Merge pull request #15 from eSlider/feat/oo-catalog-clients-contacts

feat(oo): catalog scan/match/apply for CRM clients and contacts

AGENTS.md

@@ -14,8 +14,9 @@ Canonical Go client for OnlyOffice Workspace (Projects + Calendar + CRM) and the
 - **CLI — `cmd/oo/` as `package main`.** Cobra wrapper that loads `.env` via `godotenv` at startup. **Subject-based command tree** mirroring [`tea`](https://gitea.com/gitea/tea):
   - `main.go` — entry point (docstring lists the command tree).
   - `common.go` — `rootCmd`, `newOO`, `printTable`/`printObject`, `--output table|json` flag.
-  - `calendar.go`, `projects.go`, `projects_files.go`, `tasks.go`, `tasks_files.go`, `users.go`, `contacts.go`, `opportunities.go`, `cases.go`, `crm_tasks.go`, `apps.go` — one file per subject (or per subject facet), each registers in `init()`.
+  - `calendar.go`, `projects.go`, `projects_files.go`, `tasks.go`, `tasks_files.go`, `users.go`, `contacts.go`, `opportunities.go`, `cases.go`, `crm_tasks.go`, `apps.go`, `catalog.go` — one file per subject (or per subject facet), each registers in `init()`.
   - CLI-only deps (`spf13/cobra`, `joho/godotenv`) stay out of the library.
+- **Catalog inventory — `catalog/` package.** Filesystem VCF/project scan → YAML → match/apply against CRM. Used by `oo catalog`; not part of the flat `*Client` surface (uses Client as a dependency).
 - **TUI — `cmd/office/` as `package main`.** Bubble Tea three-pane browser (module tree, selectable list, markdown preview). Reuses `cmd/internal/bootstrap` for env/auth and the root `onlyoffice` library for all API calls. UI logic in `cmd/office/ui/`; preview/formatting in `cmd/office/preview/`; list loaders in `cmd/office/fetch/`.
   - **List table (`DataTable`)** — `cmd/office/ui/table*.go`. Column layout policies live in `cmd/office/model/table_layout.go` (`TableFlexLayoutFor`); cell rendering uses the bubbles/table inline pattern in `table_render.go` (`renderTableCell`, `padANSIWidth`). See `.cursor/skills/office-tui-table/SKILL.md` before changing center-pane tables.
 - **Shared bootstrap — `cmd/internal/bootstrap/`.** `LoadEnv()` + `NewClient(ctx)` extracted from `oo`; both binaries import it.

README.md

@@ -635,6 +635,7 @@ oo tasks files detach 208 12345
 | `cases` | `list`, `create`, `delete`, `member-add` |
 | `crm-tasks` | `list`, `create`, `delete`, `categories` |
 | `applications` | `sync` |
+| `catalog` | `scan-contacts`, `scan-projects`, `scan-thunderbird`, `merge`, `match`, `apply` |
 
 The CLI reads only `.env` from the current working directory (godotenv is a
 CLI-only concern — the library itself never loads dotfiles).
@@ -733,6 +734,27 @@ oo contacts get 857 -o json
 oo crm cleanup -o json
 ```
 
+### Clients/contacts inventory (`catalog`)
+
+**Problem:** Contacts live as VCF/folders on disk and project trees under
+`~/projects` / arc-1; OnlyOffice CRM is the SSOT for persons/companies but
+starts sparse. You need a reviewable inventory before writes.
+
+```bash
+oo catalog scan-contacts --root /path/to/contacts -O /tmp/contacts.yaml
+oo catalog scan-projects --root ~/projects -O /tmp/projects.yaml
+oo catalog scan-thunderbird --root /path/to/thunderbird-profile -O /tmp/thunderbird.yaml
+oo catalog merge -i /tmp/contacts.yaml -i /tmp/projects.yaml -i /tmp/thunderbird.yaml -O clients-contacts.yaml
+oo catalog match -i clients-contacts.yaml
+# set approve: true on pilot rows in the YAML
+oo catalog apply --dry-run -i clients-contacts.yaml
+oo catalog apply --apply -i clients-contacts.yaml
+oo crm cleanup
+```
+
+YAML schema and ops runbook live in the inventar repo under `docs/catalog/` and
+`docs/ops/oo-clients-contacts-sync.md`. Library helper: `FindPersonByEmail`.
+
 ### Job applications → CRM (`applications sync`)
 
 **Problem:** You keep CVs in a folder tree (`applications/2026/Acme/README.md`)

catalog/apply.go

@@ -0,0 +1,199 @@
+package catalog
+
+import (
+	"context"
+	"fmt"
+	"strconv"
+	"strings"
+
+	onlyoffice "github.com/eslider/go-onlyoffice"
+)
+
+// ApplyResult summarizes one apply pass.
+type ApplyResult struct {
+	Created int      `json:"created"`
+	Updated int      `json:"updated"`
+	Skipped int      `json:"skipped"`
+	Errors  []string `json:"errors,omitempty"`
+	DryRun  bool     `json:"dry_run"`
+}
+
+// ApplyApproved creates/updates OO contacts for entries with approve=true.
+// Companies are applied before persons so Org links resolve in the same run.
+func ApplyApproved(ctx context.Context, client *onlyoffice.Client, doc *Document, dryRun bool) (*ApplyResult, error) {
+	res := &ApplyResult{DryRun: dryRun}
+	order := make([]int, 0, len(doc.Entries))
+	for i, e := range doc.Entries {
+		if e.Kind == "company" {
+			order = append(order, i)
+		}
+	}
+	for i, e := range doc.Entries {
+		if e.Kind != "company" {
+			order = append(order, i)
+		}
+	}
+	for _, i := range order {
+		e := &doc.Entries[i]
+		if !e.Approve {
+			res.Skipped++
+			continue
+		}
+		if e.Status == "conflict" {
+			res.Skipped++
+			res.Errors = append(res.Errors, e.ID+": conflict — resolve manually")
+			continue
+		}
+		if dryRun {
+			if e.OOID != "" || e.Status == "exists" {
+				res.Updated++
+			} else {
+				res.Created++
+			}
+			continue
+		}
+		created, err := applyOne(ctx, client, e)
+		if err != nil {
+			res.Errors = append(res.Errors, e.ID+": "+err.Error())
+			continue
+		}
+		if created {
+			res.Created++
+		} else {
+			res.Updated++
+		}
+	}
+	return res, nil
+}
+
+func applyOne(ctx context.Context, client *onlyoffice.Client, e *Entry) (created bool, err error) {
+	switch e.Kind {
+	case "company":
+		return applyCompany(ctx, client, e)
+	case "person":
+		return applyPerson(ctx, client, e)
+	default:
+		return false, fmt.Errorf("unknown kind %q", e.Kind)
+	}
+}
+
+func applyCompany(ctx context.Context, client *onlyoffice.Client, e *Entry) (bool, error) {
+	name := strings.TrimSpace(e.Name)
+	if name == "" {
+		return false, fmt.Errorf("company missing name")
+	}
+	var co map[string]any
+	var err error
+	if e.OOID != "" {
+		co, err = client.GetContact(ctx, e.OOID)
+	} else {
+		co, err = client.FindCompany(ctx, name)
+	}
+	if err != nil {
+		return false, err
+	}
+	created := false
+	if co == nil {
+		co, err = client.CreateCompany(ctx, name)
+		if err != nil {
+			return false, err
+		}
+		created = true
+	}
+	id := contactIDString(co)
+	e.OOID = id
+	if err := ensureContactInfos(ctx, client, id, e); err != nil {
+		return created, err
+	}
+	e.Status = "applied"
+	if created {
+		e.Notes = "created"
+	} else {
+		e.Notes = "updated"
+	}
+	return created, nil
+}
+
+func applyPerson(ctx context.Context, client *onlyoffice.Client, e *Entry) (bool, error) {
+	first := strings.TrimSpace(e.First)
+	last := strings.TrimSpace(e.Last)
+	if first == "" && last == "" {
+		first, last = SplitDisplayName(e.Name)
+	}
+	if first == "" {
+		first = strings.TrimSpace(e.Name)
+	}
+	if first == "" {
+		return false, fmt.Errorf("person missing name")
+	}
+	if last == "" {
+		last = "-"
+	}
+
+	var p map[string]any
+	var err error
+	if e.OOID != "" {
+		p, err = client.GetContact(ctx, e.OOID)
+	} else if len(e.Emails) > 0 {
+		p, err = client.FindPersonByEmail(ctx, e.Emails[0])
+	} else {
+		p, err = client.FindPerson(ctx, first, last)
+	}
+	if err != nil {
+		return false, err
+	}
+	created := false
+	companyID := 0
+	if e.Org != "" {
+		if co, ferr := client.FindCompany(ctx, e.Org); ferr == nil && co != nil {
+			companyID, _ = strconv.Atoi(contactIDString(co))
+		}
+	}
+	if p == nil {
+		about := ""
+		if e.Org != "" {
+			about = "org: " + e.Org
+		}
+		p, err = client.CreatePerson(ctx, first, last, companyID, "", about)
+		if err != nil {
+			return false, err
+		}
+		created = true
+	}
+	id := contactIDString(p)
+	e.OOID = id
