58d4f6aa2cCutting the URL at the first quote dropped the rest of valid links such as ?q="cat"&page=2. Percent-encoding ", < and > keeps the URL complete and still keeps it inside the attribute.
318 lines
9.9 KiB
Elixir
318 lines
9.9 KiB
Elixir
# Pleroma: A lightweight social networking server
|
|
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
|
|
# SPDX-License-Identifier: AGPL-3.0-only
|
|
|
|
defmodule Pleroma.Formatter do
|
|
require Logger
|
|
|
|
alias Pleroma.HTML
|
|
alias Pleroma.User
|
|
|
|
@safe_mention_regex ~r/^(\s*(?<mentions>(@.+?\s+){1,})+)(?<rest>.*)/s
|
|
@link_regex ~r"((?:http(s)?:\/\/)?[\w.-]+(?:\.[\w\.-]+)+[\w\-\._~%:/?#[\]@!\$&'\(\)\*\+,;=.]+)|[0-9a-z+\-\.]+:[0-9a-z$-_.+!*'(),]+"ui
|
|
@markdown_characters_regex ~r/(`|\*|_|{|}|[|]|\(|\)|#|\+|-|\.|!)/
|
|
@markdown_void_elements ~w[area base br col command embed hr img input keygen link meta param source track wbr]
|
|
|
|
defp linkify_opts do
|
|
Pleroma.Config.get(Pleroma.Formatter) ++
|
|
[
|
|
hashtag: true,
|
|
hashtag_handler: &Pleroma.Formatter.hashtag_handler/4,
|
|
mention: true,
|
|
mention_handler: &Pleroma.Formatter.mention_handler/4
|
|
]
|
|
end
|
|
|
|
def escape_mention_handler("@" <> nickname = mention, buffer, _, _) do
|
|
case User.get_cached_by_nickname(nickname) do
|
|
%User{} ->
|
|
# escape markdown characters with `\\`
|
|
# (we don't want something like @user__name to be parsed by markdown)
|
|
String.replace(mention, @markdown_characters_regex, "\\\\\\1")
|
|
|
|
_ ->
|
|
buffer
|
|
end
|
|
end
|
|
|
|
def mention_handler("@" <> nickname, buffer, opts, acc) do
|
|
case User.get_cached_by_nickname(nickname) do
|
|
%User{} = user ->
|
|
{mention_from_user(user, opts),
|
|
%{acc | mentions: MapSet.put(acc.mentions, {"@" <> nickname, user})}}
|
|
|
|
_ ->
|
|
{buffer, acc}
|
|
end
|
|
end
|
|
|
|
def mention_from_user(%User{id: id} = user, opts \\ %{mentions_format: :full}) do
|
|
user_url = user.uri || user.ap_id
|
|
nickname_text = get_nickname_text(user.nickname, opts)
|
|
|
|
Phoenix.HTML.Tag.content_tag(
|
|
:span,
|
|
Phoenix.HTML.Tag.content_tag(
|
|
:a,
|
|
["@", Phoenix.HTML.Tag.content_tag(:span, nickname_text)],
|
|
"data-user": id,
|
|
class: "u-url mention",
|
|
href: user_url,
|
|
rel: "ugc"
|
|
),
|
|
class: "h-card"
|
|
)
|
|
|> Phoenix.HTML.safe_to_string()
|
|
end
|
|
|
|
def hashtag_handler("#" <> tag = tag_text, _buffer, _opts, acc) do
|
|
tag = String.downcase(tag)
|
|
url = "#{Pleroma.Web.Endpoint.url()}/tag/#{tag}"
|
|
|
|
link =
|
|
Phoenix.HTML.Tag.content_tag(:a, tag_text,
|
|
class: "hashtag",
|
|
"data-tag": tag,
|
|
href: url,
|
|
rel: "tag ugc"
|
|
)
|
|
|> Phoenix.HTML.safe_to_string()
|
|
|
|
{link, %{acc | tags: MapSet.put(acc.tags, {tag_text, tag})}}
|
|
end
|
|
|
|
@doc """
|
|
Parses a text and replace plain text links with HTML. Returns a tuple with a result text, mentions, and hashtags.
|
|
|
|
If the 'safe_mention' option is given, only consecutive mentions at the start the post are actually mentioned.
