feat(13-01): implement NpmSource and PyPISource with httptest tests
- NpmSource searches npm registry JSON API for provider keywords - PyPISource scrapes pypi.org search HTML for project links - Both credentialless, rate-limited at 1 req/2s, burst 2 - httptest-based tests verify Sweep, ctx cancellation, Name/Rate/Burst
This commit is contained in:
salvacybersec
114
pkg/recon/sources/npm.go
Normal file
114
pkg/recon/sources/npm.go
Normal file
@@ -0,0 +1,114 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"time"
|
||||
|
||||
"golang.org/x/time/rate"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
// NpmSource searches the npm registry for packages matching provider keywords.
|
||||
// No credentials required. Emits findings tagged SourceType=recon:npm.
|
||||
type NpmSource struct {
|
||||
BaseURL string
|
||||
Registry *providers.Registry
|
||||
Limiters *recon.LimiterRegistry
|
||||
Client *Client
|
||||
}
|
||||
|
||||
var _ recon.ReconSource = (*NpmSource)(nil)
|
||||
|
||||
func (s *NpmSource) Name() string { return "npm" }
|
||||
func (s *NpmSource) RateLimit() rate.Limit { return rate.Every(2 * time.Second) }
|
||||
func (s *NpmSource) Burst() int { return 2 }
|
||||
func (s *NpmSource) RespectsRobots() bool { return false }
|
||||
func (s *NpmSource) Enabled(_ recon.Config) bool { return true }
|
||||
|
||||
// npm search JSON response structs.
|
||||
type npmSearchResponse struct {
|
||||
Objects []npmObject `json:"objects"`
|
||||
}
|
||||
|
||||
type npmObject struct {
|
||||
Package npmPackage `json:"package"`
|
||||
}
|
||||
|
||||
type npmPackage struct {
|
||||
Name string `json:"name"`
|
||||
Links npmLinks `json:"links"`
|
||||
}
|
||||
|
||||
type npmLinks struct {
|
||||
Npm string `json:"npm"`
|
||||
}
|
||||
|
||||
func (s *NpmSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
|
||||
base := s.BaseURL
|
||||
if base == "" {
|
||||
base = "https://registry.npmjs.org"
|
||||
}
|
||||
client := s.Client
|
||||
if client == nil {
|
||||
client = NewClient()
|
||||
}
|
||||
|
||||
queries := BuildQueries(s.Registry, "npm")
|
||||
if len(queries) == 0 {
|
||||
return nil
|
||||
}
|
||||
|
||||
for _, q := range queries {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
if s.Limiters != nil {
|
||||
if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
searchURL := fmt.Sprintf("%s/-/v1/search?text=%s&size=20", base, url.QueryEscape(q))
|
||||
req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
|
||||
if err != nil {
|
||||
return fmt.Errorf("npm: build req: %w", err)
|
||||
}
|
||||
|
||||
resp, err := client.Do(ctx, req)
|
||||
if err != nil {
|
||||
return fmt.Errorf("npm: fetch: %w", err)
|
||||
}
|
||||
|
||||
var result npmSearchResponse
|
||||
if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
|
||||
_ = resp.Body.Close()
|
||||
return fmt.Errorf("npm: decode json: %w", err)
|
||||
}
|
||||
_ = resp.Body.Close()
|
||||
|
||||
for _, obj := range result.Objects {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
source := obj.Package.Links.Npm
|
||||
if source == "" {
|
||||
source = fmt.Sprintf("https://www.npmjs.com/package/%s", obj.Package.Name)
|
||||
}
|
||||
out <- recon.Finding{
|
||||
ProviderName: "",
|
||||
Source: source,
|
||||
SourceType: "recon:npm",
|
||||
Confidence: "low",
|
||||
DetectedAt: time.Now(),
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
141
pkg/recon/sources/npm_test.go
Normal file
141
pkg/recon/sources/npm_test.go
Normal file
@@ -0,0 +1,141 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
func npmTestRegistry() *providers.Registry {
|
||||
return providers.NewRegistryFromProviders([]providers.Provider{
|
||||
{Name: "openai", Keywords: []string{"sk-proj-"}},
|
||||
})
|
||||
}
|
||||
|
||||
const npmFixtureJSON = `{
|
||||
"objects": [
|
||||
{
|
||||
"package": {
|
||||
"name": "openai-key-checker",
|
||||
"links": {"npm": "https://www.npmjs.com/package/openai-key-checker"}
|
||||
}
|
||||
},
|
||||
{
|
||||
"package": {
|
||||
"name": "sk-proj-util",
|
||||
"links": {"npm": ""}
|
||||
}
|
||||
}
|
||||
]
|
||||
}`
|
||||
|
||||
func newNpmTestSource(srvURL string) *NpmSource {
|
||||
return &NpmSource{
|
||||
BaseURL: srvURL,
|
||||
Registry: npmTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
}
|
||||
|
||||
func TestNpm_Sweep_ExtractsFindings(t *testing.T) {
|
||||
var hits int
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
if r.URL.Path != "/-/v1/search" {
|
||||
t.Errorf("unexpected path: %s", r.URL.Path)
|
||||
}
|
||||
if r.URL.Query().Get("text") == "" {
|
||||
t.Errorf("missing text param")
|
||||
}
|
||||
hits++
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
_, _ = w.Write([]byte(npmFixtureJSON))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newNpmTestSource(srv.URL)
|
||||
out := make(chan recon.Finding, 16)
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 10*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)
|
||||
}
|
||||
if len(findings) != 2 {
|
||||
t.Fatalf("expected 2 findings, got %d", len(findings))
|
||||
}
|
||||
|
||||
got := map[string]bool{}
|
||||
for _, f := range findings {
|
||||
got[f.Source] = true
|
||||
if f.SourceType != "recon:npm" {
|
||||
t.Errorf("unexpected SourceType: %s", f.SourceType)
|
||||
}
|
||||
if f.Confidence != "low" {
|
||||
t.Errorf("unexpected Confidence: %s", f.Confidence)
|
||||
}
|
||||
}
|
||||
if !got["https://www.npmjs.com/package/openai-key-checker"] {
|
||||
t.Error("missing finding with npm link")
|
||||
}
|
||||
// Second package has empty links.npm — should get constructed URL.