+	if err := ensureContactInfos(ctx, client, id, e); err != nil {
+		return created, err
+	}
+	e.Status = "applied"
+	if created {
+		e.Notes = "created"
+	} else {
+		e.Notes = "updated"
+	}
+	return created, nil
+}
+
+func ensureContactInfos(ctx context.Context, client *onlyoffice.Client, contactID string, e *Entry) error {
+	existing, err := client.GetContact(ctx, contactID)
+	if err != nil {
+		return err
+	}
+	for i, em := range e.Emails {
+		if onlyoffice.HasContactInfo(existing, "Email", em) {
+			continue
+		}
+		if _, err := client.AddContactInfo(ctx, contactID, "Email", em, "Work", i == 0); err != nil {
+			return fmt.Errorf("add email %s: %w", em, err)
+		}
+	}
+	for _, ph := range e.Phones {
+		if onlyoffice.HasContactInfo(existing, "Phone", ph) {
+			continue
+		}
+		if _, err := client.AddContactInfo(ctx, contactID, "Phone", ph, "Work", false); err != nil {
+			return fmt.Errorf("add phone %s: %w", ph, err)
+		}
+	}
+	return nil
+}

catalog/catalog_test.go

@@ -0,0 +1,132 @@
+package catalog
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+)
+
+func TestParseVCFFile(t *testing.T) {
+	dir := t.TempDir()
+	path := filepath.Join(dir, "sample.vcf")
+	body := `BEGIN:VCARD
+VERSION:3.0
+FN:Ada Lovelace
+N:Lovelace;Ada;;;
+EMAIL;TYPE=INTERNET:ada@example.com
+TEL;TYPE=CELL:+1-555-0100
+ORG:Analytical Engines
+END:VCARD
+`
+	if err := os.WriteFile(path, []byte(body), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	entries, err := ParseVCFFile(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(entries) != 1 {
+		t.Fatalf("got %d entries", len(entries))
+	}
+	e := entries[0]
+	if e.Kind != "person" || e.First != "Ada" || e.Last != "Lovelace" {
+		t.Fatalf("unexpected entry: %+v", e)
+	}
+	if len(e.Emails) != 1 || e.Emails[0] != "ada@example.com" {
+		t.Fatalf("emails: %v", e.Emails)
+	}
+	if e.Org != "Analytical Engines" {
+		t.Fatalf("org: %q", e.Org)
+	}
+	if e.Approve {
+		t.Fatal("approve should default false")
+	}
+}
+
+func TestMergeDocs(t *testing.T) {
+	a := &Document{Entries: []Entry{{
+		ID: "person:ada@example.com", Kind: "person", First: "Ada", Emails: []string{"ada@example.com"},
+		Sources: []string{"a.vcf"}, Zone: "private", Role: "unknown",
+	}}}
+	b := &Document{Entries: []Entry{{
+		ID: "person:ada@example.com", Kind: "person", Last: "Lovelace", Phones: []string{"+1"},
+		Sources: []string{"folder/Ada"}, Zone: "warm", Role: "work", Approve: true,
+	}}}
+	m := MergeDocs(a, b)
+	if len(m.Entries) != 1 {
+		t.Fatalf("len=%d", len(m.Entries))
+	}
+	e := m.Entries[0]
+	if e.First != "Ada" || e.Last != "Lovelace" {
+		t.Fatalf("%+v", e)
+	}
+	if len(e.Sources) != 2 || len(e.Phones) != 1 || !e.Approve {
+		t.Fatalf("%+v", e)
+	}
+	if e.Zone != "warm" || e.Role != "work" {
+		t.Fatalf("zone/role: %s/%s", e.Zone, e.Role)
+	}
+}
+
+func TestScanContactsRoot(t *testing.T) {
+	root := t.TempDir()
+	vcfDir := filepath.Join(root, "Contacts VCF's")
+	if err := os.MkdirAll(vcfDir, 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(filepath.Join(vcfDir, "bob.vcf"), []byte(`BEGIN:VCARD
+VERSION:3.0
+FN:Bob Builder
+EMAIL:bob@work.test
+END:VCARD
+`), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	personDir := filepath.Join(root, "Carol Smith")
+	if err := os.MkdirAll(personDir, 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(filepath.Join(personDir, "carol@example.com.txt"), []byte(""), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	doc, err := ScanContactsRoot(root)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(doc.Entries) < 2 {
+		t.Fatalf("expected >=2 entries, got %d: %+v", len(doc.Entries), doc.Entries)
+	}
+}
+
+func TestScanProjectsRoot(t *testing.T) {
+	root := t.TempDir()
+	repo := filepath.Join(root, "produktor-demo")
+	if err := os.MkdirAll(filepath.Join(repo, ".git"), 0o755); err != nil {
+		t.Fatal(err)
+	}
+	doc, err := ScanProjectsRoot(root, 3)
+	if err != nil {
+		t.Fatal(err)
+	}
+	found := false
+	for _, e := range doc.Entries {
+		if e.Kind == "company" && e.Name == "produktor-demo" {
+			found = true
+			if e.Role != "work" {
+				t.Fatalf("role=%q", e.Role)
+			}
+		}
+	}
+	if !found {
+		t.Fatalf("missing company: %+v", doc.Entries)
+	}
+}
+
+func TestEntryID(t *testing.T) {
+	if got := EntryID("person", "A@B.COM", ""); got != "person:a@b.com" {
+		t.Fatal(got)
+	}
+	if got := EntryID("company", "", "Acme Corp"); got != "company:acme corp" {
+		t.Fatal(got)
+	}
+}

catalog/match.go

@@ -0,0 +1,133 @@
+package catalog
+
+import (
+	"context"
+	"fmt"
+	"strconv"
+	"strings"
+
+	onlyoffice "github.com/eslider/go-onlyoffice"
+)
+
+// MatchAgainstOO sets status/oo_id by email then normalized name.
+func MatchAgainstOO(ctx context.Context, client *onlyoffice.Client, doc *Document) error {
+	all, err := client.ListAllContacts(ctx)
+	if err != nil {
+		return err
+	}
+
+	byEmail := map[string]map[string]any{}
+	byPersonName := map[string]map[string]any{}
+	byCompanyName := map[string]map[string]any{}
+
+	for _, c := range all {
+		id := contactIDString(c)
+		isCo, _ := c["isCompany"].(bool)
+		for _, em := range contactEmails(c) {
+			byEmail[NormalizeEmail(em)] = c
+		}
+		if isCo {
+			key := NormalizeName(fmt.Sprint(c["displayName"]))
+			if key != "" {
+				byCompanyName[key] = c
+			}
+			continue
+		}
+		first := strings.TrimSpace(fmt.Sprint(c["firstName"]))
+		last := strings.TrimSpace(fmt.Sprint(c["lastName"]))
+		key := NormalizeName(first + " " + last)
+		if key == "" {
+			key = NormalizeName(fmt.Sprint(c["displayName"]))
+		}
+		if key != "" {
+			byPersonName[key] = c
+		}
+		_ = id
+	}
+
+	for i := range doc.Entries {
+		e := &doc.Entries[i]
+		// preserve approve
+		matched := false
+		conflict := false
+		var oo map[string]any
+
+		for _, em := range e.Emails {
+			if c, ok := byEmail[NormalizeEmail(em)]; ok {
+				if oo != nil && contactIDString(oo) != contactIDString(c) {
+					conflict = true
+				}
+				oo = c
+				matched = true
+			}
+		}
+		if !matched {
+			if e.Kind == "company" {
+				if c, ok := byCompanyName[NormalizeName(e.Name)]; ok {
+					oo = c
+					matched = true
+				}
+			} else {
+				key := NormalizeName(strings.TrimSpace(e.First + " " + e.Last))
+				if key == "" {
+					key = NormalizeName(e.Name)
+				}
+				if c, ok := byPersonName[key]; ok {
+					oo = c
+					matched = true
+				}
+			}
+		}
+
+		if conflict {
+			e.Status = "conflict"
+			e.Notes = strings.TrimSpace(e.Notes + " email_matches_multiple_oo")
+			if oo != nil {
+				e.OOID = contactIDString(oo)
+			}
+			continue
+		}
+		if matched && oo != nil {
+			e.Status = "exists"
+			e.OOID = contactIDString(oo)
+			continue
+		}
+		e.Status = "new"
+		e.OOID = ""
+	}
+	return nil
+}
+
+func contactIDString(c map[string]any) string {
+	switch v := c["id"].(type) {
+	case float64:
+		return strconv.FormatInt(int64(v), 10)
+	case int:
+		return strconv.Itoa(v)
+	case string:
+		return v
+	default:
+		return fmt.Sprint(v)
+	}
+}
+
+func contactEmails(c map[string]any) []string {
+	var out []string
+	if em := strings.TrimSpace(fmt.Sprint(c["email"])); em != "" && em != "<nil>" {
+		out = append(out, em)
+	}
+	if em := strings.TrimSpace(fmt.Sprint(c["primaryEmail"])); em != "" && em != "<nil>" {
+		out = append(out, em)
+	}
+	for _, row := range onlyoffice.ContactInfoRows(c) {
+		t := strings.ToLower(fmt.Sprint(row["infoType"]))
+		if t != "email" {
+			continue
+		}
+		data := strings.TrimSpace(fmt.Sprint(row["data"]))
+		if data != "" {
+			out = append(out, data)
+		}
+	}
+	return uniqueEmails(out)
+}

catalog/match_test.go

@@ -0,0 +1,26 @@
+package catalog
+
+import (
+	"testing"
+)
+
+func TestMatchAgainstOO_emailThenName(t *testing.T) {
+	// MatchAgainstOO needs a live client; unit-test the helpers used by matching.
