Skip to content

A platform for community discussion. Free, open, simple.

FIX: Escape markdown characters in upload filenames (#39133)

Régis HANOL13280c102312
diff --git a/app/services/inline_uploads.rb b/app/services/inline_uploads.rb
index ca58c977701..7b3c35b182f 100644
--- a/app/services/inline_uploads.rb
+++ b/app/services/inline_uploads.rb
@@ -199,7 +199,7 @@ class InlineUploads
has_attachment = node.attributes["class"]&.value
index = $~.offset(0)[0]
text = match[2].strip.gsub("\n", "").gsub(/ +/, " ")
- text = "#{text}|attachment" if has_attachment
+ text = "#{UploadMarkdown.escape_markdown(text)}|attachment" if has_attachment
yield(match[0], href, +"[#{text}](#{PLACEHOLDER})", index) if block_given?
end
@@ -255,9 +255,9 @@ class InlineUploads
raw =
raw.gsub(%r{^(https?://\S+)(\s?)$}) do |match|
if upload = blk.call(match)
- filename_modified = upload.original_filename&.gsub(/[\[\]\|]/, "").to_s
+ filename_modified = upload.original_filename.to_s
filename_modified = File.basename(filename_modified, File.extname(filename_modified))
- "![#{filename_modified}](#{upload.short_url})"
+ "![#{UploadMarkdown.escape_markdown(filename_modified)}](#{upload.short_url})"
else
match
end
diff --git a/frontend/discourse-markdown-it/src/engine.js b/frontend/discourse-markdown-it/src/engine.js
index 21a1ddf4d97..c8e6e5df7e8 100644
--- a/frontend/discourse-markdown-it/src/engine.js
+++ b/frontend/discourse-markdown-it/src/engine.js
@@ -173,16 +173,35 @@ function audioHTML(token) {
const IMG_SIZE_REGEX =
/^([1-9]+[0-9]*)x([1-9]+[0-9]*)(\s*,\s*(x?)([1-9][0-9]{0,2}?)([%x]?))?$/;
+function isKnownImageSuffix(segment) {
+ return (
+ segment === "video" ||
+ segment === "audio" ||
+ segment === "thumbnail" ||
+ IMG_SIZE_REGEX.test(segment) ||
+ !!extractDataAttribute(segment)
+ );
+}
+
function renderImageOrPlayableMedia(tokens, idx, options, env, slf) {
const token = tokens[idx];
const alt = slf.renderInlineAsText(token.children, options, env);
const split = alt.split("|");
- const altSplit = [split[0]];
+
+ // Find where structured suffixes begin by scanning from the right.
+ // Everything before that boundary is alt text (which may contain pipes).
+ let suffixStart = split.length;
+ while (suffixStart > 1 && isKnownImageSuffix(split[suffixStart - 1])) {
+ suffixStart--;
+ }
+
+ const altText = split.slice(0, suffixStart).join("|");
+ const suffixes = split.slice(suffixStart);
// markdown-it supports returning HTML instead of continuing to render the current token
// see https://github.com/markdown-it/markdown-it/blob/master/docs/architecture.md#renderer
// handles |video and |audio alt transformations for image tags
- if (split[1] === "video") {
+ if (suffixes[0] === "video") {
if (
options.discourse.previewing &&
!options.discourse.limitedSiteSettings.enableDiffhtmlPreview
@@ -197,13 +216,13 @@ function renderImageOrPlayableMedia(tokens, idx, options, env, slf) {
} else {
return videoHTML(token);
}
- } else if (split[1] === "audio") {
+ } else if (suffixes[0] === "audio") {
return audioHTML(token);
}
// parsing ![myimage|500x300]() or ![myimage|75%]() or ![myimage|500x300, 75%]
- for (let i = 1, match, data; i < split.length; ++i) {
- if ((match = split[i].match(IMG_SIZE_REGEX)) && match[1] && match[2]) {
