enhancement(ruleset): add support for category-level ruleset expansion with docker/all syntax

python-sdk/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "codepathfinder"
-version = "1.2.0"
+version = "1.2.1"
 description = "Python SDK for code-pathfinder static analysis for modern security teams"
 readme = "README.md"
 requires-python = ">=3.8"

sast-engine/VERSION

@@ -1 +1 @@
-1.2.0
+1.2.1

sast-engine/cmd/scan.go

@@ -561,11 +561,21 @@ func prepareRules(localRulesPath string, rulesetSpecs []string, refresh bool, lo
 			// This is a rule ID (e.g., docker/DOCKER-BP-007)
 			ruleIDSpecs = append(ruleIDSpecs, spec)
 		} else {
-			// This is a bundle (e.g., docker/security)
+			// This is a bundle (e.g., docker/security) or category expansion (e.g., docker/all)
 			bundleSpecs = append(bundleSpecs, spec)
 		}
 	}
 
+	// Expand "category/all" specs to individual bundle specs
+	if len(bundleSpecs) > 0 {
+		manifestLoader := ruleset.NewManifestLoader("https://assets.codepathfinder.dev/rules", getCacheDir())
+		expanded, err := expandBundleSpecs(bundleSpecs, manifestLoader, logger)
+		if err != nil {
+			return "", "", err
+		}
+		bundleSpecs = expanded
+	}
+
 	// Download remote bundles
 	var downloadedPaths []string
 	if len(bundleSpecs) > 0 {
@@ -736,6 +746,46 @@ func copyRules(src, dest, subdir string) error {
 	return nil
 }
 
+// expandBundleSpecs expands "category/all" specs into individual bundle specs.
+// This function is extracted for testability with mock manifest providers.
+func expandBundleSpecs(bundleSpecs []string, manifestProvider ruleset.ManifestProvider, logger *output.Logger) ([]string, error) {
+	expandedBundleSpecs := make([]string, 0, len(bundleSpecs))
+
+	for _, spec := range bundleSpecs {
+		parsed, err := ruleset.ParseSpec(spec)
+		if err != nil {
+			return nil, fmt.Errorf("invalid ruleset spec %s: %w", spec, err)
+		}
+
+		// Check if this is a category expansion (bundle == "*")
+		if parsed.Bundle == "*" {
+			// Load category manifest to get all bundle names
+			manifest, err := manifestProvider.LoadCategoryManifest(parsed.Category)
+			if err != nil {
+				return nil, fmt.Errorf("failed to load manifest for category %s: %w", parsed.Category, err)
+			}
+
+			// Expand to all bundles in category
+			bundleNames := manifest.GetAllBundleNames()
+			if len(bundleNames) == 0 {
+				logger.Warning("Category %s has no bundles", parsed.Category)
+				continue
+			}
+
+			logger.Progress("Expanding %s/all to %d bundles: %v", parsed.Category, len(bundleNames), bundleNames)
+
+			for _, bundleName := range bundleNames {
+				expandedBundleSpecs = append(expandedBundleSpecs, fmt.Sprintf("%s/%s", parsed.Category, bundleName))
+			}
+		} else {
+			// Regular bundle spec, keep as-is
+			expandedBundleSpecs = append(expandedBundleSpecs, spec)
+		}
+	}
+
+	return expandedBundleSpecs, nil
+}
+
 // copyFile copies a single file from src to dest.
 func copyFile(src, dest string) error {
 	sourceFile, err := os.Open(src)

sast-engine/cmd/scan_test.go

@@ -2,13 +2,16 @@ package cmd
 
 import (
 	"bytes"
+	"fmt"
 	"io"
 	"os"
 	"testing"
 
