Add Slate extractor
This commit is contained in:
parent
fd75482779
commit
5990d0e4c2
|
@ -0,0 +1,39 @@
|
|||
defmodule Frenzy.Pipeline.Extractor.Slate do
|
||||
@moduledoc """
|
||||
Extractor for https://slate.com
|
||||
"""
|
||||
|
||||
alias Frenzy.Pipeline.Extractor
|
||||
@behaviour Extractor
|
||||
|
||||
@impl Extractor
|
||||
def extract(html_tree) do
|
||||
case get_article_content(html_tree) do
|
||||
nil ->
|
||||
{:error, "no matching elements"}
|
||||
|
||||
elem ->
|
||||
{:ok, elem}
|
||||
end
|
||||
end
|
||||
|
||||
defp get_article_content(html_tree) do
|
||||
case Floki.find(html_tree, ".article__content") do
|
||||
[el] ->
|
||||
article_content = Floki.filter_out(el, ".slate-ad, .in-article-recirc, .social-share")
|
||||
|
||||
image = Floki.find(html_tree, ".article__top-image img")
|
||||
|
||||
case image do
|
||||
[] ->
|
||||
article_content
|
||||
|
||||
[image | _] ->
|
||||
[image, article_content]
|
||||
end
|
||||
|
||||
_ ->
|
||||
nil
|
||||
end
|
||||
end
|
||||
end
|
|
@ -113,13 +113,7 @@ defmodule Frenzy.Pipeline.ScrapeStage do
|
|||
|
||||
html = Floki.map(html, rewrite_image_urls(convert_to_data_uris, URI.parse(url)))
|
||||
|
||||
case opts["extractor"] do
|
||||
"builtin" ->
|
||||
{:ok, Readability.readable_html(html)}
|
||||
|
||||
_ ->
|
||||
{:ok, Floki.raw_html(html)}
|
||||
end
|
||||
{:ok, Readability.readable_html(html)}
|
||||
|
||||
res ->
|
||||
res
|
||||
|
|
|
@ -9,6 +9,7 @@ defmodule FrenzyWeb.ConfigureStage.ScrapeStageLive do
|
|||
{"finertech.com", Frenzy.Pipeline.Extractor.FinerTech},
|
||||
{"macstories.net", Frenzy.Pipeline.Extractor.MacStories},
|
||||
{"om.co", Frenzy.Pipeline.Extractor.OmMalik},
|
||||
{"slate.com", Frenzy.Pipeline.Extractor.Slate},
|
||||
{"whatever.scalzi.com", Frenzy.Pipeline.Extractor.WhateverScalzi}
|
||||
]
|
||||
|> Enum.map(fn {pretty_name, module} ->
|
||||
|
|
Loading…
Reference in New Issue