Remove suffix matching from host whitelist #32

Closed
clawbot wants to merge 3 commits from fix/remove-whitelist-suffix-matching into main
4 changed files with 54 additions and 48 deletions

View File

@@ -98,9 +98,7 @@ expiration 1704067200:
**Whitelist patterns:** **Whitelist patterns:**
- **Exact match**: `cdn.example.com` — matches only that host - **Exact match only**: `cdn.example.com` — matches only that host
- **Suffix match**: `.example.com` — matches `cdn.example.com`,
`images.example.com`, and `example.com`
### Configuration ### Configuration

View File

@@ -13,8 +13,7 @@ state_dir: ./data
# Generate with: openssl rand -base64 32 # Generate with: openssl rand -base64 32
signing_key: "CHANGE_ME_generate_with_openssl_rand_base64_32" signing_key: "CHANGE_ME_generate_with_openssl_rand_base64_32"
# Hosts that don't require signatures # Hosts that don't require signatures (exact match only)
# Use "." prefix for wildcard subdomain matching (e.g., ".example.com" matches "cdn.example.com")
whitelist_hosts: whitelist_hosts:
- s3.sneak.cloud - s3.sneak.cloud
- static.sneak.cloud - static.sneak.cloud

View File

@@ -6,22 +6,19 @@ import (
) )
// HostWhitelist implements the Whitelist interface for checking allowed source hosts. // HostWhitelist implements the Whitelist interface for checking allowed source hosts.
// Only exact host matches are supported. Leading dots in patterns are stripped
// (e.g. ".example.com" becomes an exact match for "example.com").
type HostWhitelist struct { type HostWhitelist struct {
// exactHosts contains hosts that must match exactly (e.g., "cdn.example.com") // hosts contains hosts that must match exactly (e.g., "cdn.example.com")
exactHosts map[string]struct{} hosts map[string]struct{}
// suffixHosts contains domain suffixes to match (e.g., ".example.com" matches "cdn.example.com")
suffixHosts []string
} }
// NewHostWhitelist creates a whitelist from a list of host patterns. // NewHostWhitelist creates a whitelist from a list of host patterns.
// Patterns starting with "." are treated as suffix matches. // All patterns are treated as exact matches. Leading dots are stripped
// Examples: // for backwards compatibility (e.g. ".example.com" matches "example.com" only).
// - "cdn.example.com" - exact match only
// - ".example.com" - matches cdn.example.com, images.example.com, etc.
func NewHostWhitelist(patterns []string) *HostWhitelist { func NewHostWhitelist(patterns []string) *HostWhitelist {
w := &HostWhitelist{ w := &HostWhitelist{
exactHosts: make(map[string]struct{}), hosts: make(map[string]struct{}),
suffixHosts: make([]string, 0),
} }
for _, pattern := range patterns { for _, pattern := range patterns {
@@ -30,17 +27,22 @@ func NewHostWhitelist(patterns []string) *HostWhitelist {
continue continue
} }
if strings.HasPrefix(pattern, ".") { // Strip leading dot — suffix matching is not supported.
w.suffixHosts = append(w.suffixHosts, pattern) // ".example.com" is treated as exact match for "example.com".
} else { pattern = strings.TrimPrefix(pattern, ".")
w.exactHosts[pattern] = struct{}{}
if pattern == "" {
continue
} }
w.hosts[pattern] = struct{}{}
} }
return w return w
} }
// IsWhitelisted checks if a URL's host is in the whitelist. // IsWhitelisted checks if a URL's host is in the whitelist.
// Only exact host matches are supported.
func (w *HostWhitelist) IsWhitelisted(u *url.URL) bool { func (w *HostWhitelist) IsWhitelisted(u *url.URL) bool {
if u == nil { if u == nil {
return false return false
@@ -51,32 +53,17 @@ func (w *HostWhitelist) IsWhitelisted(u *url.URL) bool {
return false return false
} }
// Check exact match _, ok := w.hosts[host]
if _, ok := w.exactHosts[host]; ok {
return true
}
// Check suffix match return ok
for _, suffix := range w.suffixHosts {
if strings.HasSuffix(host, suffix) {
return true
}
// Also match if host equals the suffix without the leading dot
// e.g., pattern ".example.com" should match "example.com"
if host == strings.TrimPrefix(suffix, ".") {
return true
}
}
return false
} }
// IsEmpty returns true if the whitelist has no entries. // IsEmpty returns true if the whitelist has no entries.
func (w *HostWhitelist) IsEmpty() bool { func (w *HostWhitelist) IsEmpty() bool {
return len(w.exactHosts) == 0 && len(w.suffixHosts) == 0 return len(w.hosts) == 0
} }
// Count returns the total number of whitelist entries. // Count returns the total number of whitelist entries.
func (w *HostWhitelist) Count() int { func (w *HostWhitelist) Count() int {
return len(w.exactHosts) + len(w.suffixHosts) return len(w.hosts)
} }

