Merge pull request #7 from eSlider/fix/oo-applications-discover

fix(oo): skip junk dirs in applications Discover

README.md

@@ -751,7 +751,11 @@ oo applications sync --path ./applications/2026 --verbose
 oo applications sync --path ./applications/2026 --apply --verbose
 ```
 
-Each `README.md` is parsed for company, role, email, phone, LinkedIn, etc.
+Each dossier `README.md` is parsed for company, role, email, phone, LinkedIn, etc.
+Discovery walks `--path` and **skips** junk trees (`node_modules`, `tools`, `.venv`,
+`pdfs`, …). Only folders that look like CV application slugs (`source-NNN-…`, or
+long hyphenated dossiers) are synced — not npm package READMEs.
+
 The sync creates or finds contacts, opens a deal, adds members, and appends a
 history note. Re-running is safe: duplicate members and duplicate deal titles
 are skipped when already present.

cmd/oo/applications/applications.go

@@ -24,6 +24,33 @@ import (
 
 var docExt = map[string]bool{".pdf": true, ".docx": true, ".xlsx": true, ".doc": true, ".xls": true}
 
+// Directories that are never application dossiers (WalkDir SkipDir).
+var skipDirNames = map[string]bool{
+	"pdfs": true, "node_modules": true, ".git": true, ".venv": true, "venv": true,
+	"vendor": true, "tools": true, "dist": true, "build": true, ".pnpm": true,
+	"coverage": true, "__pycache__": true, "out": true, ".turbo": true, ".next": true,
+}
+
+var appFolderNum = regexp.MustCompile(`^\d{2,4}$`)
+
+// looksLikeAppFolder accepts CV dossier slugs (uae-009-…, djinni-001-…, linkedin-elastic-…).
+func looksLikeAppFolder(name string) bool {
+	if name == "" || strings.HasPrefix(name, ".") {
+		return false
+	}
+	parts := strings.Split(name, "-")
+	if len(parts) < 3 {
+		return false
+	}
+	for _, p := range parts {
+		if appFolderNum.MatchString(p) {
+			return true
+		}
+	}
+	// Rare dossiers without NNN (e.g. alex-staff-cloud-platform-engineer)
+	return len(parts) >= 4 && len(name) >= 24
+}
+
 type RecruiterInfo struct {
 	First, Last, JobTitle, Company, Email, Phone, LinkedIn string
 }
@@ -51,7 +78,7 @@ func Discover(base string) ([]string, error) {
 			return err
 		}
 		if d.IsDir() {
-			if d.Name() == "pdfs" {
+			if skipDirNames[d.Name()] {
 				return fs.SkipDir
 			}
 			return nil
@@ -63,6 +90,9 @@ func Discover(base string) ([]string, error) {
 		if filepath.Clean(dir) == filepath.Clean(base) {
 			return nil
 		}
+		if !looksLikeAppFolder(filepath.Base(dir)) {
+			return nil
+		}
 		out = append(out, path)
 		return nil
 	})
@@ -97,9 +127,13 @@ func ParseReadme(path string) (Data, error) {
 		app.Source = filepath.Base(filepath.Dir(filepath.Dir(path)))
 	}
 	if app.Company == "" || strings.EqualFold(app.Company, "undisclosed") || app.Company == "?" || strings.EqualFold(app.Company, "tbd") {
+		// Prefer slug token after NNN (uae-009-dizzaract-… → Dizzaract), not source prefix.
 		parts := strings.Split(filepath.Base(filepath.Dir(path)), "-")
-		if len(parts) >= 2 {
-			app.Company = titleWord(parts[0])
+		for i, p := range parts {
+			if appFolderNum.MatchString(p) && i+1 < len(parts) {
+				app.Company = titleWord(parts[i+1])
+				break
+			}
 		}
 	}
 	app.Summary = buildSummary(app, text)

cmd/oo/applications/applications_test.go

@@ -1,6 +1,8 @@
 package applications
 
 import (
+	"os"
+	"path/filepath"
 	"testing"
 
 	onlyoffice "github.com/eslider/go-onlyoffice"
@@ -15,6 +17,47 @@ func TestDealTitleForApplication(t *testing.T) {
 	}
 }
 
+func TestLooksLikeAppFolder(t *testing.T) {
+	cases := map[string]bool{
+		"uae-009-dizzaract-senior-backend-devops-abu-dhabi":    true,
+		"djinni-001-devops-engineer-windows-cloud-justmarkets": true,
+		"linkedin-elastic-senior-go-intake-observability":      true, // no NNN but long
+		"alex-staff-cloud-platform-engineer":                   true,
+		"http-client":                                          false,
+		"node_modules":                                         false,
+		"tools":                                                false,
+		"README":                                               false,
+		"uae-009":                                              false, // too short
+	}
+	for name, want := range cases {
+		if got := looksLikeAppFolder(name); got != want {
+			t.Errorf("looksLikeAppFolder(%q)=%v want %v", name, got, want)
+		}
+	}
+}
+
+func TestDiscoverSkipsNodeModules(t *testing.T) {
+	root := t.TempDir()
+	appDir := filepath.Join(root, "uae-009-dizzaract-senior-backend-devops-abu-dhabi")
+	junk := filepath.Join(root, "node_modules", "http-client")
+	tools := filepath.Join(root, "tools", "gmail-oo-reconcile")
+	for _, d := range []string{appDir, junk, tools} {
+		if err := os.MkdirAll(d, 0o755); err != nil {
+			t.Fatal(err)
+		}
+		if err := os.WriteFile(filepath.Join(d, "README.md"), []byte("# x\n"), 0o644); err != nil {
+			t.Fatal(err)
+		}
+	}
+	got, err := Discover(root)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(got) != 1 || filepath.Base(filepath.Dir(got[0])) != filepath.Base(appDir) {
+		t.Fatalf("Discover=%v want only %s", got, appDir)
+	}
+}
+
 func TestHasContactInfoHelper(t *testing.T) {
 	contact := map[string]any{
 		"commonData": []any{