+	c := map[string]any{
+		"id":        float64(42),
+		"isCompany": false,
+		"firstName": "Ada",
+		"lastName":  "Lovelace",
+		"email":     "ada@example.com",
+		"commonData": []any{
+			map[string]any{"infoType": "Email", "data": "ada.alt@example.com"},
+		},
+	}
+	if contactIDString(c) != "42" {
+		t.Fatal(contactIDString(c))
+	}
+	emails := contactEmails(c)
+	if len(emails) < 2 {
+		t.Fatalf("%v", emails)
+	}
+}

catalog/mbox.go

@@ -0,0 +1,238 @@
+package catalog
+
+import (
+	"bufio"
+	"fmt"
+	"io"
+	"os"
+	"path/filepath"
+	"strings"
+)
+
+// ScanOptions controls Thunderbird / mbox extraction depth.
+type ScanOptions struct {
+	// MboxHeaders, when true, also walks mbox-like files under the root and
+	// extracts From/To/Cc/Reply-To addresses (headers only, no bodies).
+	MboxHeaders bool
+	// MboxMaxBytes skips individual mbox files larger than this (0 = 256 MiB default).
+	MboxMaxBytes int64
+}
+
+// ScanThunderbirdRoot finds Thunderbird profiles under root and emits person rows
+// from address books (*.mab) and Gloda global-messages-db.sqlite.
+func ScanThunderbirdRoot(root string) (*Document, error) {
+	return ScanThunderbirdRootOpts(root, ScanOptions{})
+}
+
+// ScanThunderbirdRootOpts is ScanThunderbirdRoot with optional mbox header pass.
+func ScanThunderbirdRootOpts(root string, opts ScanOptions) (*Document, error) {
+	root = filepath.Clean(root)
+	st, err := os.Stat(root)
+	if err != nil {
+		return nil, err
+	}
+	if !st.IsDir() {
+		return nil, fmt.Errorf("not a directory: %s", root)
+	}
+	if opts.MboxMaxBytes <= 0 {
+		opts.MboxMaxBytes = 256 << 20
+	}
+
+	var entries []Entry
+	seenDB := map[string]struct{}{}
+	seenMAB := map[string]struct{}{}
+	seenMbox := map[string]struct{}{}
+
+	err = filepath.WalkDir(root, func(path string, d os.DirEntry, err error) error {
+		if err != nil {
+			return nil
+		}
+		name := d.Name()
+		if d.IsDir() {
+			switch name {
+			case "Cache", "cache2", "startupCache", "OfflineCache", "minidumps",
+				"crashes", "safebrowsing", "thumbnails", "chrome", "extensions",
+				"node_modules", ".git":
+				return filepath.SkipDir
+			}
+			return nil
+		}
+		lower := strings.ToLower(name)
+		switch {
+		case lower == "global-messages-db.sqlite":
+			if _, ok := seenDB[path]; ok {
+				return nil
+			}
+			seenDB[path] = struct{}{}
+			parsed, perr := parseGlodaContacts(path)
+			if perr != nil {
+				entries = append(entries, Entry{
+					ID:      EntryID("person", "", filepath.Base(path)),
+					Kind:    "person",
+					Name:    "gloda",
+					Sources: []string{path},
+					Zone:    "private",
+					Role:    "unknown",
+					Notes:   "gloda_parse_error: " + perr.Error(),
+					Status:  "new",
+				})
+				return nil
+			}
+			entries = append(entries, parsed...)
+		case strings.HasSuffix(lower, ".mab"):
+			if _, ok := seenMAB[path]; ok {
+				return nil
+			}
+			seenMAB[path] = struct{}{}
+			parsed, perr := parseMABEmails(path)
+			if perr != nil {
+				return nil
+			}
+			entries = append(entries, parsed...)
+		case opts.MboxHeaders && isLikelyMboxFile(name, path):
+			if _, ok := seenMbox[path]; ok {
+				return nil
+			}
+			seenMbox[path] = struct{}{}
+			info, ierr := d.Info()
+			if ierr != nil {
+				return nil
+			}
+			if info.Size() > opts.MboxMaxBytes {
+				return nil
+			}
+			parsed, perr := parseMboxHeaderEmails(path)
+			if perr != nil {
+				return nil
+			}
+			entries = append(entries, parsed...)
+		}
+		return nil
+	})
+	if err != nil {
+		return nil, err
+	}
+
+	return MergeDocs(&Document{Entries: entries}), nil
+}
+
+func isLikelyMboxFile(name, path string) bool {
+	lower := strings.ToLower(name)
+	if strings.HasSuffix(lower, ".msf") || strings.HasSuffix(lower, ".sqlite") ||
+		strings.HasSuffix(lower, ".mab") || strings.HasSuffix(lower, ".json") ||
+		strings.HasSuffix(lower, ".dat") || strings.HasSuffix(lower, ".ini") ||
+		strings.HasSuffix(lower, ".log") {
+		return false
+	}
+	if strings.HasSuffix(lower, ".mbox") || strings.HasSuffix(lower, ".mbx") {
+		return true
+	}
+	// Thunderbird: …/ImapMail/<server>/INBOX or …/Mail/Local Folders/Inbox
+	// or …/INBOX.sbd/<folder>
+	sep := string(filepath.Separator)
+	norm := filepath.ToSlash(path)
+	if strings.Contains(norm, "/ImapMail/") || strings.Contains(norm, "/Mail/") {
+		if !strings.Contains(name, ".") {
+			return true
+		}
+		known := map[string]struct{}{
+			"inbox": {}, "sent": {}, "drafts": {}, "trash": {}, "archives": {}, "junk": {},
+		}
+		if _, ok := known[lower]; ok {
+			return true
+		}
+	}
+	if strings.Contains(filepath.Base(filepath.Dir(path)), ".sbd") && !strings.Contains(name, ".") {
+		return true
+	}
+	_ = sep
+	return false
+}
+
+// parseMboxHeaderEmails extracts addresses from From/To/Cc/Reply-To headers only.