View File

@@ -31,41 +31,47 @@ func TestHostWhitelist_IsWhitelisted(t *testing.T) {
want: false, want: false,
}, },
{ {
name: "suffix match", name: "dot prefix does not enable suffix matching",
patterns: []string{".example.com"}, patterns: []string{".example.com"},
testURL: "https://cdn.example.com/image.jpg", testURL: "https://cdn.example.com/image.jpg",
want: true, want: false,
}, },
{ {
name: "suffix match deep subdomain", name: "dot prefix does not match deep subdomain",
patterns: []string{".example.com"}, patterns: []string{".example.com"},
testURL: "https://cdn.images.example.com/image.jpg", testURL: "https://cdn.images.example.com/image.jpg",
want: true, want: false,
}, },
{ {
name: "suffix match apex domain", name: "dot prefix stripped matches apex domain exactly",
patterns: []string{".example.com"}, patterns: []string{".example.com"},
testURL: "https://example.com/image.jpg", testURL: "https://example.com/image.jpg",
want: true, want: true,
}, },
{ {
name: "suffix match not found", name: "dot prefix does not match unrelated domain",
patterns: []string{".example.com"}, patterns: []string{".example.com"},
testURL: "https://notexample.com/image.jpg", testURL: "https://notexample.com/image.jpg",
want: false, want: false,
}, },
{ {
name: "suffix match partial not allowed", name: "dot prefix does not match partial domain",
patterns: []string{".example.com"}, patterns: []string{".example.com"},
testURL: "https://fakeexample.com/image.jpg", testURL: "https://fakeexample.com/image.jpg",
want: false, want: false,
}, },
{ {
name: "multiple patterns", name: "multiple patterns exact only",
patterns: []string{"cdn.example.com", ".images.org", "static.test.net"}, patterns: []string{"cdn.example.com", "photos.images.org", "static.test.net"},
testURL: "https://photos.images.org/image.jpg", testURL: "https://photos.images.org/image.jpg",
want: true, want: true,
}, },
{
name: "multiple patterns no suffix match",
patterns: []string{"cdn.example.com", ".images.org", "static.test.net"},
testURL: "https://photos.images.org/image.jpg",
want: false,
},
{ {
name: "empty whitelist", name: "empty whitelist",
patterns: []string{}, patterns: []string{},
@@ -90,6 +96,12 @@ func TestHostWhitelist_IsWhitelisted(t *testing.T) {
testURL: "https://cdn.example.com/image.jpg", testURL: "https://cdn.example.com/image.jpg",
want: true, want: true,
}, },
{
name: "whitespace dot prefix stripped matches exactly",
patterns: []string{" .other.com "},
testURL: "https://other.com/image.jpg",
want: true,
},
} }
for _, tt := range tests { for _, tt := range tests {
@@ -139,6 +151,11 @@ func TestHostWhitelist_IsEmpty(t *testing.T) {
patterns: []string{"example.com"}, patterns: []string{"example.com"},
want: false, want: false,
}, },
{
name: "dot prefix entry still counts",
patterns: []string{".example.com"},
want: false,
},
} }
for _, tt := range tests { for _, tt := range tests {
@@ -168,7 +185,7 @@ func TestHostWhitelist_Count(t *testing.T) {
want: 3, want: 3,
}, },
{ {
name: "suffix hosts only", name: "dot prefix hosts treated as exact",
patterns: []string{".a.com", ".b.com"}, patterns: []string{".a.com", ".b.com"},
want: 2, want: 2,
}, },
@@ -177,6 +194,11 @@ func TestHostWhitelist_Count(t *testing.T) {
patterns: []string{"exact.com", ".suffix.com"}, patterns: []string{"exact.com", ".suffix.com"},
want: 2, want: 2,
}, },
{
name: "dot prefix deduplicates with exact",
patterns: []string{"example.com", ".example.com"},
want: 1,
},
} }
for _, tt := range tests { for _, tt := range tests {