|
||||
if !got["https://www.npmjs.com/package/sk-proj-util"] {
|
||||
t.Error("missing finding with constructed URL")
|
||||
}
|
||||
if hits == 0 {
|
||||
t.Fatal("server was never hit")
|
||||
}
|
||||
}
|
||||
|
||||
func TestNpm_EnabledAlwaysTrue(t *testing.T) {
|
||||
s := &NpmSource{}
|
||||
if !s.Enabled(recon.Config{}) {
|
||||
t.Fatal("expected Enabled=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestNpm_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(npmFixtureJSON))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newNpmTestSource(srv.URL)
|
||||
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")
|
||||
}
|
||||
}
|
||||
|
||||
func TestNpm_NameAndRate(t *testing.T) {
|
||||
s := &NpmSource{}
|
||||
if s.Name() != "npm" {
|
||||
t.Errorf("unexpected name: %s", s.Name())
|
||||
}
|
||||
if s.Burst() != 2 {
|
||||
t.Errorf("burst: %d", s.Burst())
|
||||
}
|
||||
if s.RespectsRobots() {
|
||||
t.Error("expected RespectsRobots=false")
|
||||
}
|
||||
want := float64(1) / 2
|
||||
got := float64(s.RateLimit())
|
||||
if got < want-0.01 || got > want+0.01 {
|
||||
t.Errorf("rate limit=%v want~%v", got, want)
|
||||
}
|
||||
}
|
||||
102
pkg/recon/sources/pypi.go
Normal file
102
pkg/recon/sources/pypi.go
Normal file
@@ -0,0 +1,102 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"time"
|
||||
|
||||
"golang.org/x/time/rate"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
// PyPISource searches pypi.org for packages matching provider keywords.
|
||||
// Scrapes the HTML search page since PyPI has no public search JSON API.
|
||||
// No credentials required. Emits findings tagged SourceType=recon:pypi.
|
||||
type PyPISource struct {
|
||||
BaseURL string
|
||||
Registry *providers.Registry
|
||||
Limiters *recon.LimiterRegistry
|
||||
Client *Client
|
||||
}
|
||||
|
||||
var _ recon.ReconSource = (*PyPISource)(nil)
|
||||
|
||||
// pypiProjectRE matches /project/{name}/ hrefs in search results.
|
||||
var pypiProjectRE = regexp.MustCompile(`^/project/[^/]+/?$`)
|
||||
|
||||
func (s *PyPISource) Name() string { return "pypi" }
|
||||
func (s *PyPISource) RateLimit() rate.Limit { return rate.Every(2 * time.Second) }
|
||||
func (s *PyPISource) Burst() int { return 2 }
|
||||
func (s *PyPISource) RespectsRobots() bool { return false }
|
||||
func (s *PyPISource) Enabled(_ recon.Config) bool { return true }
|
||||
|
||||
func (s *PyPISource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
|
||||
base := s.BaseURL
|
||||
if base == "" {
|
||||
base = "https://pypi.org"
|
||||
}
|
||||
client := s.Client
|
||||
if client == nil {
|
||||
client = NewClient()
|
||||
}
|
||||
|
||||
queries := BuildQueries(s.Registry, "pypi")
|
||||
if len(queries) == 0 {
|
||||
return nil
|
||||
}
|
||||
|
||||
for _, q := range queries {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
if s.Limiters != nil {
|
||||
if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
searchURL := fmt.Sprintf("%s/search/?q=%s", base, url.QueryEscape(q))
|
||||
req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
|
||||
if err != nil {
|
||||
return fmt.Errorf("pypi: build req: %w", err)
|
||||
}
|
||||
|
||||
resp, err := client.Do(ctx, req)
|
||||
if err != nil {
|
||||
return fmt.Errorf("pypi: fetch: %w", err)
|
||||
}
|
||||
|
||||
hrefs, err := extractPyPIProjectLinks(resp.Body)
|
||||
_ = resp.Body.Close()
|
||||
if err != nil {
|
||||
return fmt.Errorf("pypi: parse html: %w", err)
|
||||
}
|
||||
|
||||
for _, href := range hrefs {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
absURL := base + href
|
||||
out <- recon.Finding{
|
||||
ProviderName: "",
|
||||
Source: absURL,
|
||||
SourceType: "recon:pypi",
|
||||
Confidence: "low",
|
||||
DetectedAt: time.Now(),
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// extractPyPIProjectLinks extracts unique /project/{name}/ hrefs from HTML.