+ for (let i = 0, match, data; i < suffixes.length; ++i) {
+ if ((match = suffixes[i].match(IMG_SIZE_REGEX)) && match[1] && match[2]) {
let width = match[1];
let height = match[2];
@@ -243,20 +262,17 @@ function renderImageOrPlayableMedia(tokens, idx, options, env, slf) {
) {
token.attrs.push(["class", "resizable"]);
}
- } else if ((data = extractDataAttribute(split[i]))) {
+ } else if ((data = extractDataAttribute(suffixes[i]))) {
token.attrs.push(data);
- } else if (split[i] === "thumbnail") {
+ } else if (suffixes[i] === "thumbnail") {
token.attrs.push(["data-thumbnail", "true"]);
- } else {
- altSplit.push(split[i]);
}
}
- const altValue = altSplit.join("|").trim();
- if (altValue === "") {
+ if (altText === "") {
token.attrSet("role", "presentation");
} else {
- token.attrSet("alt", altValue);
+ token.attrSet("alt", altText);
}
return slf.renderToken(tokens, idx, options);
@@ -274,23 +290,36 @@ export const ATTACHMENT_CSS_CLASS = "attachment";
function renderAttachment(tokens, idx, options, env, slf) {
const linkToken = tokens[idx];
- const textToken = tokens[idx + 1];
-
- const split = textToken.content.split("|");
- const contentSplit = [];
- for (let i = 0, data; i < split.length; ++i) {
- if (split[i] === ATTACHMENT_CSS_CLASS) {
- linkToken.attrs.unshift(["class", split[i]]);
- } else if ((data = extractDataAttribute(split[i]))) {
- linkToken.attrs.push(data);
- } else {
- contentSplit.push(split[i]);
+ // Link text may span multiple tokens when it contains inline formatting.
+ // Scan forward to find the last text token which holds the |attachment marker.
+ let lastTextToken = null;
+ for (let i = idx + 1; i < tokens.length; i++) {
+ if (tokens[i].type === "link_close") {
+ break;
+ }
+ if (tokens[i].type === "text" && tokens[i].content) {
+ lastTextToken = tokens[i];
}
}
- if (contentSplit.length > 0) {
- textToken.content = contentSplit.join("|");
+ if (lastTextToken) {
+ const split = lastTextToken.content.split("|");
+ const contentSplit = [];
+
+ for (let i = 0, data; i < split.length; ++i) {
+ if (split[i] === ATTACHMENT_CSS_CLASS) {
+ linkToken.attrs.unshift(["class", split[i]]);
+ } else if ((data = extractDataAttribute(split[i]))) {
+ linkToken.attrs.push(data);
+ } else {
+ contentSplit.push(split[i]);
+ }
+ }
+
+ if (contentSplit.length > 0) {
+ lastTextToken.content = contentSplit.join("|");
+ }
}
return slf.renderToken(tokens, idx, options);
diff --git a/frontend/discourse/app/lib/markdown-image-builder.js b/frontend/discourse/app/lib/markdown-image-builder.js
index 1e578033ba2..18e15b15f78 100644
--- a/frontend/discourse/app/lib/markdown-image-builder.js
+++ b/frontend/discourse/app/lib/markdown-image-builder.js
@@ -1,3 +1,7 @@
+export function escapeMarkdownCharacters(text) {
+ return text.replace(/[\\*_~`\[\]|]/g, "\\$&");
+}
+
export function sanitizeAlt(text, options = {}) {
const fallback = options.fallback ?? "";
@@ -10,7 +14,7 @@ export function sanitizeAlt(text, options = {}) {
return fallback;
}
- return trimmed.replace(/\|/g, "&#124;").replace(/([\\\[\]])/g, "\\$1");
+ return escapeMarkdownCharacters(trimmed);
}
/**
diff --git a/frontend/discourse/app/lib/to-markdown.js b/frontend/discourse/app/lib/to-markdown.js
index 440b392b2ec..5c43217e190 100644
--- a/frontend/discourse/app/lib/to-markdown.js
