PonyMirrors/philomena
1
0
Fork 0
mirror of https://github.com/philomena-dev/philomena.git synced 2025-02-20 04:14:23 +01:00
Code Issues Projects Releases Packages Wiki Activity

fix tumblr scrapes for text posts

Browse source
This commit is contained in:
byte[] 2020-08-28 13:51:26 -04:00
parent 6a3d292983
commit 2347d018ca
1 changed files with 5 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
lib/philomena/scrapers/tumblr.ex
View file

@ -1,6 +1,6 @@
defmodule Philomena.Scrapers.Tumblr do defmodule Philomena.Scrapers.Tumblr do
@url_regex ~r|\Ahttps?://(?:.*)/(?:image\|post)/(\d+)(?:\z\|[/?#])| @url_regex ~r|\Ahttps?://(?:.*)/(?:image\|post)/(\d+)(?:\z\|[/?#])|
@inline_regex ~r|https?://(?:\d+\.)?media\.tumblr\.com\/[a-f\d]+\/tumblr(?:_inline)?_[a-z\d]+_\d+\.(?:png\|jpe?g\|gif)|i @media_regex ~r|https?://(?:\d+\.)?media\.tumblr\.com/[a-f\d]+/[a-f\d]+-[a-f\d]+/s\d+x\d+/[a-f\d]+\.(?:png\|jpe?g\|gif)|i
@size_regex ~r|_(\d+)(\..+)\z| @size_regex ~r|_(\d+)(\..+)\z|
@sizes [1280, 540, 500, 400, 250, 100, 75] @sizes [1280, 540, 500, 400, 250, 100, 75]
@tumblr_ranges [ @tumblr_ranges [
@ -54,10 +54,10 @@ defmodule Philomena.Scrapers.Tumblr do
defp process_post!(%{"type" => "text"} = post) do defp process_post!(%{"type" => "text"} = post) do
images = images =
@inline_regex @media_regex
|> Regex.scan(post["text"]) |> Regex.scan(post["body"])
|> Enum.map(fn url -> |> Enum.map(fn [url | _captures] ->
%{url: upsize(url), camo_url: Camo.Image.image_url(url)} %{url: url, camo_url: Camo.Image.image_url(url)}
end) end)
add_meta(post, images) add_meta(post, images)