annot.at is a service for syncing static websites, such as blogs, to atproto and standard.site
annot.at
elixir
atproto
standardsite
Grab feed categories and strip text content
I realized I was passing the HTML string to textContent and it feels more accurate to extract the text instead.
Also, adds feed tag extraction so the standard.site documents get nice tags!
Johanna Larsson
(Jul 1, 2026, 7:15 AM +0100)
58fd1bbc
09e71ce3
+30
-8
+1
-1
README.md
+1
-1
README.md
···
14
14
- [x] Maybe some design?
15
15
- [x] Add site including verification
16
16
- [ ] Write a guide for using this
17
-
- [ ] Document cover images
17
+
- [x] Document cover images
18
18
- [ ] Document content with html content type
19
19
- [ ] RSS poller
20
20
- [ ] standard.site document reader
+3
-1
lib/annot_at/feeds/entry.ex
+3
-1
lib/annot_at/feeds/entry.ex
···
15
15
:content,
16
16
:rkey,
17
17
:image,
18
+
:categories,
18
19
cover_status: :none
19
20
]
20
21
···
29
30
content: String.t() | nil,
30
31
rkey: String.t() | nil,
31
32
image: String.t() | nil,
32
-
cover_status: cover_status()
33
+
cover_status: cover_status(),
34
+
categories: [String.t()]
33
35
}
34
36
35
37
def hash(%__MODULE__{} = entry) do
+14
-5
lib/annot_at/feeds/rss.ex
+14
-5
lib/annot_at/feeds/rss.ex
···
53
53
{:ok,
54
54
%{
55
55
state
56
-
| entries: [%Entry{} | state.entries],
56
+
| entries: [%Entry{categories: []} | state.entries],
57
57
stack: ["item" | state.stack],
58
58
current_text: []
59
59
}}
···
105
105
defp apply_entry_field(entry, "description", text), do: %{entry | summary: text}
106
106
defp apply_entry_field(entry, "content:encoded", text), do: %{entry | content: text}
107
107
defp apply_entry_field(entry, "pubDate", text), do: %{entry | published_at: parse_date(text)}
108
+
109
+
defp apply_entry_field(entry, "category", text),
110
+
do: %{entry | categories: [text | entry.categories]}
111
+
108
112
defp apply_entry_field(entry, _name, _text), do: entry
109
113
110
114
defp apply_feed_field(feed, "title", text), do: %{feed | title: text}
···
112
116
defp apply_feed_field(feed, "link", text), do: %{feed | url: text}
113
117
defp apply_feed_field(feed, _name, _text), do: feed
114
118
115
-
defp finalize_entry(%Entry{id: nil, url: url} = entry) when is_binary(url) do
116
-
%{entry | id: url}
117
-
end
119
+
defp finalize_entry(%Entry{} = entry) do
120
+
id =
121
+
if is_nil(entry.id) and is_binary(entry.url) do
122
+
entry.url
123
+
else
124
+
entry.id
125
+
end
118
126
119
-
defp finalize_entry(entry), do: entry
127
+
%{entry | id: id, categories: Enum.reverse(entry.categories)}
128
+
end
120
129
121
130
defp text(parts) do
122
131
result =
+9
-1
lib/annot_at_web/controllers/live/posts_live.ex
+9
-1
lib/annot_at_web/controllers/live/posts_live.ex
···
363
363
path: path_of(entry.url),
364
364
published_at: entry.published_at,
365
365
description: entry.summary,
366
-
text_content: entry.content,
366
+
text_content: text_content_of(entry.content),
367
367
cover_image: fetch_cover(entry)
368
368
}
369
369
end
370
370
371
371
defp path_of(nil), do: nil
372
372
defp path_of(url), do: URI.parse(url).path
373
+
374
+
defp text_content_of(nil), do: nil
375
+
376
+
defp text_content_of(content) do
377
+
content
378
+
|> LazyHTML.from_fragment()
379
+
|> LazyHTML.text()
380
+
end
373
381
374
382
defp create_document(scope, site, entry) do
375
383
document = to_document(entry, site, scope.user)
+1
test/annot_at/feeds/rss_test.exs
+1
test/annot_at/feeds/rss_test.exs
···
27
27
assert "<p>The full <strong>content</strong> of the first post.</p>" == first.content
28
28
assert %DateTime{} = first.published_at
29
29
assert ~U[2024-10-02 13:00:00Z] == first.published_at
30
+
assert ["category-1", "category-2"] = first.categories
30
31
end
31
32
32
33
test "falls back to URL as ID when guid is missing" do
+2
test/support/fixtures/feeds/rss_sample.xml
+2
test/support/fixtures/feeds/rss_sample.xml
···
17
17
<content:encoded
18
18
><![CDATA[<p>The full <strong>content</strong> of the first post.</p>]]></content:encoded>
19
19
<pubDate>Wed, 02 Oct 2024 13:00:00 GMT</pubDate>
20
+
<category>category-1</category>
21
+
<category>category-2</category>
20
22
</item>
21
23
<item>
22
24
<title>Second Post</title>