+++ b/frontend/discourse/app/lib/to-markdown.js
@@ -1,5 +1,6 @@
import {
buildImageMarkdown,
+ escapeMarkdownCharacters,
extensionFromUrl,
} from "discourse/lib/markdown-image-builder";
@@ -396,6 +397,7 @@ export class Tag {
let linkModifier = "";
if (attr.class?.includes("attachment")) {
+ text = escapeMarkdownCharacters(text);
linkModifier = "|attachment";
}
diff --git a/frontend/discourse/app/lib/uploads.js b/frontend/discourse/app/lib/uploads.js
index 052c530aeed..7b5b1f77fdf 100644
--- a/frontend/discourse/app/lib/uploads.js
+++ b/frontend/discourse/app/lib/uploads.js
@@ -1,5 +1,6 @@
import deprecated from "discourse/lib/deprecated";
import { getOwnerWithFallback } from "discourse/lib/get-owner";
+import { escapeMarkdownCharacters } from "discourse/lib/markdown-image-builder";
import { humanizeList } from "discourse/lib/text";
import { capabilities } from "discourse/services/capabilities";
import I18n, { i18n } from "discourse-i18n";
@@ -284,7 +285,7 @@ function markdownNameFromFileName(fileName) {
name = i18n("upload_selector.default_image_alt_text");
}
- return name.replace(/\[|\]|\|/g, "");
+ return escapeMarkdownCharacters(name);
}
function imageMarkdown(upload) {
@@ -300,7 +301,7 @@ function playableMediaMarkdown(upload, type) {
}
function attachmentMarkdown(upload) {
- return `[${upload.original_filename}|attachment](${
+ return `[${escapeMarkdownCharacters(upload.original_filename)}|attachment](${
upload.short_url
}) (${I18n.toHumanSize(upload.filesize)})`;
}
diff --git a/frontend/discourse/app/static/prosemirror/extensions/link.js b/frontend/discourse/app/static/prosemirror/extensions/link.js
index 01013bf36fc..ff3fa9d750c 100644
--- a/frontend/discourse/app/static/prosemirror/extensions/link.js
+++ b/frontend/discourse/app/static/prosemirror/extensions/link.js
@@ -53,10 +53,20 @@ const extension = {
link: {
mark: "link",
getAttrs(tok, tokens, i) {
- const nextContent = tokens[i + 1].content;
- const attachment = nextContent.endsWith("|attachment");
- if (attachment) {
- tokens[i + 1].content = nextContent.slice(0, -11);
+ // Link text may span multiple tokens when it contains inline formatting.
+ let attachment = false;
+ for (let j = i + 1; j < tokens.length; j++) {
+ if (tokens[j].type === "link_close") {
+ break;
+ }
+ if (
+ tokens[j].type === "text" &&
+ tokens[j].content.endsWith("|attachment")
+ ) {
+ tokens[j].content = tokens[j].content.slice(0, -11);
+ attachment = true;
+ break;
+ }
}
return {
diff --git a/frontend/discourse/tests/unit/lib/markdown-image-builder-test.js b/frontend/discourse/tests/unit/lib/markdown-image-builder-test.js
index 7d1933da750..872e0e75e1e 100644
--- a/frontend/discourse/tests/unit/lib/markdown-image-builder-test.js
+++ b/frontend/discourse/tests/unit/lib/markdown-image-builder-test.js
@@ -23,14 +23,11 @@ module("Unit | Lib | markdown-image-builder", function () {
assert.strictEqual(sanitizeAlt(" ", { fallback: "image" }), "image");
});
- test("escapes pipes for markdown", function (assert) {
+ test("escapes markdown special characters", function (assert) {
assert.strictEqual(
sanitizeAlt("alt|text|with|pipes"),
- "alt&#124;text&#124;with&#124;pipes"
+ "alt\\|text\\|with\\|pipes"
);
- });
-
- test("escapes backslashes, brackets", function (assert) {
assert.strictEqual(