+func parseMboxHeaderEmails(path string) ([]Entry, error) {
+	f, err := os.Open(path)
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+
+	emails := map[string]struct{}{}
+	br := bufio.NewReaderSize(f, 1<<20)
+	inHeaders := false
+	var headerBuf strings.Builder
+
+	flushHeaders := func() {
+		if headerBuf.Len() == 0 {
+			return
+		}
+		block := headerBuf.String()
+		headerBuf.Reset()
+		for _, line := range strings.Split(block, "\n") {
+			lower := strings.ToLower(strings.TrimSpace(line))
+			if strings.HasPrefix(lower, "from:") || strings.HasPrefix(lower, "to:") ||
+				strings.HasPrefix(lower, "cc:") || strings.HasPrefix(lower, "reply-to:") ||
+				strings.HasPrefix(lower, "sender:") {
+				for _, m := range emailRE.FindAllString(line, -1) {
+					em := NormalizeEmail(m)
+					if !noisyEmail(em) {
+						emails[em] = struct{}{}
+					}
+				}
+			}
+		}
+	}
+
+	for {
+		line, err := br.ReadString('\n')
+		if len(line) > 0 {
+			// Classic mbox separator
+			if strings.HasPrefix(line, "From ") && (len(line) == 5 || line[5] != ':') {
+				flushHeaders()
+				inHeaders = true
+				continue
+			}
+			if inHeaders {
+				trimmed := strings.TrimRight(line, "\r\n")
+				if trimmed == "" {
+					flushHeaders()
+					inHeaders = false
+					continue
+				}
+				headerBuf.WriteString(trimmed)
+				headerBuf.WriteByte('\n')
+				// cap pathological headers
+				if headerBuf.Len() > 64<<10 {
+					flushHeaders()
+					inHeaders = false
+				}
+			}
+		}
+		if err == io.EOF {
+			flushHeaders()
+			break
+		}
+		if err != nil {
+			flushHeaders()
+			return nil, err
+		}
+	}
+
+	var out []Entry
+	for em := range emails {
+		org, zone, role := classifyMailIdentity("", em)
+		out = append(out, Entry{
+			ID:      EntryID("person", em, ""),
+			Kind:    "person",
+			Emails:  []string{em},
+			Org:     org,
+			Sources: []string{path},
+			Zone:    zone,
+			Role:    role,
+			Approve: false,
+			Status:  "new",
+			Notes:   "thunderbird_mbox_header",
+		})
+	}
+	return out, nil
+}

catalog/mbox_test.go

@@ -0,0 +1,67 @@
+package catalog
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+)
+
+func TestParseMboxHeaderEmails(t *testing.T) {
+	dir := t.TempDir()
+	path := filepath.Join(dir, "INBOX")
+	body := `From - Mon Jul 1 00:00:00 2016
+From: Axel Schaefer <axel.schaefer@wheregroup.com>
+To: Andriy Oblivantsev <andriy.oblivantsev@wheregroup.com>
+Cc: noreply@example.com, client@stadt-example.de
+Subject: test
+
+Body line ignored
+From - Mon Jul 2 00:00:00 2016
+From: Someone <paul.schmidt@wheregroup.com>
+To: list@wheregroup.com
+
+more body
+`
+	if err := os.WriteFile(path, []byte(body), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	ents, err := parseMboxHeaderEmails(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	got := map[string]bool{}
+	for _, e := range ents {
+		if len(e.Emails) > 0 {
+			got[e.Emails[0]] = true
+		}
+	}
+	if !got["axel.schaefer@wheregroup.com"] || !got["andriy.oblivantsev@wheregroup.com"] {
+		t.Fatalf("%v", got)
+	}
+	if got["noreply@example.com"] {
+		t.Fatal("noreply should be filtered")
+	}
+	if !got["client@stadt-example.de"] {
+		t.Fatal("missing client")
+	}
+}
+
+func TestScanThunderbirdRootOptsMbox(t *testing.T) {
+	root := t.TempDir()
+	imap := filepath.Join(root, "ImapMail", "mail.example.com")
+	if err := os.MkdirAll(imap, 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(filepath.Join(imap, "INBOX"), []byte(
+		"From - x\nFrom: a@wheregroup.com\nTo: b@wheregroup.com\n\nbody\n",
+	), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	doc, err := ScanThunderbirdRootOpts(root, ScanOptions{MboxHeaders: true})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(doc.Entries) < 2 {
+		t.Fatalf("%+v", doc.Entries)
+	}
+}

catalog/scan_projects.go

@@ -0,0 +1,127 @@
+package catalog
+
+import (
+	"fmt"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"strings"
+)
+
+// ScanProjectsRoot finds git roots under root (max depth) and emits company rows.
+func ScanProjectsRoot(root string, maxDepth int) (*Document, error) {
+	root = filepath.Clean(root)
+	if maxDepth <= 0 {
+		maxDepth = 4
+	}
+	st, err := os.Stat(root)
+	if err != nil {
+		return nil, err
+	}
+	if !st.IsDir() {
+		return nil, fmt.Errorf("not a directory: %s", root)
+	}
+
+	var entries []Entry
+	err = walkGitRoots(root, root, 0, maxDepth, &entries)
+	if err != nil {
+		return nil, err
+	}
+
+	// Also top-level dirs as company stubs (even without git) — coverage 1B.
+	dents, _ := os.ReadDir(root)
+	for _, d := range dents {
+		if !d.IsDir() || strings.HasPrefix(d.Name(), ".") {
+			continue
+		}
+		name := d.Name()
+		skip := map[string]struct{}{
+			"node_modules": {}, "vendor": {}, ".cache": {},
+		}
+		if _, ok := skip[name]; ok {
+			continue
+		}
+		path := filepath.Join(root, name)
+		id := EntryID("company", "", name)
+		role, zone := classifyProjectName(name, "")
+		entries = append(entries, Entry{
+			ID:      id,
+			Kind:    "company",
+			Name:    name,
+			Sources: []string{path},
+			Zone:    zone,
+			Role:    role,
+			Status:  "new",
+			Notes:   "top_level_dir",
+		})
+	}
+
+	return MergeDocs(&Document{Entries: entries}), nil
+}
+
+func walkGitRoots(root, dir string, depth, maxDepth int, out *[]Entry) error {
+	if depth > maxDepth {
+		return nil
+	}
+	// is this a git root?
+	if st, err := os.Stat(filepath.Join(dir, ".git")); err == nil && (st.IsDir() || st.Mode().IsRegular()) {
+		name := filepath.Base(dir)
+		remote := gitRemoteOrigin(dir)
+		role, zone := classifyProjectName(name, remote)
+		*out = append(*out, Entry{
+			ID:      EntryID("company", "", name),
+			Kind:    "company",
+			Name:    name,
+			Sources: []string{dir},
+			Remote:  remote,
+			GitRoot: dir,
+			Zone:    zone,
+			Role:    role,
+			Status:  "new",
+		})
+		return nil // do not descend into nested repos from a git root
+	}
+	dents, err := os.ReadDir(dir)
+	if err != nil {
+		return nil
+	}
+	for _, d := range dents {
+		if !d.IsDir() {
+			continue
+		}
+		name := d.Name()
+		if name == ".git" || name == "node_modules" || name == "vendor" || name == ".venv" || name == "dist" {
+			continue
+		}
+		_ = walkGitRoots(root, filepath.Join(dir, name), depth+1, maxDepth, out)
+	}
+	return nil
+}
+
+func gitRemoteOrigin(dir string) string {
+	cmd := exec.Command("git", "-C", dir, "remote", "get-url", "origin")
+	b, err := cmd.Output()
+	if err != nil {
+		return ""
+	}
+	return strings.TrimSpace(string(b))
+}
+
+func classifyProjectName(name, remote string) (role, zone string) {
+	lower := strings.ToLower(name)
+	remoteL := strings.ToLower(remote)
+	switch {
+	case strings.Contains(lower, "experiment") || strings.HasPrefix(lower, "test"):
+		return "experiment", "cold"
+	case lower == "mama" || lower == "personal" || strings.Contains(lower, "private"):
+		return "personal", "private"
+	case strings.Contains(remoteL, "git.produktor.io") || strings.Contains(remoteL, "github.com/eslider"):
+		return "work", "hot"
+	case strings.Contains(lower, "produktor") || strings.Contains(lower, "eslider") ||
+		strings.Contains(lower, "asesoria") || strings.Contains(lower, "dyvenia") ||
+		strings.Contains(lower, "onlyoffice"):
+		return "work", "warm"
+	default:
+		return "unknown", "warm"
+	}
+}

catalog/thunderbird.go

@@ -0,0 +1,191 @@
+package catalog
+
+import (
+	"database/sql"
+	"os"
+	"regexp"
+	"strings"
+
+	_ "modernc.org/sqlite"
+)
+
+var emailRE = regexp.MustCompile(`(?i)[a-z0-9._%+\-]+@[a-z0-9.\-]+\.[a-z]{2,}`)
+
+// noisyEmail reports addresses that should not become CRM catalog persons.
