sanitizer.go 1.7 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455
  1. // Copyright 2017 The Gogs Authors. All rights reserved.
  2. // Use of this source code is governed by a MIT-style
  3. // license that can be found in the LICENSE file.
  4. package markup
  5. import (
  6. "regexp"
  7. "sync"
  8. "github.com/microcosm-cc/bluemonday"
  9. "github.com/gogs/gogs/pkg/setting"
  10. )
  11. // Sanitizer is a protection wrapper of *bluemonday.Policy which does not allow
  12. // any modification to the underlying policies once it's been created.
  13. type Sanitizer struct {
  14. policy *bluemonday.Policy
  15. init sync.Once
  16. }
  17. var sanitizer = &Sanitizer{
  18. policy: bluemonday.UGCPolicy(),
  19. }
  20. // NewSanitizer initializes sanitizer with allowed attributes based on settings.
  21. // Multiple calls to this function will only create one instance of Sanitizer during
  22. // entire application lifecycle.
  23. func NewSanitizer() {
  24. sanitizer.init.Do(func() {
  25. // We only want to allow HighlightJS specific classes for code blocks
  26. sanitizer.policy.AllowAttrs("class").Matching(regexp.MustCompile(`^language-\w+$`)).OnElements("code")
  27. // Checkboxes
  28. sanitizer.policy.AllowAttrs("type").Matching(regexp.MustCompile(`^checkbox$`)).OnElements("input")
  29. sanitizer.policy.AllowAttrs("checked", "disabled").OnElements("input")
  30. // Data URLs
  31. sanitizer.policy.AllowURLSchemes("data")
  32. // Custom URL-Schemes
  33. sanitizer.policy.AllowURLSchemes(setting.Markdown.CustomURLSchemes...)
  34. })
  35. }
  36. // Sanitize takes a string that contains a HTML fragment or document and applies policy whitelist.
  37. func Sanitize(s string) string {
  38. return sanitizer.policy.Sanitize(s)
  39. }
  40. // SanitizeBytes takes a []byte slice that contains a HTML fragment or document and applies policy whitelist.
  41. func SanitizeBytes(b []byte) []byte {
  42. return sanitizer.policy.SanitizeBytes(b)
  43. }