feed_parser/lib/parser/atom.ex

92 lines
2.0 KiB
Elixir
Raw Normal View History

2019-09-01 03:25:10 +00:00
defmodule FeedParser.Parser.Atom do
@moduledoc """
A `FeedParser.Parser` that handles [Atom feeds](https://validator.w3.org/feed/docs/atom.html).
"""
2019-08-31 23:02:41 +00:00
alias FeedParser.XML
require XML
@behaviour FeedParser.Parser
@impl FeedParser.Parser
def accepts(data, content_type) do
case content_type do
"application/atom+xml" ->
2019-09-01 19:49:30 +00:00
{true, XML.parse(data)}
2019-08-31 23:02:41 +00:00
_ when content_type in ["text/xml", "application/xml"] ->
doc = XML.parse(data)
if XML.xmlElement(doc, :name) == :feed do
{true, doc}
else
false
end
_ ->
false
end
end
@impl FeedParser.Parser
def parse_feed(feed) do
title = text('/feed/title/text()', feed)
link = attr('/feed/link/@href', feed)
icon = text('/feed/icon/text()', feed)
items =
:xmerl_xpath.string('/feed/entry', feed)
|> Enum.map(fn entry ->
id = text('/entry/id/text()', entry)
title = text('/entry/title/text()', entry)
link = attr('/entry/link/@href', entry)
updated =
text('/entry/updated/text()', entry)
|> Timex.parse("{ISO:Extended}")
|> case do
{:ok, date} -> date
_ -> nil
end
content = text('/entry/content/text()', entry) || text('/entry/summary/text()', entry)
%FeedParser.Item{
guid: id,
title: title,
url: link,
content: content,
date: updated
}
end)
{:ok,
%FeedParser.Feed{
site_url: link,
title: title,
image_url: icon,
items: items
}}
end
defp text(xpath, element) do
case :xmerl_xpath.string(xpath, element) do
[el] ->
XML.xmlText(el, :value) |> List.to_string() |> String.trim()
_ ->
nil
end
end
defp attr(xpath, element) do
case :xmerl_xpath.string(xpath, element) do
[attr] ->
XML.xmlAttribute(attr, :value) |> List.to_string() |> String.trim()
_ ->
nil
end
end
end