review-bot/review/docmap_test.go

package review

import (
	"context"
	"errors"
	"os"
	"path/filepath"
	"strings"
	"testing"
)

// fakeDocFetcher is a mock DocFetcher for tests.
type fakeDocFetcher struct {
	files map[string]string // path -> content
	dirs  map[string]map[string]string // dir path -> (file path -> content)
}

func (f *fakeDocFetcher) GetFileContent(_ context.Context, _, _, path string) (string, error) {
	if content, ok := f.files[path]; ok {
		return content, nil
	}
	return "", errors.New("file not found: " + path)
}

func (f *fakeDocFetcher) GetAllFilesInPath(_ context.Context, _, _, path string) (map[string]string, error) {
	if files, ok := f.dirs[path]; ok {
		return files, nil
	}
	// Return empty (not an error) for unknown directories.
	return nil, nil
}

// ============================================================
// ParseDocMapConfig
// ============================================================

func TestParseDocMapConfig_Valid(t *testing.T) {
	yaml := `
mappings:
  - paths:
      - "lib/foo/**"
    docs:
      - docs/foo.md
  - paths:
      - "lib/bar/**"
      - "lib/baz.go"
    docs:
      - docs/bar.md
      - docs/shared/
`
	f := writeTempYAML(t, yaml)
	cfg, err := ParseDocMapConfig(f)
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if len(cfg.Mappings) != 2 {
		t.Fatalf("expected 2 mappings, got %d", len(cfg.Mappings))
	}
	if cfg.Mappings[0].Paths[0] != "lib/foo/**" {
		t.Errorf("unexpected path: %q", cfg.Mappings[0].Paths[0])
	}
	if cfg.Mappings[1].Docs[1] != "docs/shared/" {
		t.Errorf("unexpected doc: %q", cfg.Mappings[1].Docs[1])
	}
}

func TestParseDocMapConfig_InvalidYAML(t *testing.T) {
	f := writeTempYAML(t, "mappings: [{{invalid")
	_, err := ParseDocMapConfig(f)
	if err == nil {
		t.Fatal("expected error for invalid YAML, got nil")
	}
}

func TestParseDocMapConfig_EmptyMappings(t *testing.T) {
	f := writeTempYAML(t, "mappings: []\n")
	cfg, err := ParseDocMapConfig(f)
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if len(cfg.Mappings) != 0 {
		t.Errorf("expected 0 mappings, got %d", len(cfg.Mappings))
	}
}

func TestParseDocMapConfig_UnknownKeys(t *testing.T) {
	// Unknown keys should produce a warning but not fail.
	yaml := `
mappings:
  - paths: ["lib/foo/**"]
    docs: ["docs/foo.md"]
extra_key: ignored
`
	f := writeTempYAML(t, yaml)
	// Should succeed (lenient parsing).
	cfg, err := ParseDocMapConfig(f)
	if err != nil {
		t.Fatalf("unexpected error for unknown keys: %v", err)
	}
	if len(cfg.Mappings) != 1 {
		t.Errorf("expected 1 mapping, got %d", len(cfg.Mappings))
	}
}

func TestParseDocMapConfig_FileNotFound(t *testing.T) {
	_, err := ParseDocMapConfig("/nonexistent/path/doc-map.yml")
	if err == nil {
		t.Fatal("expected error for missing file, got nil")
	}
}

// ============================================================
// MatchDocs
// ============================================================

func TestMatchDocs_NoMatch(t *testing.T) {
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{"lib/foo/**"}, Docs: []string{"docs/foo.md"}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/bar/baz.go"})
	if len(got) != 0 {
		t.Errorf("expected no matches, got %v", got)
	}
}

func TestMatchDocs_SingleMatch(t *testing.T) {
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{"lib/foo/**"}, Docs: []string{"docs/foo.md"}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/foo/bar.go"})
	if len(got) != 1 || got[0] != "docs/foo.md" {
		t.Errorf("expected [docs/foo.md], got %v", got)
	}
}

func TestMatchDocs_MultipleMatchesDeduplicated(t *testing.T) {
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{"lib/foo/**"}, Docs: []string{"docs/shared.md", "docs/foo.md"}},
			{Paths: []string{"lib/bar/**"}, Docs: []string{"docs/shared.md", "docs/bar.md"}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/foo/a.go", "lib/bar/b.go"})
	// Both match; docs/shared.md should appear only once.
	wantSet := map[string]bool{
		"docs/shared.md": true,
		"docs/foo.md":    true,
		"docs/bar.md":    true,
	}
	if len(got) != 3 {
		t.Errorf("expected 3 docs, got %d: %v", len(got), got)
	}
	for _, d := range got {
		if !wantSet[d] {
			t.Errorf("unexpected doc: %q", d)
		}
	}
}