+func noisyEmail(email string) bool {
+	e := NormalizeEmail(email)
+	if e == "" || !strings.Contains(e, "@") {
+		return true
+	}
+	local, domain, ok := strings.Cut(e, "@")
+	if !ok {
+		return true
+	}
+	noiseLocal := []string{
+		"noreply", "no-reply", "donotreply", "mailer-daemon", "postmaster",
+		"bounce", "notifications", "newsletter", "unsubscribe",
+		"root", "admin", "abuse", "webmaster", "hostmaster",
+	}
+	for _, p := range noiseLocal {
+		if local == p || strings.Contains(local, p) {
+			return true
+		}
+	}
+	// role / shared mailboxes (keep human dotted names)
+	roleExact := map[string]struct{}{
+		"info": {}, "alle": {}, "all": {}, "office": {}, "office@": {},
+		"wartung": {}, "starface": {}, "gitlab": {}, "pl": {}, "gf": {},
+		"umsetzung": {}, "support": {}, "sales": {}, "billing": {},
+	}
+	if _, ok := roleExact[local]; ok {
+		return true
+	}
+	if strings.HasSuffix(local, "-request") || strings.HasSuffix(local, "-owner") {
+		return true
+	}
+	noiseDomain := []string{
+		"marketplace.amazon.", "reply.github.com", "users.noreply.github.com",
+		"groups.facebook.com", "googlegroups.com", "yahoogroups.",
+		"email.apple.com", "amazonses.com", "mailchimp.com",
+		"sendgrid.net", "mailgun.org", "postmarkapp.com",
+		"property.booking.com", "transvascular.com",
+	}
+	for _, p := range noiseDomain {
+		if strings.Contains(domain, p) {
+			return true
+		}
+	}
+	// random marketplace / tracking locals
+	if len(local) >= 16 && !strings.Contains(local, ".") && !strings.Contains(local, "-") {
+		if strings.Contains(domain, "amazon.") {
+			return true
+		}
+	}
+	return false
+}
+
+func parseMABEmails(path string) ([]Entry, error) {
+	b, err := os.ReadFile(path)
+	if err != nil {
+		return nil, err
+	}
+	found := emailRE.FindAllString(string(b), -1)
+	var out []Entry
+	for _, raw := range found {
+		em := NormalizeEmail(raw)
+		if noisyEmail(em) {
+			continue
+		}
+		org, zone, role := classifyMailIdentity("", em)
+		out = append(out, Entry{
+			ID:      EntryID("person", em, ""),
+			Kind:    "person",
+			Emails:  []string{em},
+			Org:     org,
+			Sources: []string{path},
+			Zone:    zone,
+			Role:    role,
+			Approve: false,
+			Status:  "new",
+			Notes:   "thunderbird_mab",
+		})
+	}
+	return out, nil
+}
+
+func parseGlodaContacts(dbPath string) ([]Entry, error) {
+	// read-only URI; immutable=1 helps when WAL/shm are missing
+	dsn := "file:" + dbPath + "?mode=ro&_pragma=query_only(1)"
+	db, err := sql.Open("sqlite", dsn)
+	if err != nil {
+		return nil, err
+	}
+	defer db.Close()
+
+	rows, err := db.Query(`
+		SELECT COALESCE(c.name, ''), i.value
+		FROM identities i
+		LEFT JOIN contacts c ON c.id = i.contactID
+		WHERE lower(i.kind) = 'email' AND i.value IS NOT NULL AND i.value != ''
+	`)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	var out []Entry
+	for rows.Next() {
+		var name, value string
+		if err := rows.Scan(&name, &value); err != nil {
+			return out, err
+		}
+		em := NormalizeEmail(value)
+		// Gloda sometimes stores "Name <email>" in value
+		if m := emailRE.FindString(em); m != "" {
+			em = NormalizeEmail(m)
+		}
+		if noisyEmail(em) {
+			continue
+		}
+		display := strings.TrimSpace(name)
+		// strip wrapping quotes / email leftovers
+		display = strings.Trim(display, `"'`)
+		if idx := strings.Index(display, "<"); idx > 0 {
+			display = strings.TrimSpace(display[:idx])
+		}
+		first, last := "", ""
+		if display != "" {
+			first, last = SplitDisplayName(display)
+		}
+		org, zone, role := classifyMailIdentity(display, em)
+		out = append(out, Entry{
+			ID:      EntryID("person", em, display),
+			Kind:    "person",
+			Name:    display,
+			First:   first,
+			Last:    last,
+			Emails:  []string{em},
+			Org:     org,
+			Sources: []string{dbPath},
+			Zone:    zone,
+			Role:    role,
+			Approve: false,
+			Status:  "new",
+			Notes:   "thunderbird_gloda",
+		})
+	}
+	return out, rows.Err()
+}
+
+func classifyMailIdentity(name, email string) (org, zone, role string) {
+	em := NormalizeEmail(email)
+	_, domain, _ := strings.Cut(em, "@")
+	switch {
+	case domain == "wheregroup.com" || strings.Contains(strings.ToLower(name), "wheregroup"):
+		return "WhereGroup", "warm", "work"
+	case domain == "produktor.io" || domain == "eslider.de" || strings.HasSuffix(domain, ".produktor.io"):
+		return "produktor.io", "hot", "work"
+	case domain == "dyvenia.com":
+		return "Dyvenia", "warm", "work"
+	case domain == "immowelt.de" || domain == "immowelt.com":
+		return "Immowelt", "warm", "work"
+	case strings.HasSuffix(domain, ".de") && looksPublicSector(domain):
+		return domain, "warm", "work"
+	default:
+		return "", "private", "unknown"
+	}
+}
+
+func looksPublicSector(domain string) bool {
+	d := strings.ToLower(domain)
+	hints := []string{
+		"stadt-", "stadt.", "kreis-", "gemeinde", "landkreis",
+		"bund.", "bundes", "lvermgeo", "rvr.", "eba.",
+	}
+	for _, h := range hints {
+		if strings.Contains(d, h) {
+			return true
+		}
+	}
+	return false
+}

catalog/thunderbird_test.go

@@ -0,0 +1,97 @@
+package catalog
+
+import (
+	"database/sql"
+	"os"
+	"path/filepath"
+	"testing"
+
+	_ "modernc.org/sqlite"
+)
+
+func TestNoisyEmail(t *testing.T) {
+	if !noisyEmail("noreply@example.com") {
+		t.Fatal("expected noisy")
+	}
+	if !noisyEmail("x@marketplace.amazon.de") {
+		t.Fatal("amazon marketplace")
+	}
+	if noisyEmail("andriy.oblivantsev@wheregroup.com") {
+		t.Fatal("should keep wheregroup")
+	}
+}
+
+func TestParseMABEmails(t *testing.T) {
+	dir := t.TempDir()
+	path := filepath.Join(dir, "abook.mab")
+	body := `// mork junk
+		PrimaryEmail=andriy.oblivantsev@wheregroup.com
+		noreply@github.com
+		axel.schaefer@wheregroup.com
+`
+	if err := os.WriteFile(path, []byte(body), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	ents, err := parseMABEmails(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(ents) != 2 {
+		t.Fatalf("got %d: %+v", len(ents), ents)
+	}
+	for _, e := range ents {
+		if e.Org != "WhereGroup" || e.Role != "work" {
+			t.Fatalf("%+v", e)
+		}
+	}
+}
+
+func TestParseGlodaContacts(t *testing.T) {
+	dir := t.TempDir()
+	dbPath := filepath.Join(dir, "global-messages-db.sqlite")
+	db, err := sql.Open("sqlite", dbPath)
+	if err != nil {
+		t.Fatal(err)
+	}
+	_, err = db.Exec(`
+		CREATE TABLE contacts (id INTEGER PRIMARY KEY, name TEXT);
+		CREATE TABLE identities (id INTEGER PRIMARY KEY, contactID INTEGER, kind TEXT, value TEXT);
+		INSERT INTO contacts VALUES (1, 'Axel Schaefer');
+		INSERT INTO identities VALUES (1, 1, 'email', 'axel.schaefer@wheregroup.com');
+		INSERT INTO contacts VALUES (2, 'Noise Bot');
+		INSERT INTO identities VALUES (2, 2, 'email', 'noreply@example.com');
+	`)
+	if err != nil {
+		t.Fatal(err)
+	}
+	_ = db.Close()
+
+	ents, err := parseGlodaContacts(dbPath)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(ents) != 1 {
+		t.Fatalf("got %d %+v", len(ents), ents)
+	}
+	if ents[0].First != "Axel" || ents[0].Emails[0] != "axel.schaefer@wheregroup.com" {
+		t.Fatalf("%+v", ents[0])
+	}
+}
+
+func TestScanThunderbirdRoot(t *testing.T) {
+	root := t.TempDir()
+	prof := filepath.Join(root, "profile")
+	if err := os.MkdirAll(prof, 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(filepath.Join(prof, "abook.mab"), []byte("mail=paul.schmidt@wheregroup.com\n"), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	doc, err := ScanThunderbirdRoot(root)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(doc.Entries) < 1 {
+		t.Fatal(doc.Entries)
+	}
+}

catalog/types.go

@@ -0,0 +1,211 @@
+// Package catalog builds a YAML inventory of CRM persons/companies from
+// filesystem contacts (VCF, folders) and project trees, then matches/applies
+// against OnlyOffice CRM.
