package postcard
import (
"html"
"regexp"
"strings"
xhtml "golang.org/x/net/html"
)
// scriptURLText finds a script URL in text: javascript: or vbscript: at a
// scheme boundary, or data: at a scheme boundary directly followed by a
// non-space character (a data URL). A boundary is the start of the text or a
// character that cannot belong to a URL scheme, so "Metadata: Weekly" and
// "Big Data: Summit" do not match.
var scriptURLText = regexp.MustCompile(`(?i)(?:^|[^a-z0-9+.\-])(?:(?:javascript|vbscript):|data:\S)`)
// UnsafeText reports whether a text that a layout prints, such as the site
// name of the built-in layouts, must be kept out of mail. A text is unsafe
// when, HTML-escaped as html/template prints text, it fails the rendered-HTML
// safety checks (for example event-handler-shaped text such as
// "Shop onclick=x"), or when it holds a script URL: javascript: or vbscript:
// at a scheme boundary, or a data: URL. "Big Data: Summit" is safe;
// "javascript: Acme" is not.
func UnsafeText(s string) bool {
return validateHTML(html.EscapeString(s)) != nil || scriptURLText.MatchString(s)
}
// urlAttributes are the attributes that hold one URL. Their decoded value is
// refused when it starts with a script or data scheme.
var urlAttributes = map[string]bool{
"action": true,
"background": true,
"cite": true,
"codebase": true,
"data": true,
"dynsrc": true,
"formaction": true,
"href": true,
"icon": true,
"longdesc": true,
"lowsrc": true,
"manifest": true,
"poster": true,
"profile": true,
"src": true,
"usemap": true,
"xlink:href": true,
// SVG animation targets can set an href.
"from": true,
"to": true,
"by": true,
}
// urlListAttributes hold several URLs (or values that can be URLs). Their
// decoded value is refused when dangerousScheme matches anywhere in it.
var urlListAttributes = map[string]bool{
"srcset": true,
"ping": true,
"archive": true,
"values": true,
}
// hasDangerousScheme reports whether a decoded URL attribute value starts
// with javascript:, vbscript: or data:, after every ASCII whitespace and
// control character is removed and the value is lower-cased, the way
// browsers read it ("java\tscript:" is javascript:).
func hasDangerousScheme(val string) bool {
var b strings.Builder
b.Grow(len(val))
for i := 0; i < len(val); i++ {
c := val[i]
if c <= 0x20 || c == 0x7f {
continue
}
if 'A' <= c && c <= 'Z' {
c += 'a' - 'A'
}
b.WriteByte(c)
}
v := b.String()
return strings.HasPrefix(v, "javascript:") || strings.HasPrefix(v, "vbscript:") || strings.HasPrefix(v, "data:")
}
// unsafeMarkup checks the markup of rendered HTML: event-handler attributes
// and javascript:, vbscript: or data: URLs. It reads doc with x/net/html's
// tokenizer, which lower-cases attribute names and decodes character
// references in attribute values, so href="javascript:x" is read as a
// mail client reads it.
//
// URL schemes are looked for only where a URL can be: single-URL attributes
// (urlAttributes, by prefix), list-valued attributes (urlListAttributes),
// style attributes, the content of meta elements, the text of style
// elements and comments (anywhere in the value). Text content is not a URL,
// so prose such as "metadata:" or "Big Data: Summit" is not checked. Any
// attribute whose name starts with "on" is an event handler, with or
// without whitespace before it.
func unsafeMarkup(doc string) error {
z := xhtml.NewTokenizer(strings.NewReader(doc))
offset := 0
inStyle := false
for {
tt := z.Next()
if tt == xhtml.ErrorToken {
return nil
}
start := offset
offset += len(z.Raw())
loc := []int{start, offset}
t := z.Token()
switch tt {
case xhtml.StartTagToken, xhtml.SelfClosingTagToken:
for _, a := range t.Attr {
key := a.Key
if a.Namespace != "" {
key = a.Namespace + ":" + a.Key
}
switch {
case strings.HasPrefix(key, "on"):
return &unsafeError{msg: "rendered HTML contains event handlers", snippet: snippetAt(doc, loc)}
case urlAttributes[key] && hasDangerousScheme(a.Val),
urlListAttributes[key] && dangerousScheme.MatchString(a.Val),
key == "style" && dangerousScheme.MatchString(a.Val),
t.Data == "meta" && key == "content" && dangerousScheme.MatchString(a.Val):
return &unsafeError{msg: "rendered HTML contains a dangerous URL scheme", snippet: snippetAt(doc, loc)}
}
}
if tt == xhtml.StartTagToken && t.Data == "style" {
inStyle = true
}
case xhtml.EndTagToken:
if t.Data == "style" {
inStyle = false
}
case xhtml.TextToken:
if inStyle && dangerousScheme.MatchString(t.Data) {
return &unsafeError{msg: "rendered HTML contains a dangerous URL scheme", snippet: snippetAt(doc, loc)}
}
case xhtml.CommentToken:
if dangerousScheme.MatchString(t.Data) {
return &unsafeError{msg: "rendered HTML contains a dangerous URL scheme", snippet: snippetAt(doc, loc)}
}
}
}
}