 	"github.com/shivasurya/code-pathfinder/sast-engine/dsl"
 	"github.com/shivasurya/code-pathfinder/sast-engine/graph"
 	"github.com/shivasurya/code-pathfinder/sast-engine/graph/callgraph/core"
+	"github.com/shivasurya/code-pathfinder/sast-engine/output"
+	"github.com/shivasurya/code-pathfinder/sast-engine/ruleset"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -407,3 +410,195 @@ line 7`
 		assert.Equal(t, 0, len(snippet.Lines))
 	})
 }
+
+// mockManifestProvider is a mock implementation of ruleset.ManifestProvider for testing.
+type mockManifestProvider struct {
+	manifests map[string]*ruleset.Manifest
+	errors    map[string]error
+}
+
+func newMockManifestProvider() *mockManifestProvider {
+	return &mockManifestProvider{
+		manifests: make(map[string]*ruleset.Manifest),
+		errors:    make(map[string]error),
+	}
+}
+
+func (m *mockManifestProvider) LoadCategoryManifest(category string) (*ruleset.Manifest, error) {
+	if err, exists := m.errors[category]; exists {
+		return nil, err
+	}
+	if manifest, exists := m.manifests[category]; exists {
+		return manifest, nil
+	}
+	return nil, fmt.Errorf("category not found: %s", category)
+}
+
+func (m *mockManifestProvider) addManifest(category string, bundleNames []string) {
+	manifest := &ruleset.Manifest{
+		Category: category,
+		Bundles:  make(map[string]*ruleset.Bundle),
+	}
+	for _, name := range bundleNames {
+		manifest.Bundles[name] = &ruleset.Bundle{Name: name}
+	}
+	m.manifests[category] = manifest
+}
+
+func (m *mockManifestProvider) addError(category string, err error) {
+	m.errors[category] = err
+}
+
+func TestExpandBundleSpecs(t *testing.T) {
+	t.Run("expands docker/all to multiple bundles", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{"security", "best-practice", "performance"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 3, len(expanded))
+		assert.Contains(t, expanded, "docker/best-practice")
+		assert.Contains(t, expanded, "docker/performance")
+		assert.Contains(t, expanded, "docker/security")
+	})
+
+	t.Run("expands python/all to multiple bundles", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("python", []string{"deserialization", "django", "flask"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"python/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 3, len(expanded))
+		assert.Contains(t, expanded, "python/deserialization")
+		assert.Contains(t, expanded, "python/django")
+		assert.Contains(t, expanded, "python/flask")
+	})
+
+	t.Run("keeps regular bundle specs unchanged", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/security", "python/django"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 2, len(expanded))
+		assert.Equal(t, "docker/security", expanded[0])
+		assert.Equal(t, "python/django", expanded[1])
+	})
+
+	t.Run("mixes category expansion with regular specs", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{"security", "best-practice"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/all", "python/django"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 3, len(expanded))
+		assert.Contains(t, expanded, "docker/best-practice")
+		assert.Contains(t, expanded, "docker/security")
+		assert.Contains(t, expanded, "python/django")
+	})
+
+	t.Run("handles category with single bundle", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{"security"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 1, len(expanded))
+		assert.Equal(t, "docker/security", expanded[0])
+	})
+
+	t.Run("handles category with no bundles", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{}) // Empty bundles
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 0, len(expanded))
+	})
+
+	t.Run("returns error when category manifest fails to load", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addError("nonexistent", fmt.Errorf("HTTP 404: not found"))
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"nonexistent/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.Error(t, err)
+		assert.Nil(t, expanded)
+		assert.Contains(t, err.Error(), "failed to load manifest for category nonexistent")
+	})
+
+	t.Run("returns error for invalid spec format", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"invalid-spec-no-slash"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.Error(t, err)
+		assert.Nil(t, expanded)
+		assert.Contains(t, err.Error(), "invalid ruleset spec")
+	})
+
+	t.Run("handles multiple category expansions", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{"security", "best-practice"})
+		mock.addManifest("python", []string{"django", "flask"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"docker/all", "python/all"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 4, len(expanded))
+		assert.Contains(t, expanded, "docker/best-practice")
+		assert.Contains(t, expanded, "docker/security")
+		assert.Contains(t, expanded, "python/django")
+		assert.Contains(t, expanded, "python/flask")
+	})
+
+	t.Run("handles empty input specs", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 0, len(expanded))
+	})
+
+	t.Run("preserves order for mixed specs", func(t *testing.T) {
+		mock := newMockManifestProvider()
+		mock.addManifest("docker", []string{"security"})
+		logger := output.NewLogger(output.VerbosityDefault)
+
+		specs := []string{"python/django", "docker/all", "java/security"}
+		expanded, err := expandBundleSpecs(specs, mock, logger)
+
+		require.NoError(t, err)
+		assert.Equal(t, 3, len(expanded))
+		// Order should be: python/django, docker/security (expanded), java/security
+		assert.Equal(t, "python/django", expanded[0])
+		assert.Equal(t, "docker/security", expanded[1])
+		assert.Equal(t, "java/security", expanded[2])
+	})
+}

sast-engine/ruleset/manifest.go

@@ -6,6 +6,7 @@ import (
 	"fmt"
 	"io"
 	"net/http"
+	"sort"
 	"time"
 )
 
@@ -68,3 +69,15 @@ func (m *Manifest) GetBundle(bundleName string) (*Bundle, error) {
 	}
 	return bundle, nil
 }
+
+// GetAllBundleNames returns a sorted list of all bundle names in this category.
+// Used for expanding "category/all" specs to all available bundles.
+func (m *Manifest) GetAllBundleNames() []string {
+	names := make([]string, 0, len(m.Bundles))
+	for name := range m.Bundles {
+		names = append(names, name)
+	}
+	// Sort for consistent ordering across runs
+	sort.Strings(names)
+	return names
+}

sast-engine/ruleset/manifest_test.go

@@ -111,3 +111,57 @@ func TestManifestGetBundle(t *testing.T) {
 		t.Errorf("expected error for non-existent bundle, got nil")
 	}
 }
+
+func TestManifestGetAllBundleNames(t *testing.T) {
+	tests := []struct {
+		name     string
+		manifest *Manifest
+		want     []string
+	}{
+		{
+			name: "multiple bundles",
+			manifest: &Manifest{
+				Bundles: map[string]*Bundle{
+					"security":      {Name: "Security Rules"},
+					"best-practice": {Name: "Best Practices"},
+					"performance":   {Name: "Performance Rules"},
+				},
+			},
+			want: []string{"best-practice", "performance", "security"},
+		},
+		{
+			name: "single bundle",
+			manifest: &Manifest{
+				Bundles: map[string]*Bundle{
+					"security": {Name: "Security Rules"},
+				},
+			},
+			want: []string{"security"},
+		},
+		{
+			name: "empty bundles",
+			manifest: &Manifest{
+				Bundles: map[string]*Bundle{},
+			},
+			want: []string{},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := tt.manifest.GetAllBundleNames()
+
+			if len(got) != len(tt.want) {
+				t.Errorf("expected %d bundles, got %d", len(tt.want), len(got))
+				return
+			}
+
+			// Check each expected bundle name
+			for i, name := range tt.want {
+				if got[i] != name {
+					t.Errorf("expected bundle[%d] = %s, got %s", i, name, got[i])
+				}
+			}
+		})
+	}
+}