+package catalog
+
+import (
+	"fmt"
+	"os"
+	"strings"
+	"time"
+
+	"gopkg.in/yaml.v3"
+)
+
+// Entry is one catalog row (person or company).
+type Entry struct {
+	ID      string   `yaml:"id"`
+	Kind    string   `yaml:"kind"` // person | company
+	Name    string   `yaml:"name,omitempty"`
+	First   string   `yaml:"first,omitempty"`
+	Last    string   `yaml:"last,omitempty"`
+	Emails  []string `yaml:"emails,omitempty"`
+	Phones  []string `yaml:"phones,omitempty"`
+	Org     string   `yaml:"org,omitempty"`
+	Sources []string `yaml:"sources,omitempty"`
+	Zone    string   `yaml:"zone"`
+	Role    string   `yaml:"role"`
+	OOID    string   `yaml:"oo_id,omitempty"`
+	Approve bool     `yaml:"approve"`
+	Status  string   `yaml:"status,omitempty"` // new | exists | conflict | applied | skipped
+	Notes   string   `yaml:"notes,omitempty"`
+	Remote  string   `yaml:"remote,omitempty"`
+	GitRoot string   `yaml:"git_root,omitempty"`
+}
+
+// Document is the on-disk catalog file.
+type Document struct {
+	GeneratedAt string  `yaml:"generated_at"`
+	Entries     []Entry `yaml:"entries"`
+}
+
+// LoadYAML reads a catalog document from path.
+func LoadYAML(path string) (*Document, error) {
+	b, err := os.ReadFile(path)
+	if err != nil {
+		return nil, err
+	}
+	var doc Document
+	if err := yaml.Unmarshal(b, &doc); err != nil {
+		return nil, err
+	}
+	return &doc, nil
+}
+
+// SaveYAML writes the catalog document.
+func SaveYAML(path string, doc *Document) error {
+	doc.GeneratedAt = time.Now().UTC().Format(time.RFC3339)
+	b, err := yaml.Marshal(doc)
+	if err != nil {
+		return err
+	}
+	return os.WriteFile(path, b, 0o644)
+}
+
+// NormalizeEmail lowercases and trims.
+func NormalizeEmail(s string) string {
+	return strings.ToLower(strings.TrimSpace(s))
+}
+
+// NormalizeName collapses whitespace and lowercases for matching.
+func NormalizeName(s string) string {
+	fields := strings.Fields(strings.ToLower(strings.TrimSpace(s)))
+	return strings.Join(fields, " ")
+}
+
+// SplitDisplayName splits "First Last …" into first/last (last = remainder).
+func SplitDisplayName(name string) (first, last string) {
+	parts := strings.Fields(strings.TrimSpace(name))
+	if len(parts) == 0 {
+		return "", ""
+	}
+	if len(parts) == 1 {
+		return parts[0], ""
+	}
+	return parts[0], strings.Join(parts[1:], " ")
+}
+
+// EntryID builds a stable-ish id from kind + email or name.
+func EntryID(kind, email, name string) string {
+	if e := NormalizeEmail(email); e != "" {
+		return fmt.Sprintf("%s:%s", kind, e)
+	}
+	return fmt.Sprintf("%s:%s", kind, NormalizeName(name))
+}
+
+// MergeDocs unions entries by id; later sources append sources[] and fill empties.
+func MergeDocs(docs ...*Document) *Document {
+	byID := map[string]*Entry{}
+	order := []string{}
+	for _, d := range docs {
+		if d == nil {
+			continue
+		}
+		for i := range d.Entries {
+			e := d.Entries[i]
+			id := e.ID
+			if id == "" {
+				primary := ""
+				if len(e.Emails) > 0 {
+					primary = e.Emails[0]
+				}
+				name := e.Name
+				if name == "" {
+					name = strings.TrimSpace(e.First + " " + e.Last)
+				}
+				id = EntryID(e.Kind, primary, name)
+				e.ID = id
+			}
+			if prev, ok := byID[id]; ok {
+				mergeEntry(prev, &e)
+			} else {
+				cp := e
+				byID[id] = &cp
+				order = append(order, id)
+			}
+		}
+	}
+	out := &Document{Entries: make([]Entry, 0, len(order))}
+	for _, id := range order {
+		out.Entries = append(out.Entries, *byID[id])
+	}
+	return out
+}
+
+func mergeEntry(dst, src *Entry) {
+	dst.Sources = uniqueStrings(append(dst.Sources, src.Sources...))
+	dst.Emails = uniqueEmails(append(dst.Emails, src.Emails...))
+	dst.Phones = uniqueStrings(append(dst.Phones, src.Phones...))
+	if dst.First == "" {
+		dst.First = src.First
+	}
+	if dst.Last == "" {
+		dst.Last = src.Last
+	}
+	if dst.Name == "" {
+		dst.Name = src.Name
+	}
+	if dst.Org == "" {
+		dst.Org = src.Org
+	}
+	if dst.Remote == "" {
+		dst.Remote = src.Remote
+	}
+	if dst.GitRoot == "" {
+		dst.GitRoot = src.GitRoot
+	}
+	if dst.Notes == "" {
+		dst.Notes = src.Notes
+	}
+	// Prefer more specific zone/role from project scans over default private.
+	if dst.Zone == "private" && src.Zone != "" && src.Zone != "private" {
+		dst.Zone = src.Zone
+	}
+	if dst.Role == "" || (dst.Role == "unknown" && src.Role != "" && src.Role != "unknown") {
+		dst.Role = src.Role
+	}
+	// Preserve approve/oo_id/status from whichever already set.
+	if !dst.Approve && src.Approve {
+		dst.Approve = true
+	}
+	if dst.OOID == "" {
+		dst.OOID = src.OOID
+	}
+	if dst.Status == "" {
+		dst.Status = src.Status
+	}
+}
+
+func uniqueEmails(in []string) []string {
+	seen := map[string]struct{}{}
+	var out []string
+	for _, e := range in {
+		e = NormalizeEmail(e)
+		if e == "" {
+			continue
+		}
+		if _, ok := seen[e]; ok {
+			continue
+		}
+		seen[e] = struct{}{}
+		out = append(out, e)
+	}
+	return out
+}
+
+func uniqueStrings(in []string) []string {
+	seen := map[string]struct{}{}
+	var out []string
+	for _, s := range in {
+		s = strings.TrimSpace(s)
+		if s == "" {
+			continue
+		}
+		if _, ok := seen[s]; ok {
+			continue
+		}
+		seen[s] = struct{}{}
+		out = append(out, s)
+	}
+	return out
+}

catalog/vcf.go

@@ -0,0 +1,218 @@
+package catalog
+
+import (
+	"bufio"
+	"errors"
+	"fmt"
+	"io"
+	"os"
+	"path/filepath"
+	"strings"
+
+	"github.com/emersion/go-vcard"
+)
+
+// ParseVCFFile extracts person candidates from a vCard file (may contain many cards).
+func ParseVCFFile(path string) ([]Entry, error) {
+	f, err := os.Open(path)
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+
+	dec := vcard.NewDecoder(f)
+	var out []Entry
+	for {
+		card, err := dec.Decode()
+		if err != nil {
+			if errors.Is(err, io.EOF) {
+				break
+			}
+			return out, fmt.Errorf("%s: %w", path, err)
+		}
+		e := entryFromCard(card, path)
+		if e.Name == "" && e.First == "" && len(e.Emails) == 0 {
+			continue
+		}
+		out = append(out, e)
+	}
+	return out, nil
+}
+
+func entryFromCard(card vcard.Card, source string) Entry {
+	fn := strings.TrimSpace(card.PreferredValue(vcard.FieldFormattedName))
+	n := card.Name()
+	first, last := "", ""
+	if n != nil {
+		first = strings.TrimSpace(n.GivenName)
+		last = strings.TrimSpace(n.FamilyName)
+	}
+	if fn == "" {
+		fn = strings.TrimSpace(first + " " + last)
+	}
+	if first == "" && last == "" && fn != "" {
+		first, last = SplitDisplayName(fn)
+	}
+	var emails, phones []string
+	for _, v := range card.Values(vcard.FieldEmail) {
+		if e := NormalizeEmail(v); e != "" {
+			emails = append(emails, e)
+		}
+	}
+	for _, v := range card.Values(vcard.FieldTelephone) {
+		p := strings.TrimSpace(v)
+		if p != "" {
+			phones = append(phones, p)
+		}
+	}
+	org := strings.TrimSpace(card.PreferredValue(vcard.FieldOrganization))
+	primary := ""
+	if len(emails) > 0 {
+		primary = emails[0]
+	}
+	return Entry{
+		ID:      EntryID("person", primary, fn),
+		Kind:    "person",
+		Name:    fn,
+		First:   first,
+		Last:    last,
+		Emails:  uniqueEmails(emails),
+		Phones:  uniqueStrings(phones),
+		Org:     org,
+		Sources: []string{source},
+		Zone:    "private",
+		Role:    "unknown",
+		Approve: false,
+		Status:  "new",
+	}
+}
+
+// ScanContactsRoot walks a contacts directory: *.vcf anywhere, person-named
+// subdirs, and *@*.txt email hint files.
