forgejo/modules/markup/sanitizer.go
guillep2k 15809d81f7 Rewrite reference processing code in preparation for opening/closing from comment references (#8261)
* Add a markdown stripper for mentions and xrefs

* Improve comments

* Small code simplification

* Move reference code to modules/references

* Fix typo

* Make MarkdownStripper return [][]byte

* Implement preliminary keywords parsing

* Add FIXME comment

* Fix comment

* make fmt

* Fix permissions check

* Fix text assumptions

* Fix imports

* Fix lint, fmt

* Fix unused import

* Add missing export comment

* Bypass revive on implemented interface

* Move mdstripper into its own package

* Support alphanumeric patterns

* Refactor FindAllMentions

* Move mentions test to references

* Parse mentions from reference package

* Refactor code to implement renderizable references

* Fix typo

* Move patterns and tests to the references package

* Fix nil reference

* Preliminary rendering attempt of closing keywords

* Normalize names, comments, general tidy-up

* Add CSS style for action keywords

* Fix permission for admin and owner

* Fix golangci-lint

* Fix golangci-lint
2019-10-13 23:29:10 +01:00

62 lines
1.9 KiB
Go

// Copyright 2017 The Gitea Authors. All rights reserved.
// Copyright 2017 The Gogs Authors. All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package markup
import (
"regexp"
"sync"
"code.gitea.io/gitea/modules/setting"
"github.com/microcosm-cc/bluemonday"
)
// Sanitizer is a protection wrapper of *bluemonday.Policy which does not allow
// any modification to the underlying policies once it's been created.
type Sanitizer struct {
policy *bluemonday.Policy
init sync.Once
}
var sanitizer = &Sanitizer{}
// NewSanitizer initializes sanitizer with allowed attributes based on settings.
// Multiple calls to this function will only create one instance of Sanitizer during
// entire application lifecycle.
func NewSanitizer() {
sanitizer.init.Do(func() {
sanitizer.policy = bluemonday.UGCPolicy()
// We only want to allow HighlightJS specific classes for code blocks
sanitizer.policy.AllowAttrs("class").Matching(regexp.MustCompile(`^language-\w+$`)).OnElements("code")
// Checkboxes
sanitizer.policy.AllowAttrs("type").Matching(regexp.MustCompile(`^checkbox$`)).OnElements("input")
sanitizer.policy.AllowAttrs("checked", "disabled").OnElements("input")
// Custom URL-Schemes
sanitizer.policy.AllowURLSchemes(setting.Markdown.CustomURLSchemes...)
// Allow keyword markup
sanitizer.policy.AllowAttrs("class").Matching(regexp.MustCompile(`^` + keywordClass + `$`)).OnElements("span")
})
}
// Sanitize takes a string that contains a HTML fragment or document and applies policy whitelist.
func Sanitize(s string) string {
NewSanitizer()
return sanitizer.policy.Sanitize(s)
}
// SanitizeBytes takes a []byte slice that contains a HTML fragment or document and applies policy whitelist.
func SanitizeBytes(b []byte) []byte {
if len(b) == 0 {
// nothing to sanitize
return b
}
NewSanitizer()
return sanitizer.policy.SanitizeBytes(b)
}