|
|
"""
|
|
@spec linkify(String.t(), keyword()) ::
|
|
{String.t(), [{String.t(), User.t()}], [{String.t(), String.t()}]}
|
|
def linkify(text, options \\ []) do
|
|
options = linkify_opts() ++ options
|
|
|
|
if options[:safe_mention] && Regex.named_captures(@safe_mention_regex, text) do
|
|
%{"mentions" => mentions, "rest" => rest} = Regex.named_captures(@safe_mention_regex, text)
|
|
acc = %{mentions: MapSet.new(), tags: MapSet.new()}
|
|
|
|
{text_mentions, %{mentions: mentions}} = Linkify.link_map(mentions, acc, options)
|
|
{text_rest, %{tags: tags}} = Linkify.link_map(rest, acc, options)
|
|
|
|
{text_mentions <> text_rest, MapSet.to_list(mentions), MapSet.to_list(tags)}
|
|
else
|
|
acc = %{mentions: MapSet.new(), tags: MapSet.new()}
|
|
{text, %{mentions: mentions, tags: tags}} = Linkify.link_map(text, acc, options)
|
|
|
|
{text, MapSet.to_list(mentions), MapSet.to_list(tags)}
|
|
end
|
|
end
|
|
|
|
@doc """
|
|
Escapes a special characters in mention names.
|
|
"""
|
|
def mentions_escape(text, options \\ []) do
|
|
options =
|
|
Keyword.merge(options,
|
|
mention: true,
|
|
url: false,
|
|
mention_handler: &Pleroma.Formatter.escape_mention_handler/4
|
|
)
|
|
|
|
if options[:safe_mention] && Regex.named_captures(@safe_mention_regex, text) do
|
|
%{"mentions" => mentions, "rest" => rest} = Regex.named_captures(@safe_mention_regex, text)
|
|
Linkify.link(mentions, options) <> Linkify.link(rest, options)
|
|
else
|
|
Linkify.link(text, options)
|
|
end
|
|
end
|
|
|
|
def markdown_to_html(text) do
|
|
markdown_to_html(text, %{})
|
|
end
|
|
|
|
def markdown_to_html(text, opts) do
|
|
breaks = Map.get(opts, :breaks, false)
|
|
|
|
{_status, ast, _messages} = EarmarkParser.as_ast(text, breaks: breaks, gfm: true)
|
|
|
|
ast
|
|
|> maybe_strip_email_autolinks(text, Map.get(opts, :strip_email_autolinks, false))
|
|
|> render_markdown_nodes(false)
|
|
|> IO.iodata_to_binary()
|
|
rescue
|
|
exception ->
|
|
Logger.warning("Markdown rendering failed in #{inspect(exception.__struct__)}")
|
|
markdown_fallback(text, Map.get(opts, :breaks, false))
|
|
end
|
|
|
|
defp markdown_fallback(text, breaks) do
|
|
text = text |> escape_html() |> IO.iodata_to_binary() |> maybe_render_markdown_breaks(breaks)
|
|
"<p>" <> text <> "</p>"
|
|
end
|
|
|
|
defp maybe_render_markdown_breaks(text, true), do: String.replace(text, ~r/\r?\n/, "<br>")
|
|
defp maybe_render_markdown_breaks(text, false), do: text
|
|
|
|
defp maybe_strip_email_autolinks(ast, _text, false), do: ast
|
|
|
|
defp maybe_strip_email_autolinks(ast, text, true) do
|
|
emails =
|
|
~r/<([^ >]+@[^ >]+)>/
|
|
|> Regex.scan(text, capture: :all_but_first)
|
|
|> Enum.map(fn [email] -> String.replace_prefix(email, "mailto:", "") end)
|
|
|> MapSet.new()
|
|
|
|
strip_email_autolinks(ast, emails)
|
|
end
|
|
|
|
defp strip_email_autolinks(nodes, emails) when is_list(nodes) do
|
|
Enum.map(nodes, &strip_email_autolinks(&1, emails))
|
|
end
|
|
|
|
defp strip_email_autolinks({"a", attrs, [email], _meta} = node, emails)
|
|
when is_binary(email) do
|
|
if List.keyfind(attrs, "href", 0) == {"href", "mailto:" <> email} and
|
|
MapSet.member?(emails, email) do
|
|
email
|
|
else
|
|
node
|
|
end
|
|
end
|
|
|
|
defp strip_email_autolinks({tag, attrs, children, meta}, emails) do
|
|
{tag, attrs, strip_email_autolinks(children, emails), meta}
|
|
end
|
|
|
|
defp strip_email_autolinks(text, _emails) when is_binary(text), do: text
|
|
|
|
defp render_markdown_nodes(nodes, verbatim) when is_list(nodes) do
|
|
Enum.map(nodes, &render_markdown_node(&1, verbatim))
|
|
end
|
|
|
|
defp render_markdown_node(text, true) when is_binary(text),
|
|
do: strip_dangerous_markdown_html(text)
|
|
|
|
defp render_markdown_node(text, false) when is_binary(text) do