sanitizeAlt("text\\with\\slashes"),
"text\\\\with\\\\slashes"
@@ -39,6 +36,14 @@ module("Unit | Lib | markdown-image-builder", function () {
sanitizeAlt("text[with]brackets"),
"text\\[with\\]brackets"
);
+ assert.strictEqual(
+ sanitizeAlt("_underscores_ and *stars*"),
+ "\\_underscores\\_ and \\*stars\\*"
+ );
+ assert.strictEqual(
+ sanitizeAlt("~~strike~~ and `code`"),
+ "\\~\\~strike\\~\\~ and \\`code\\`"
+ );
});
test("trims whitespace", function (assert) {
@@ -137,7 +142,7 @@ module("Unit | Lib | markdown-image-builder", function () {
src: "/uploads/image.png",
alt: "text|with|pipes",
}),
- "![text&#124;with&#124;pipes](/uploads/image.png)"
+ "![text\\|with\\|pipes](/uploads/image.png)"
);
});
diff --git a/frontend/discourse/tests/unit/lib/uploads-test.js b/frontend/discourse/tests/unit/lib/uploads-test.js
index ff3d421fe59..666900802d7 100644
--- a/frontend/discourse/tests/unit/lib/uploads-test.js
+++ b/frontend/discourse/tests/unit/lib/uploads-test.js
@@ -321,7 +321,7 @@ module("Unit | Utility | uploads", function (hooks) {
);
assert.strictEqual(
testUploadMarkdown("[foo|bar].png"),
- "![foobar|100x200](/uploads/123/abcdef.ext)"
+ "![\\[foo\\|bar\\]|100x200](/uploads/123/abcdef.ext)"
);
assert.strictEqual(
testUploadMarkdown("file name with space.png"),
@@ -339,6 +339,11 @@ module("Unit | Utility | uploads", function (hooks) {
testUploadMarkdown("important.txt", { short_url }),
`[important.txt|attachment](${short_url}) (42 Bytes)`
);
+
+ assert.strictEqual(
+ testUploadMarkdown("_test_file_.txt", { short_url }),
+ `[\\_test\\_file\\_.txt|attachment](${short_url}) (42 Bytes)`
+ );
});
test("getUploadMarkdown - replaces GUID in image alt text on iOS", function (assert) {
diff --git a/lib/upload_markdown.rb b/lib/upload_markdown.rb
index 5fdecb14b52..4d83fccc339 100644
--- a/lib/upload_markdown.rb
+++ b/lib/upload_markdown.rb
@@ -17,14 +17,14 @@ class UploadMarkdown
def image_markdown(display_name: nil)
display_name ||= @upload.original_filename
- "![#{display_name}|#{@upload.width}x#{@upload.height}](#{@upload.short_url})"
+ "![#{self.class.escape_markdown(display_name)}|#{@upload.width}x#{@upload.height}](#{@upload.short_url})"
end
def attachment_markdown(display_name: nil, with_filesize: true)
human_filesize = with_filesize ? " (#{@upload.human_filesize})" : ""
display_name ||= @upload.original_filename
- "[#{display_name}|attachment](#{@upload.short_url})#{human_filesize}"
+ "[#{self.class.escape_markdown(display_name)}|attachment](#{@upload.short_url})#{human_filesize}"
end
def playable_media_markdown(display_name: nil)
@@ -36,6 +36,10 @@ class UploadMarkdown
end
return attachment_markdown if !type
display_name ||= @upload.original_filename
- "![#{display_name}|#{type}](#{@upload.short_url})"
+ "![#{self.class.escape_markdown(display_name)}|#{type}](#{@upload.short_url})"
+ end
+
+ def self.escape_markdown(text)
+ text.gsub(/[\\*_~`\[\]|]/) { |c| "\\#{c}" }
end
end
diff --git a/plugins/chat/spec/models/chat/message_spec.rb b/plugins/chat/spec/models/chat/message_spec.rb
index 50573e7bb18..4d23aa01864 100644
--- a/plugins/chat/spec/models/chat/message_spec.rb
+++ b/plugins/chat/spec/models/chat/message_spec.rb
@@ -661,7 +661,7 @@ describe Chat::Message do
expect(message.to_markdown).to eq(<<~MSG.chomp)
hey friend, what's up?!
