom.atom_to_activity(u"""\
+    self.assert_equals(expected, atom.atom_to_activity("""\
 <?xml version="1.0" encoding="UTF-8"?>
 <entry xmlns:thr="http://purl.org/syndication/thread/1.0">
-<thr:in-reply-to>my-inreplyto</thr:in-reply-to>
+  <thr:in-reply-to>my-inreplyto</thr:in-reply-to>
 </entry>
 """))
 
@@ -267,13 +269,46 @@ def test_atom_to_activity_unicode_title(self):
         'objectType': 'article',
         'title': 'How quill’s editor looks',
       },
-    }, atom.atom_to_activity(u"""\
+    }, atom.atom_to_activity("""\
 <?xml version='1.0' encoding='UTF-8'?>
 <entry xmlns='http://www.w3.org/2005/Atom'>
-<title>How quill’s editor looks</title>
+  <title>How quill’s editor looks</title>
 </entry>
 """))
 
+  def test_atom_to_object_uri(self):
+    self.assert_equals({
+      'objectType': 'note',
+      'id': 'http://post',
+      'url': 'http://post',
+    }, atom.atom_to_activity("""\
+<entry xmlns="http://www.w3.org/2005/Atom">
+  <uri>http://post</uri>
+</entry>
+""")['object'])
+
+  def test_atom_to_object_link_self(self):
+    self.assert_equals({
+      'objectType': 'note',
+      'id': 'http://post',
+      'url': 'http://post',
+    }, atom.atom_to_activity("""\
+<entry xmlns="http://www.w3.org/2005/Atom">
+  <link rel="self" href="http://post" />
+</entry>
+""")['object'])
+
+  def test_atom_to_object_link_alternate(self):
+    self.assert_equals({
+      'objectType': 'note',
+      'id': 'http://post',
+      'url': 'http://post',
+    }, atom.atom_to_activity("""\
+<entry xmlns="http://www.w3.org/2005/Atom">
+  <link rel="alternate" type="text/html" href="http://post" />
+</entry>
+""")['object'])
+
   def test_atom_to_activity_use_feed_author_id_url(self):
     expected_author = {
       'id': 'id:ryan',
@@ -716,7 +751,7 @@ def test_html_to_atom_fetch_author(self):
       # fetch_author requires url
       atom.html_to_atom('', fetch_author=True)
 
-    html = u"""\
+    html = """\
 <body class="h-card vcard">
 <img class="photo u-photo" src="photo.jpg" alt=""/>
 <a class="u-url u-uid" rel="author" href="/author"></a>
@@ -737,7 +772,7 @@ def test_html_to_atom_fetch_author(self):
     self.mox.ReplayAll()
 
     got = atom.html_to_atom(html, 'https://my.site/', fetch_author=True)
-    self.assert_multiline_in(u"""\
+    self.assert_multiline_in("""\
 <author>
 <activity:object-type>http://activitystrea.ms/schema/1.0/person</activity:object-type>
 <uri>https://my.site/author</uri>

test_app.py

@@ -536,6 +536,7 @@ def test_url_atom_to_as1(self):
     self.assert_equals({
       'items': [{
         'id': 'https://perma/link',
+        'url': 'https://perma/link',
         'objectType': 'activity',
         'verb': 'post',
         'actor': {
@@ -545,6 +546,7 @@ def test_url_atom_to_as1(self):
         },
         'object': {
           'id': 'https://perma/link',
+          'url': 'https://perma/link',
           'objectType': 'note',
           'author': {
             'objectType': 'person',