feat(11-02): add PasteSitesSource multi-paste aggregator
- Aggregates dpaste, paste.ee, rentry, hastebin into single source - Follows SandboxesSource multi-platform pattern with per-platform error isolation - Two-phase search+raw-fetch with keyword matching against provider registry
This commit is contained in:
salvacybersec
190
pkg/recon/sources/pastesites_test.go
Normal file
190
pkg/recon/sources/pastesites_test.go
Normal file
@@ -0,0 +1,190 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
func pasteSitesTestRegistry() *providers.Registry {
|
||||
return providers.NewRegistryFromProviders([]providers.Provider{
|
||||
{Name: "openai", Keywords: []string{"sk-proj-"}},
|
||||
})
|
||||
}
|
||||
|
||||
// Fixture HTML for each sub-platform search result page.
|
||||
const dpasteSearchHTML = `<html><body><a href="/AbcDef12">dpaste hit</a></body></html>`
|
||||
const pasteEeSearchHTML = `<html><body><a href="/p/Xyz789">paste.ee hit</a></body></html>`
|
||||
const rentrySearchHTML = `<html><body><a href="/my-paste">rentry hit</a></body></html>`
|
||||
const hastebinSearchHTML = `<html><body><a href="/abcdef">hastebin hit</a></body></html>`
|
||||
|
||||
// Raw content fixtures -- some match, some don't.
|
||||
const dpasteRaw = `leaked: sk-proj-AAAA1234 oops`
|
||||
const pasteEeRaw = `config sk-proj-BBBBB5678 here`
|
||||
const rentryRaw = `has sk-proj-CCCC9012 inside`
|
||||
const hastebinRaw = `nothing interesting`
|
||||
|
||||
func TestPasteSites_Sweep_ExtractsFindings(t *testing.T) {
|
||||
mux := http.NewServeMux()
|
||||
|
||||
// dpaste routes
|
||||
mux.HandleFunc("/dpaste-search/", func(w http.ResponseWriter, r *http.Request) {
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(dpasteSearchHTML))
|
||||
})
|
||||
mux.HandleFunc("/AbcDef12/raw", func(w http.ResponseWriter, r *http.Request) {
|
||||
_, _ = w.Write([]byte(dpasteRaw))
|
||||
})
|
||||
|
||||
// paste.ee routes
|
||||
mux.HandleFunc("/pasteee-search", func(w http.ResponseWriter, r *http.Request) {
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(pasteEeSearchHTML))
|
||||
})
|
||||
mux.HandleFunc("/r/p/Xyz789", func(w http.ResponseWriter, r *http.Request) {
|
||||
_, _ = w.Write([]byte(pasteEeRaw))
|
||||
})
|
||||
|
||||
// rentry routes
|
||||
mux.HandleFunc("/rentry-search", func(w http.ResponseWriter, r *http.Request) {
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(rentrySearchHTML))
|
||||
})
|
||||
mux.HandleFunc("/my-paste/raw", func(w http.ResponseWriter, r *http.Request) {
|
||||
_, _ = w.Write([]byte(rentryRaw))
|
||||
})
|
||||
|
||||
// hastebin routes
|
||||
mux.HandleFunc("/hastebin-search", func(w http.ResponseWriter, r *http.Request) {
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(hastebinSearchHTML))
|
||||
})
|
||||
mux.HandleFunc("/raw/abcdef", func(w http.ResponseWriter, r *http.Request) {
|
||||
_, _ = w.Write([]byte(hastebinRaw))
|
||||
})
|
||||
|
||||
srv := httptest.NewServer(mux)
|
||||
defer srv.Close()
|
||||
|
||||
// Override platforms to use test server with relative paths.