- ![test_image.jpg|400x300](#{image.short_url})
+ ![test\\_image.jpg|400x300](#{image.short_url})
![meme.jpg|10x10](#{image2.short_url})
MSG
end
diff --git a/spec/jobs/export_csv_file_spec.rb b/spec/jobs/export_csv_file_spec.rb
index ce24c2c65d7..d9e150c1516 100644
--- a/spec/jobs/export_csv_file_spec.rb
+++ b/spec/jobs/export_csv_file_spec.rb
@@ -34,8 +34,7 @@ RSpec.describe Jobs::ExportCsvFile do
expect(system_message.first_post.raw).to eq(
I18n.t(
"system_messages.csv_export_succeeded.text_body_template",
- download_link:
- "[#{upload.original_filename}|attachment](#{upload.short_url}) (#{upload.filesize} Bytes)",
+ download_link: UploadMarkdown.new(upload).attachment_markdown,
).chomp,
)
diff --git a/spec/jobs/export_user_archive_spec.rb b/spec/jobs/export_user_archive_spec.rb
index 54055141c9a..455a7889255 100644
--- a/spec/jobs/export_user_archive_spec.rb
+++ b/spec/jobs/export_user_archive_spec.rb
@@ -85,8 +85,7 @@ RSpec.describe Jobs::ExportUserArchive do
expect(system_message.first_post.raw).to eq(
I18n.t(
"system_messages.csv_export_succeeded.text_body_template",
- download_link:
- "[#{upload.original_filename}|attachment](#{upload.short_url}) (#{upload.human_filesize})",
+ download_link: UploadMarkdown.new(upload).attachment_markdown,
).chomp,
)
@@ -145,8 +144,7 @@ RSpec.describe Jobs::ExportUserArchive do
expect(system_message.first_post.raw).to eq(
I18n.t(
"system_messages.csv_export_succeeded.text_body_template",
- download_link:
- "[#{upload.original_filename}|attachment](#{upload.short_url}) (#{upload.human_filesize})",
+ download_link: UploadMarkdown.new(upload).attachment_markdown,
).chomp,
)
end
diff --git a/spec/lib/pretty_text_spec.rb b/spec/lib/pretty_text_spec.rb
index 59ab4a2f8c5..72580a0a5f6 100644
--- a/spec/lib/pretty_text_spec.rb
+++ b/spec/lib/pretty_text_spec.rb
@@ -2593,6 +2593,27 @@ HTML
expect(PrettyText.cook(raw)).to eq(cooked.strip)
end
+ it "handles attachment filenames with markdown characters" do
+ SiteSetting.authorized_extensions = "txt"
+
+ {
+ "_test_file_.txt" => "<em>",
+ "*test*.txt" => "<em>",
+ "**bold**.txt" => "<strong>",
+ "~~strike~~.txt" => "<s>",
+ "`code`.txt" => "<code>",
+ }.each do |filename, bad_tag|
+ upload = Fabricate(:upload, original_filename: filename, extension: "txt")
+ markdown = UploadMarkdown.new(upload).to_markdown
+ cooked = PrettyText.cook(markdown)
+
+ expect(cooked).to include('class="attachment"'),
+ "expected attachment class for filename: #{filename}\nmarkdown: #{markdown}\ncooked: #{cooked}"
+ expect(cooked).not_to include(bad_tag),
+ "unexpected #{bad_tag} for filename: #{filename}\nmarkdown: #{markdown}\ncooked: #{cooked}"
+ end
+ end
+
it "can place a blank image if we can not find the upload" do
raw = <<~MD
![upload](upload://abcABC.png)
diff --git a/spec/lib/upload_markdown_spec.rb b/spec/lib/upload_markdown_spec.rb