func TestMatchDocs_EmptyPaths(t *testing.T) {
	// Mapping with empty paths list should not match anything.
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{}, Docs: []string{"docs/foo.md"}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/foo/bar.go"})
	if len(got) != 0 {
		t.Errorf("expected no matches for empty paths, got %v", got)
	}
}

func TestMatchDocs_EmptyDocs(t *testing.T) {
	// Mapping with empty docs list should produce nothing.
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{"lib/foo/**"}, Docs: []string{}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/foo/bar.go"})
	if len(got) != 0 {
		t.Errorf("expected no docs for empty docs list, got %v", got)
	}
}

func TestMatchDocs_ExactMatch(t *testing.T) {
	cfg := &DocMapConfig{
		Mappings: []DocMapping{
			{Paths: []string{"lib/baz.go"}, Docs: []string{"docs/baz.md"}},
		},
	}
	got := MatchDocs(cfg, []string{"lib/baz.go"})
	if len(got) != 1 || got[0] != "docs/baz.md" {
		t.Errorf("expected [docs/baz.md], got %v", got)
	}
}

// ============================================================
// globMatch
// ============================================================

func TestGlobMatch(t *testing.T) {
	tests := []struct {
		name    string
		pattern string
		path    string
		want    bool
	}{
		{"exact match", "lib/foo/bar.go", "lib/foo/bar.go", true},
		{"exact no match", "lib/foo/bar.go", "lib/foo/baz.go", false},
		{"star wildcard", "lib/foo/*.go", "lib/foo/bar.go", true},
		{"star no match cross-dir", "lib/foo/*.go", "lib/foo/sub/bar.go", false},
		{"trailing doublestar", "lib/foo/**", "lib/foo/bar.go", true},
		{"trailing doublestar nested", "lib/foo/**", "lib/foo/sub/deep/bar.go", true},
		// Note: trailing ** matches the parent path too; PR file lists contain file paths
		// (not directories), so this corner case does not arise in practice.
		{"trailing doublestar matches parent", "lib/foo/**", "lib/foo", true},
		{"doublestar in middle", "lib/**/bar.go", "lib/foo/sub/bar.go", true},
		{"doublestar in middle no match", "lib/**/bar.go", "lib/foo/sub/baz.go", false},
		{"leading doublestar", "**/bar.go", "lib/foo/bar.go", true},
		{"leading doublestar top-level", "**/bar.go", "bar.go", true},
		{"question mark", "lib/foo/ba?.go", "lib/foo/bar.go", true},
		{"question mark no match", "lib/foo/ba?.go", "lib/foo/ba.go", false},
		{"star matches none in segment", "lib/*/bar.go", "lib/bar.go", false},
		{"star single segment", "lib/*/bar.go", "lib/foo/bar.go", true},
	}

	for _, tc := range tests {
		t.Run(tc.name, func(t *testing.T) {
			got := globMatch(tc.pattern, tc.path)
			if got != tc.want {
				t.Errorf("globMatch(%q, %q) = %v, want %v", tc.pattern, tc.path, got, tc.want)
			}
		})
	}
}

// ============================================================
// LoadMatchingDocs
// ============================================================

func TestLoadMatchingDocs_FileInjection(t *testing.T) {
	fetcher := &fakeDocFetcher{
		files: map[string]string{
			"docs/foo.md": "# Foo Design\n\nThis is the foo doc.",
		},
	}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"docs/foo.md"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if !strings.Contains(content, "# Foo Design") {
		t.Errorf("expected doc content, got: %q", content)
	}
	if !strings.Contains(content, "### docs/foo.md") {
		t.Errorf("expected heading with path, got: %q", content)
	}
}

func TestLoadMatchingDocs_MissingFileSkipped(t *testing.T) {
	fetcher := &fakeDocFetcher{
		files: map[string]string{
			"docs/present.md": "present",
		},
	}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"docs/missing.md", "docs/present.md"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if !strings.Contains(content, "present") {
		t.Errorf("expected present doc content, got: %q", content)
	}
	// Missing file should be skipped, not cause a failure.
}

