aboutsummaryrefslogtreecommitdiffstats
path: root/internal/markdown
diff options
context:
space:
mode:
Diffstat (limited to 'internal/markdown')
-rw-r--r--internal/markdown/render.go38
-rw-r--r--internal/markdown/render_test.go29
2 files changed, 67 insertions, 0 deletions
diff --git a/internal/markdown/render.go b/internal/markdown/render.go
new file mode 100644
index 0000000..f9a2bfd
--- /dev/null
+++ b/internal/markdown/render.go
@@ -0,0 +1,38 @@
+// Package markdown renders untrusted Markdown to sanitized HTML.
+package markdown
+
+import (
+ "bytes"
+
+ "github.com/microcosm-cc/bluemonday"
+ "github.com/yuin/goldmark"
+ "github.com/yuin/goldmark/extension"
+ "github.com/yuin/goldmark/parser"
+ "github.com/yuin/goldmark/renderer/html"
+)
+
+var md = goldmark.New(
+ goldmark.WithExtensions(extension.GFM, extension.Typographer),
+ goldmark.WithParserOptions(parser.WithAutoHeadingID()),
+ goldmark.WithRendererOptions(html.WithHardWraps(), html.WithUnsafe()), // unsafe output is sanitized below
+)
+
+var policy = func() *bluemonday.Policy {
+ p := bluemonday.UGCPolicy()
+ p.AllowAttrs("id").OnElements("h1", "h2", "h3", "h4", "h5", "h6")
+ p.AllowAttrs("class").Matching(bluemonday.SpaceSeparatedTokens).OnElements("code", "pre", "span", "div", "table", "input", "li", "ul")
+ p.AllowAttrs("type", "checked", "disabled").OnElements("input")
+ p.AllowAttrs("align").OnElements("th", "td")
+ p.AllowAttrs("width", "height").OnElements("img")
+ p.RequireNoFollowOnLinks(false)
+ return p
+}()
+
+// Render converts Markdown to HTML that is safe to embed unescaped.
+func Render(src string) string {
+ var buf bytes.Buffer
+ if err := md.Convert([]byte(src), &buf); err != nil {
+ return "<p>(could not render content)</p>"
+ }
+ return policy.Sanitize(buf.String())
+}
diff --git a/internal/markdown/render_test.go b/internal/markdown/render_test.go
new file mode 100644
index 0000000..7240afc
--- /dev/null
+++ b/internal/markdown/render_test.go
@@ -0,0 +1,29 @@
+package markdown
+
+import (
+ "strings"
+ "testing"
+)
+
+func TestRenderSanitizes(t *testing.T) {
+ out := Render("# Hi\n\n**bold** <script>alert(1)</script> <a href=\"javascript:alert(1)\">x</a> <img src=\"/media/abc\" onerror=\"x()\">")
+ for _, bad := range []string{"<script", "javascript:", "onerror"} {
+ if strings.Contains(out, bad) {
+ t.Errorf("output contains %q: %s", bad, out)
+ }
+ }
+ for _, good := range []string{"<h1", "<strong>bold</strong>", `<img src="/media/abc"`} {
+ if !strings.Contains(out, good) {
+ t.Errorf("output missing %q: %s", good, out)
+ }
+ }
+}
+
+func TestRenderGFM(t *testing.T) {
+ out := Render("| a | b |\n|---|---|\n| 1 | 2 |\n\nline one\nline two\n\nhttps://example.org")
+ for _, want := range []string{"<table>", "<br>", `<a href="https://example.org"`} {
+ if !strings.Contains(out, want) {
+ t.Errorf("output missing %q: %s", want, out)
+ }
+ }
+}