|
||||
func extractPyPIProjectLinks(body io.Reader) ([]string, error) {
|
||||
return extractAnchorHrefs(body, pypiProjectRE)
|
||||
}
|
||||
133
pkg/recon/sources/pypi_test.go
Normal file
133
pkg/recon/sources/pypi_test.go
Normal file
@@ -0,0 +1,133 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
func pypiTestRegistry() *providers.Registry {
|
||||
return providers.NewRegistryFromProviders([]providers.Provider{
|
||||
{Name: "openai", Keywords: []string{"sk-proj-"}},
|
||||
})
|
||||
}
|
||||
|
||||
const pypiFixtureHTML = `<!doctype html>
|
||||
<html><body>
|
||||
<a href="/project/openai-leaked/">openai-leaked</a>
|
||||
<a href="/project/sk-proj-helper/">sk helper</a>
|
||||
<a href="/simple/">nope</a>
|
||||
<a href="https://external.example.com/project/nope/">external</a>
|
||||
<a href="/project/openai-leaked/">duplicate</a>
|
||||
</body></html>`
|
||||
|
||||
func newPyPITestSource(srvURL string) *PyPISource {
|
||||
return &PyPISource{
|
||||
BaseURL: srvURL,
|
||||
Registry: pypiTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
}
|
||||
|
||||
func TestPyPI_Sweep_ExtractsFindings(t *testing.T) {
|
||||
var hits int
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
if r.URL.Path != "/search/" {
|
||||
t.Errorf("unexpected path: %s", r.URL.Path)
|
||||
}
|
||||
if r.URL.Query().Get("q") == "" {
|
||||
t.Errorf("missing q param")
|
||||
}
|
||||
hits++
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(pypiFixtureHTML))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newPyPITestSource(srv.URL)
|
||||
out := make(chan recon.Finding, 16)
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 10*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)
|
||||
}
|
||||
// 2 unique /project/ links (duplicate is deduped by extractAnchorHrefs)
|
||||
if len(findings) != 2 {
|
||||
t.Fatalf("expected 2 findings, got %d", len(findings))
|
||||
}
|
||||
|
||||
got := map[string]bool{}
|
||||
for _, f := range findings {
|
||||
got[f.Source] = true
|
||||
if f.SourceType != "recon:pypi" {
|
||||
t.Errorf("unexpected SourceType: %s", f.SourceType)
|
||||
}
|
||||
if f.Confidence != "low" {
|
||||
t.Errorf("unexpected Confidence: %s", f.Confidence)
|
||||
}
|
||||
}
|
||||
if !got[srv.URL+"/project/openai-leaked/"] {
|
||||
t.Error("missing openai-leaked finding")
|
||||
}
|
||||
if !got[srv.URL+"/project/sk-proj-helper/"] {
|
||||
t.Error("missing sk-proj-helper finding")
|
||||
}
|
||||
if hits == 0 {
|
||||
t.Fatal("server was never hit")
|
||||
}
|
||||
}
|
||||
|
||||
func TestPyPI_EnabledAlwaysTrue(t *testing.T) {
|
||||
s := &PyPISource{}
|
||||
if !s.Enabled(recon.Config{}) {
|
||||
t.Fatal("expected Enabled=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestPyPI_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(pypiFixtureHTML))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newPyPITestSource(srv.URL)
|
||||
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")
|
||||
}
|
||||
}
|
||||
|
||||
func TestPyPI_NameAndRate(t *testing.T) {
|
||||
s := &PyPISource{}
|
||||
if s.Name() != "pypi" {
|
||||
t.Errorf("unexpected name: %s", s.Name())
|
||||
}
|
||||
if s.Burst() != 2 {
|
||||
t.Errorf("burst: %d", s.Burst())
|
||||
}
|
||||
if s.RespectsRobots() {
|
||||
t.Error("expected RespectsRobots=false")
|
||||
}
|
||||
want := float64(1) / 2
|
||||
got := float64(s.RateLimit())
|
||||
if got < want-0.01 || got > want+0.01 {
|
||||
t.Errorf("rate limit=%v want~%v", got, want)
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user