index 5166a10bb36..6b910fd1833 100644
--- a/spec/lib/upload_markdown_spec.rb
+++ b/spec/lib/upload_markdown_spec.rb
@@ -16,21 +16,44 @@ RSpec.describe UploadMarkdown do
)
expect(UploadMarkdown.new(video).to_markdown).to eq(<<~MD.chomp)
- ![test_video.mp4|video](#{video.short_url})
+ ![test\\_video.mp4|video](#{video.short_url})
MD
expect(UploadMarkdown.new(audio).to_markdown).to eq(<<~MD.chomp)
- ![test_audio.mp3|audio](#{audio.short_url})
+ ![test\\_audio.mp3|audio](#{audio.short_url})
MD
expect(UploadMarkdown.new(attachment).to_markdown).to eq(<<~MD.chomp)
- [test_file.pdf|attachment](#{attachment.short_url}) (#{attachment.human_filesize})
+ [test\\_file.pdf|attachment](#{attachment.short_url}) (#{attachment.human_filesize})
MD
expect(UploadMarkdown.new(image).to_markdown).to eq(<<~MD.chomp)
- ![test_img.jpg|100x200](#{image.short_url})
+ ![test\\_img.jpg|100x200](#{image.short_url})
MD
unknown = Fabricate(:upload, original_filename: "test_video.mmmppp444", extension: "mmmppp444")
expect(UploadMarkdown.new(unknown).playable_media_markdown).to eq(<<~MD.chomp)
- [test_video.mmmppp444|attachment](#{unknown.short_url}) (#{unknown.human_filesize})
+ [test\\_video.mmmppp444|attachment](#{unknown.short_url}) (#{unknown.human_filesize})
MD
end
+
+ it "escapes markdown characters in attachment filenames" do
+ SiteSetting.authorized_extensions = "txt"
+
+ {
+ "_test_file_.txt" => "<em>",
+ "*test*.txt" => "<em>",
+ "**bold**.txt" => "<strong>",
+ "~~strike~~.txt" => "<s>",
+ "`code`.txt" => "<code>",
+ }.each do |filename, bad_tag|
+ upload = Fabricate(:upload, original_filename: filename, extension: "txt")
+ markdown = UploadMarkdown.new(upload).attachment_markdown
+ cooked = PrettyText.cook(markdown)
+
+ expect(cooked).to include('class="attachment"'),
+ "expected attachment class for filename: #{filename}\ncooked: #{cooked}"
+ expect(cooked).not_to include(bad_tag),
+ "unexpected #{bad_tag} in cooked output for filename: #{filename}\ncooked: #{cooked}"
+ expect(cooked).to include(filename.gsub("\\", "")),
+ "expected display name for filename: #{filename}\ncooked: #{cooked}"
+ end
+ end
end
diff --git a/spec/services/inline_uploads_spec.rb b/spec/services/inline_uploads_spec.rb
index 50e878d8f40..fbbe7ff9842 100644
--- a/spec/services/inline_uploads_spec.rb
+++ b/spec/services/inline_uploads_spec.rb
@@ -613,7 +613,7 @@ RSpec.describe InlineUploads do
[test3|attachment](#{upload.short_url})
[test3|attachment](#{upload2.short_url})[test3|attachment](#{upload3.short_url})
- [This is some _test_ here|attachment](#{upload3.short_url})
+ [This is some \\_test\\_ here|attachment](#{upload3.short_url})
MD
end
@@ -762,7 +762,7 @@ RSpec.describe InlineUploads do
image_upload
end
- expect(raw).to eq("look at this:\n![image1](#{image_upload.short_url})")
+ expect(raw).to eq("look at this:\n![image\\]1](#{image_upload.short_url})")
end
end
end