2019-01-10 18:09:56 +00:00
|
|
|
defmodule Pleroma.Web.RichMedia.Parsers.MetaTagsParser do
|
|
|
|
def parse(html, data, prefix, error_message, key_name, value_name \\ "content") do
|
|
|
|
with elements = [_ | _] <- get_elements(html, key_name, prefix),
|
2019-06-12 22:56:51 +00:00
|
|
|
page_title = get_page_title(html),
|
2019-01-10 18:09:56 +00:00
|
|
|
meta_data =
|
|
|
|
Enum.reduce(elements, data, fn el, acc ->
|
|
|
|
attributes = normalize_attributes(el, prefix, key_name, value_name)
|
|
|
|
|
|
|
|
Map.merge(acc, attributes)
|
2019-06-12 22:56:51 +00:00
|
|
|
end)
|
|
|
|
|> Map.put_new(:title, page_title) do
|
2019-01-10 18:09:56 +00:00
|
|
|
{:ok, meta_data}
|
|
|
|
else
|
|
|
|
_e -> {:error, error_message}
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
defp get_elements(html, key_name, prefix) do
|
|
|
|
html |> Floki.find("meta[#{key_name}^='#{prefix}:']")
|
|
|
|
end
|
|
|
|
|
|
|
|
defp normalize_attributes(html_node, prefix, key_name, value_name) do
|
|
|
|
{_tag, attributes, _children} = html_node
|
|
|
|
|
|
|
|
data =
|
|
|
|
Enum.into(attributes, %{}, fn {name, value} ->
|
|
|
|
{name, String.trim_leading(value, "#{prefix}:")}
|
|
|
|
end)
|
|
|
|
|
|
|
|
%{String.to_atom(data[key_name]) => data[value_name]}
|
|
|
|
end
|
2019-06-12 22:56:51 +00:00
|
|
|
|
|
|
|
defp get_page_title(html) do
|
|
|
|
Floki.find(html, "title") |> Floki.text()
|
|
|
|
end
|
2019-01-10 18:09:56 +00:00
|
|
|
end
|