|
||||
testPlats := []pastePlatform{
|
||||
{
|
||||
Name: "dpaste",
|
||||
SearchPath: srv.URL + "/dpaste-search/?q=%s",
|
||||
ResultLinkRegex: `^/[A-Za-z0-9]+$`,
|
||||
RawPathTemplate: "%s/raw",
|
||||
},
|
||||
{
|
||||
Name: "paste.ee",
|
||||
SearchPath: srv.URL + "/pasteee-search?q=%s",
|
||||
ResultLinkRegex: `^/p/[A-Za-z0-9]+$`,
|
||||
RawPathTemplate: "/r%s",
|
||||
},
|
||||
{
|
||||
Name: "rentry",
|
||||
SearchPath: srv.URL + "/rentry-search?q=%s",
|
||||
ResultLinkRegex: `^/[a-z0-9-]+$`,
|
||||
RawPathTemplate: "%s/raw",
|
||||
},
|
||||
{
|
||||
Name: "hastebin",
|
||||
SearchPath: srv.URL + "/hastebin-search?q=%s",
|
||||
ResultLinkRegex: `^/[a-z]+$`,
|
||||
RawPathTemplate: "/raw%s",
|
||||
},
|
||||
}
|
||||
|
||||
src := &PasteSitesSource{
|
||||
Platforms: testPlats,
|
||||
Registry: pasteSitesTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
|
||||
out := make(chan recon.Finding, 32)
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
|
||||
defer cancel()
|
||||
|
||||
if err := src.Sweep(ctx, "", out); err != nil {
|
||||
t.Fatalf("Sweep err: %v", err)
|
||||
}
|
||||
close(out)
|
||||
|
||||
var findings []recon.Finding
|
||||
for f := range out {
|
||||
findings = append(findings, f)
|
||||
}
|
||||
|
||||
// dpaste, paste.ee, rentry have matching content; hastebin does not.
|
||||
if len(findings) < 3 {
|
||||
t.Fatalf("expected at least 3 findings (dpaste+paste.ee+rentry), got %d", len(findings))
|
||||
}
|
||||
|
||||
platforms := make(map[string]bool)
|
||||
for _, f := range findings {
|
||||
if f.SourceType != "recon:pastesites" {
|
||||
t.Errorf("SourceType=%s, want recon:pastesites", f.SourceType)
|
||||
}
|
||||
// Extract platform from KeyMasked.
|
||||
if len(f.KeyMasked) > len("platform=") {
|
||||
platforms[f.KeyMasked[len("platform="):]] = true
|
||||
}
|
||||
}
|
||||
for _, want := range []string{"dpaste", "paste.ee", "rentry"} {
|
||||
if !platforms[want] {
|
||||
t.Errorf("missing platform %q in findings; got platforms=%v", want, platforms)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestPasteSites_NameAndRate(t *testing.T) {
|
||||
s := &PasteSitesSource{}
|
||||
if s.Name() != "pastesites" {
|
||||
t.Errorf("Name=%s", s.Name())
|
||||
}
|
||||
if s.Burst() != 1 {
|
||||
t.Errorf("Burst=%d", s.Burst())
|
||||
}
|
||||
if !s.RespectsRobots() {
|
||||
t.Error("expected RespectsRobots=true")
|
||||
}
|
||||
if !s.Enabled(recon.Config{}) {
|
||||
t.Error("expected Enabled=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestPasteSites_Sweep_CtxCancelled(t *testing.T) {
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
time.Sleep(500 * time.Millisecond)
|
||||
_, _ = w.Write([]byte(`<html><body></body></html>`))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
testPlats := []pastePlatform{
|
||||
{
|
||||
Name: "test",
|
||||
SearchPath: srv.URL + "/search?q=%s",
|
||||
ResultLinkRegex: `^/[a-z]+$`,
|
||||
RawPathTemplate: "/raw%s",
|
||||
},
|
||||
}
|
||||
|
||||
src := &PasteSitesSource{
|
||||
Platforms: testPlats,
|
||||
Registry: pasteSitesTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
cancel()
|
||||
|
||||
out := make(chan recon.Finding, 4)
|
||||
if err := src.Sweep(ctx, "", out); err == nil {
|
||||
t.Fatal("expected ctx error")
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user