func TestLoadMatchingDocs_DirectoryExpansion(t *testing.T) {
	fetcher := &fakeDocFetcher{
		dirs: map[string]map[string]string{
			"docs/domain/": {
				"docs/domain/a.md": "# A",
				"docs/domain/b.md": "# B",
				"docs/domain/c.go": "package domain", // should be skipped (not .md)
			},
		},
	}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"docs/domain/"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if !strings.Contains(content, "# A") {
		t.Errorf("expected doc A content, got: %q", content)
	}
	if !strings.Contains(content, "# B") {
		t.Errorf("expected doc B content, got: %q", content)
	}
	if strings.Contains(content, "package domain") {
		t.Errorf("non-.md file should not be injected, got: %q", content)
	}
}

func TestLoadMatchingDocs_DirectoryNoMDFiles(t *testing.T) {
	fetcher := &fakeDocFetcher{
		dirs: map[string]map[string]string{
			"src/": {
				"src/main.go": "package main",
			},
		},
	}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"src/"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if content != "" {
		t.Errorf("expected empty content for dir with no .md files, got: %q", content)
	}
}

func TestLoadMatchingDocs_NoMatchingPaths(t *testing.T) {
	fetcher := &fakeDocFetcher{}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if content != "" {
		t.Errorf("expected empty content for no paths, got: %q", content)
	}
}

func TestLoadMatchingDocs_ContextSizeGuard(t *testing.T) {
	bigContent := strings.Repeat("x", 200)
	fetcher := &fakeDocFetcher{
		files: map[string]string{
			"docs/a.md": bigContent,
			"docs/b.md": bigContent,
			"docs/c.md": bigContent,
		},
	}
	// Limit to 350 bytes — enough for a.md fully and part of b.md.
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"docs/a.md", "docs/b.md", "docs/c.md"}, DocMapOptions{MaxBytes: 350})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if len(content) > 600 {
		t.Errorf("content too large, expected ≤600 bytes total, got %d", len(content))
	}
	if !strings.Contains(content, "truncated") {
		t.Errorf("expected truncation notice, got: %q", content)
	}
}

func TestLoadMatchingDocs_Deduplication(t *testing.T) {
	fetcher := &fakeDocFetcher{
		files: map[string]string{
			"docs/shared.md": "shared content",
		},
	}
	// MatchDocs deduplicates before calling LoadMatchingDocs, but test it with
	// duplicates in input too.
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"docs/shared.md"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected error: %v", err)
	}
	if !strings.Contains(content, "shared content") {
		t.Errorf("expected shared content, got: %q", content)
	}
}

func TestValidateDocPath(t *testing.T) {
	valid := []string{
		"docs/design.md",
		"docs/domain/contexts/risk/risk-controls.md",
		"README.md",
		"a/b/c",
	}
	for _, p := range valid {
		if err := validateDocPath(p); err != nil {
			t.Errorf("expected valid path %q to pass, got error: %v", p, err)
		}
	}

	invalid := []string{
		"/etc/passwd",
		"/docs/design.md",
		"docs/../../../etc/passwd",
		"../sibling-repo/file.md",
		"a/b/../c",
	}
	for _, p := range invalid {
		if err := validateDocPath(p); err == nil {
			t.Errorf("expected path %q to be rejected, but it was accepted", p)
		}
	}
}

func TestLoadMatchingDocs_PathTraversalRejected(t *testing.T) {
	fetcher := &fakeDocFetcher{
		files: map[string]string{
			"../secret.md": "should not be fetched",
		},
	}
	content, err := LoadMatchingDocs(context.Background(), fetcher, "owner", "repo",
		[]string{"../secret.md"}, DocMapOptions{MaxBytes: DefaultDocMapMaxBytes})
	if err != nil {
		t.Fatalf("unexpected hard error: %v", err)
	}
	// Bad path should be skipped (warned), not injected.
	if strings.Contains(content, "should not be fetched") {
		t.Errorf("path traversal doc was injected, expected it to be skipped")
	}
}

// ============================================================
// Helpers
// ============================================================

func writeTempYAML(t *testing.T, content string) string {
	t.Helper()
	f, err := os.CreateTemp(t.TempDir(), "doc-map-*.yml")
	if err != nil {
		t.Fatalf("failed to create temp file: %v", err)
	}
	defer f.Close()
	if _, err := f.WriteString(content); err != nil {
		t.Fatalf("failed to write temp file: %v", err)
	}
	return filepath.Clean(f.Name())
}