+func ScanContactsRoot(root string) (*Document, error) {
+	root = filepath.Clean(root)
+	st, err := os.Stat(root)
+	if err != nil {
+		return nil, err
+	}
+	if !st.IsDir() {
+		return nil, fmt.Errorf("not a directory: %s", root)
+	}
+
+	var entries []Entry
+	seenVCF := map[string]struct{}{}
+
+	err = filepath.WalkDir(root, func(path string, d os.DirEntry, err error) error {
+		if err != nil {
+			return nil // skip unreadable
+		}
+		if d.IsDir() {
+			base := d.Name()
+			if base == ".git" || base == "node_modules" {
+				return filepath.SkipDir
+			}
+			return nil
+		}
+		name := d.Name()
+		lower := strings.ToLower(name)
+		if strings.HasSuffix(lower, ".vcf") {
+			if _, ok := seenVCF[path]; ok {
+				return nil
+			}
+			seenVCF[path] = struct{}{}
+			parsed, perr := ParseVCFFile(path)
+			if perr != nil {
+				entries = append(entries, Entry{
+					ID:      EntryID("person", "", filepath.Base(path)),
+					Kind:    "person",
+					Name:    strings.TrimSuffix(filepath.Base(path), filepath.Ext(path)),
+					Sources: []string{path},
+					Zone:    "private",
+					Role:    "unknown",
+					Notes:   "vcf_parse_error: " + perr.Error(),
+					Status:  "new",
+				})
+				return nil
+			}
+			entries = append(entries, parsed...)
+			return nil
+		}
+		// email hint files: name contains @ and ends with .txt
+		if strings.Contains(name, "@") && strings.HasSuffix(lower, ".txt") {
+			email := NormalizeEmail(strings.TrimSuffix(name, filepath.Ext(name)))
+			if !strings.Contains(email, "@") {
+				return nil
+			}
+			parent := filepath.Base(filepath.Dir(path))
+			first, last := SplitDisplayName(parent)
+			entries = append(entries, Entry{
+				ID:      EntryID("person", email, parent),
+				Kind:    "person",
+				Name:    parent,
+				First:   first,
+				Last:    last,
+				Emails:  []string{email},
+				Sources: []string{path},
+				Zone:    "private",
+				Role:    "unknown",
+				Status:  "new",
+			})
+		}
+		return nil
+	})
+	if err != nil {
+		return nil, err
+	}
+
+	// Top-level person folders without VCF still become stub persons.
+	dents, _ := os.ReadDir(root)
+	for _, d := range dents {
+		if !d.IsDir() {
+			continue
+		}
+		name := d.Name()
+		if name == "Contacts VCF's" || strings.HasPrefix(name, ".") {
+			continue
+		}
+		// skip obvious non-person dumps
+		lower := strings.ToLower(name)
+		if strings.Contains(lower, "vcf") {
+			continue
+		}
+		first, last := SplitDisplayName(name)
+		id := EntryID("person", "", name)
+		entries = append(entries, Entry{
+			ID:      id,
+			Kind:    "person",
+			Name:    name,
+			First:   first,
+			Last:    last,
+			Sources: []string{filepath.Join(root, name)},
+			Zone:    "private",
+			Role:    "unknown",
+			Status:  "new",
+			Notes:   "folder_stub",
+		})
+	}
+
+	doc := MergeDocs(&Document{Entries: entries})
+	return doc, nil
+}
+
+// ReadEmailsFromTxt reads loose emails from a text file (one per line or free text).
+func ReadEmailsFromTxt(path string) ([]string, error) {
+	f, err := os.Open(path)
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+	var out []string
+	sc := bufio.NewScanner(f)
+	for sc.Scan() {
+		line := strings.TrimSpace(sc.Text())
+		if strings.Contains(line, "@") && !strings.Contains(line, " ") {
+			out = append(out, NormalizeEmail(line))
+		}
+	}
+	return uniqueEmails(out), sc.Err()
+}

cmd/oo/catalog.go

@@ -0,0 +1,298 @@
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"os"
+	"path/filepath"
+
+	"github.com/eslider/go-onlyoffice/catalog"
+	"github.com/spf13/cobra"
+)
+
+var catalogCmd = &cobra.Command{
+	Use:   "catalog",
+	Short: "Inventory clients/contacts from disk; match and apply to OO CRM",
+	Long: `Build a YAML catalog of persons/companies from contacts folders and project
+trees, match against live OnlyOffice CRM, then create/update only rows with
+approve: true.
+
+Typical flow:
+  oo catalog scan-contacts --root PATH -O /tmp/contacts.yaml
+  oo catalog scan-projects --root ~/projects -O /tmp/projects.yaml
+  oo catalog scan-thunderbird --root PATH -O /tmp/thunderbird.yaml
+  oo catalog merge -i /tmp/contacts.yaml -i /tmp/projects.yaml -i /tmp/thunderbird.yaml -O docs/catalog/clients-contacts.yaml
+  oo catalog match -i docs/catalog/clients-contacts.yaml
+  # edit approve: true on pilot rows
+  oo catalog apply --dry-run -i docs/catalog/clients-contacts.yaml
+  oo catalog apply --apply -i docs/catalog/clients-contacts.yaml
+`}
+
+func init() {
+	rootCmd.AddCommand(catalogCmd)
+	catalogCmd.AddCommand(catalogScanContactsCmd())
+	catalogCmd.AddCommand(catalogScanProjectsCmd())
+	catalogCmd.AddCommand(catalogScanThunderbirdCmd())
+	catalogCmd.AddCommand(catalogMergeCmd())
+	catalogCmd.AddCommand(catalogMatchCmd())
+	catalogCmd.AddCommand(catalogApplyCmd())
+}
+
+func catalogScanContactsCmd() *cobra.Command {
+	var outPath string
+	cmd := &cobra.Command{
+		Use:   "scan-contacts",
+		Short: "Parse VCF + folder stubs + email txt under a contacts root",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			root, _ := cmd.Flags().GetString("root")
+			if root == "" {
+				return fmt.Errorf("--root is required")
+			}
+			doc, err := catalog.ScanContactsRoot(root)
+			if err != nil {
+				return err
+			}
+			return writeCatalog(cmd, doc, outPath)
+		},
+	}
+	cmd.Flags().String("root", "", "contacts directory (local path)")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "write YAML to this path (default: stdout summary)")
+	_ = cmd.MarkFlagRequired("root")
+	return cmd
+}
+
+func catalogScanProjectsCmd() *cobra.Command {
+	var outPath string
+	var maxDepth int
+	cmd := &cobra.Command{
+		Use:   "scan-projects",
+		Short: "Git roots / remotes / top-level dirs → company rows",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			root, _ := cmd.Flags().GetString("root")
+			if root == "" {
+				return fmt.Errorf("--root is required")
+			}
+			doc, err := catalog.ScanProjectsRoot(root, maxDepth)
+			if err != nil {
+				return err
+			}
+			return writeCatalog(cmd, doc, outPath)
+		},
+	}
+	cmd.Flags().String("root", "", "projects directory (local path)")
+	cmd.Flags().IntVar(&maxDepth, "max-depth", 4, "max directory depth for git roots")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "write YAML to this path")
+	_ = cmd.MarkFlagRequired("root")
+	return cmd
+}
+
+func catalogScanThunderbirdCmd() *cobra.Command {
+	var outPath string
+	var mboxHeaders bool
+	cmd := &cobra.Command{
+		Use:   "scan-thunderbird",
+		Short: "Thunderbird profiles: abook/history.mab + Gloda SQLite contacts",
+		Long: `Walk a directory tree for Thunderbird profiles and extract person rows from:
+  - *.mab address books (email regex)
+  - global-messages-db.sqlite Gloda contacts/identities
+  - optional --mbox-headers: From/To/Cc/Reply-To from mbox folder files (no bodies)
+
+Noisy senders (noreply, Amazon marketplace, GitHub reply, …) are skipped.