|
|
text
|
|
|> strip_dangerous_markdown_html()
|
|
|> escape_markdown_html()
|
|
end
|
|
|
|
defp render_markdown_node({:comment, _attrs, _children, _meta}, _verbatim), do: []
|
|
|
|
defp render_markdown_node({tag, attrs, children, meta}, verbatim) do
|
|
normalized_tag = String.downcase(tag)
|
|
|
|
if normalized_tag in ["script", "style"] do
|
|
[]
|
|
else
|
|
render_markdown_element(tag, normalized_tag, attrs, children, meta, verbatim)
|
|
end
|
|
end
|
|
|
|
defp render_markdown_element("code", _normalized_tag, attrs, children, _meta, _verbatim) do
|
|
["<code", render_markdown_attrs(attrs), ">", render_markdown_code(children), "</code>"]
|
|
end
|
|
|
|
defp render_markdown_element(tag, "hr", attrs, _children, _meta, _verbatim) do
|
|
attrs = Enum.reject(attrs, &match?({"class", class} when class in ~w[thin medium thick], &1))
|
|
["<", tag, render_markdown_attrs(attrs), "/>"]
|
|
end
|
|
|
|
defp render_markdown_element(tag, normalized_tag, attrs, _children, _meta, _verbatim)
|
|
when normalized_tag in @markdown_void_elements do
|
|
["<", tag, render_markdown_attrs(attrs), ">"]
|
|
end
|
|
|
|
defp render_markdown_element(tag, _normalized_tag, attrs, children, meta, verbatim) do
|
|
children_verbatim = verbatim or Map.get(meta, :verbatim, false)
|
|
|
|
[
|
|
"<",
|
|
tag,
|
|
render_markdown_attrs(attrs),
|
|
">",
|
|
render_markdown_nodes(children, children_verbatim),
|
|
"</",
|
|
tag,
|
|
">"
|
|
]
|
|
end
|
|
|
|
defp render_markdown_attrs(attrs) do
|
|
Enum.map(attrs, fn {name, value} ->
|
|
value = value |> to_string() |> HtmlEntities.decode() |> sanitize_markdown_url(name)
|
|
[" ", to_string(name), "=\"", escape_html(value), "\""]
|
|
end)
|
|
end
|
|
|
|
defp render_markdown_code(children) do
|
|
Enum.map(children, &escape_html/1)
|
|
end
|
|
|
|
defp sanitize_markdown_url(value, name) do
|
|
if String.downcase(to_string(name)) in ["href", "src"] do
|
|
# Percent-encode instead of cutting, so the URL stays complete but can
|
|
# never close the attribute or open a tag.
|
|
String.replace(value, ~r/["<>]/, fn
|
|
"\"" -> "%22"
|
|
"<" -> "%3C"
|
|
">" -> "%3E"
|
|
end)
|
|
else
|
|
value
|
|
end
|
|
end
|
|
|
|
defp strip_dangerous_markdown_html(text) do
|
|
text
|
|
|> String.replace(~r/<(script|style)\b[^>]*>.*?<\/\1>/is, "")
|
|
|> String.replace(~r/<(script|style)\b[^>]*>.*\z/is, "")
|
|
|> String.replace(~r/<\/(script|style)\s*>/i, "")
|
|
end
|
|
|
|
defp escape_markdown_html(text) do
|
|
text |> HtmlEntities.decode() |> escape_html()
|
|
end
|
|
|
|
defp escape_html(text) do
|
|
{:safe, escaped} = Phoenix.HTML.html_escape(text)
|
|
escaped
|
|
end
|
|
|
|
def html_escape({text, mentions, hashtags}, type) do
|
|
{html_escape(text, type), mentions, hashtags}
|
|
end
|
|
|
|
def html_escape(text, "text/html") do
|
|
HTML.filter_tags(text)
|
|
end
|
|
|
|
def html_escape(text, "text/x.misskeymarkdown") do
|
|
HTML.filter_tags(text)
|
|
end
|
|
|
|
def html_escape(text, "text/plain") do
|
|
Regex.split(@link_regex, text, include_captures: true)
|
|
|> Enum.map_every(2, fn chunk ->
|
|
{:safe, part} = Phoenix.HTML.html_escape(chunk)
|
|
part
|
|
end)
|
|
|> Enum.join("")
|
|
end
|
|
|
|
def truncate(text, max_length \\ 200, omission \\ "...") do
|
|
# Remove trailing whitespace
|
|
text = Regex.replace(~r/([^ \t\r\n])([ \t]+$)/u, text, "\\g{1}")
|
|
|
|
if String.length(text) < max_length do
|
|
text
|
|
else
|
|
length_with_omission = max_length - String.length(omission)
|
|
String.slice(text, 0, length_with_omission) <> omission
|
|
end
|
|
end
|
|
|
|
defp get_nickname_text(nickname, %{mentions_format: :full}), do: User.full_nickname(nickname)
|
|
defp get_nickname_text(nickname, _), do: User.local_nickname(nickname)
|
|
end
|