+Default zone is private; known work domains (e.g. wheregroup.com) get zone=warm role=work.`,
+		RunE: func(cmd *cobra.Command, args []string) error {
+			root, _ := cmd.Flags().GetString("root")
+			if root == "" {
+				return fmt.Errorf("--root is required")
+			}
+			doc, err := catalog.ScanThunderbirdRootOpts(root, catalog.ScanOptions{MboxHeaders: mboxHeaders})
+			if err != nil {
+				return err
+			}
+			return writeCatalog(cmd, doc, outPath)
+		},
+	}
+	cmd.Flags().String("root", "", "Thunderbird profile or parent directory (local path)")
+	cmd.Flags().BoolVar(&mboxHeaders, "mbox-headers", false, "also extract emails from mbox From/To/Cc headers")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "write YAML to this path")
+	_ = cmd.MarkFlagRequired("root")
+	return cmd
+}
+
+func catalogMergeCmd() *cobra.Command {
+	var inputs []string
+	var outPath string
+	cmd := &cobra.Command{
+		Use:   "merge",
+		Short: "Union + normalize catalog YAML files by entry id",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			if len(inputs) == 0 {
+				return fmt.Errorf("at least one -i/--input required")
+			}
+			var docs []*catalog.Document
+			for _, p := range inputs {
+				d, err := catalog.LoadYAML(p)
+				if err != nil {
+					return err
+				}
+				docs = append(docs, d)
+			}
+			merged := catalog.MergeDocs(docs...)
+			if outPath == "" {
+				return fmt.Errorf("--out is required for merge")
+			}
+			if err := ensureParent(outPath); err != nil {
+				return err
+			}
+			if err := catalog.SaveYAML(outPath, merged); err != nil {
+				return err
+			}
+			fmt.Fprintf(cmd.OutOrStdout(), "merged %d entries → %s\n", len(merged.Entries), outPath)
+			return nil
+		},
+	}
+	cmd.Flags().StringArrayVarP(&inputs, "input", "i", nil, "input catalog YAML (repeatable)")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "output catalog YAML")
+	_ = cmd.MarkFlagRequired("out")
+	return cmd
+}
+
+func catalogMatchCmd() *cobra.Command {
+	var inPath, outPath string
+	cmd := &cobra.Command{
+		Use:   "match",
+		Short: "Diff catalog vs live OO; set status new|exists|conflict and oo_id",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			if inPath == "" {
+				return fmt.Errorf("--input is required")
+			}
+			doc, err := catalog.LoadYAML(inPath)
+			if err != nil {
+				return err
+			}
+			client, err := newOO(cmd)
+			if err != nil {
+				return err
+			}
+			if err := catalog.MatchAgainstOO(cmd.Context(), client, doc); err != nil {
+				return err
+			}
+			if outPath == "" {
+				outPath = inPath
+			}
+			if err := catalog.SaveYAML(outPath, doc); err != nil {
+				return err
+			}
+			counts := map[string]int{}
+			for _, e := range doc.Entries {
+				counts[e.Status]++
+			}
+			fmt.Fprintf(cmd.OutOrStdout(), "matched → %s  new=%d exists=%d conflict=%d\n",
+				outPath, counts["new"], counts["exists"], counts["conflict"])
+			return nil
+		},
+	}
+	cmd.Flags().StringVarP(&inPath, "input", "i", "", "catalog YAML")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "output path (default: overwrite input)")
+	_ = cmd.MarkFlagRequired("input")
+	return cmd
+}
+
+func catalogApplyCmd() *cobra.Command {
+	var inPath, outPath string
+	var dryRun, doApply bool
+	cmd := &cobra.Command{
+		Use:   "apply",
+		Short: "Create/update OO contacts for approve:true rows",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			if !dryRun && !doApply {
+				return fmt.Errorf("pass --dry-run or --apply")
+			}
+			if dryRun && doApply {
+				return fmt.Errorf("use either --dry-run or --apply, not both")
+			}
+			if inPath == "" {
+				return fmt.Errorf("--input is required")
+			}
+			doc, err := catalog.LoadYAML(inPath)
+			if err != nil {
+				return err
+			}
+			client, err := newOO(cmd)
+			if err != nil {
+				return err
+			}
+			res, err := catalog.ApplyApproved(cmd.Context(), client, doc, dryRun)
+			if err != nil {
+				return err
+			}
+			if !dryRun {
+				if outPath == "" {
+					outPath = inPath
+				}
+				if err := catalog.SaveYAML(outPath, doc); err != nil {
+					return err
+				}
+			}
+			if outputFormat == "json" {
+				printJSON(res)
+				return nil
+			}
+			mode := "apply"
+			if dryRun {
+				mode = "dry-run"
+			}
+			fmt.Fprintf(cmd.OutOrStdout(), "%s: created=%d updated=%d skipped=%d errors=%d\n",
+				mode, res.Created, res.Updated, res.Skipped, len(res.Errors))
+			for _, e := range res.Errors {
+				fmt.Fprintf(cmd.ErrOrStderr(), "  error: %s\n", e)
+			}
+			return nil
+		},
+	}
+	cmd.Flags().StringVarP(&inPath, "input", "i", "", "catalog YAML")
+	cmd.Flags().StringVarP(&outPath, "out", "O", "", "write updated catalog after apply (default: overwrite input)")
+	cmd.Flags().BoolVar(&dryRun, "dry-run", false, "count actions without writing to OO")
+	cmd.Flags().BoolVar(&doApply, "apply", false, "perform OO creates/updates")
+	_ = cmd.MarkFlagRequired("input")
+	return cmd
+}
+
+func writeCatalog(cmd *cobra.Command, doc *catalog.Document, outPath string) error {
+	if outPath != "" {
+		if err := ensureParent(outPath); err != nil {
+			return err
+		}
+		if err := catalog.SaveYAML(outPath, doc); err != nil {
+			return err
+		}
+		fmt.Fprintf(cmd.OutOrStdout(), "wrote %d entries → %s\n", len(doc.Entries), outPath)
+		return nil
+	}
+	if outputFormat == "json" {
+		b, err := json.MarshalIndent(doc, "", "  ")
+		if err != nil {
+			return err
+		}
+		fmt.Fprintln(cmd.OutOrStdout(), string(b))
+		return nil
+	}
+	persons, companies := 0, 0
+	for _, e := range doc.Entries {
+		if e.Kind == "company" {
+			companies++
+		} else {
+			persons++
+		}
+	}
+	fmt.Fprintf(cmd.OutOrStdout(), "entries=%d persons=%d companies=%d (pass -O PATH to write YAML)\n",
+		len(doc.Entries), persons, companies)
+	return nil
+}
+
+func ensureParent(path string) error {
+	dir := filepath.Dir(path)
+	if dir == "" || dir == "." {
+		return nil
+	}
+	return os.MkdirAll(dir, 0o755)
+}

cmd/oo/cli_test.go

@@ -11,7 +11,7 @@ func TestRootRegistersSubjects(t *testing.T) {
 	want := []string{
 		"calendar", "projects", "tasks", "users", "whoami",
 		"contacts", "persons", "companies",
-		"opportunities", "cases", "crm-tasks", "applications", "crm", "mails",
+		"opportunities", "cases", "crm-tasks", "applications", "crm", "mails", "catalog",
 	}
 	got := make(map[string]bool, len(rootCmd.Commands()))
 	for _, c := range rootCmd.Commands() {

cmd/oo/common.go

@@ -20,7 +20,7 @@ var outputFormat = "table"
 var rootCmd = &cobra.Command{
 	Use:           "oo",
 	Short:         "OnlyOffice Workspace CLI — subject-based command tree",
-	Long:          "oo is a thin CLI over github.com/eslider/go-onlyoffice.\nCommands are grouped by OnlyOffice subject (calendar, projects, tasks, users, persons, companies, opportunities, cases, crm-tasks, applications).",
+	Long:          "oo is a thin CLI over github.com/eslider/go-onlyoffice.\nCommands are grouped by OnlyOffice subject (calendar, projects, tasks, users, persons, companies, opportunities, cases, crm-tasks, applications, catalog).",
 	Version:       version,
 	SilenceUsage:  true,
 	SilenceErrors: false,

cmd/oo/main.go

@@ -15,6 +15,7 @@
 //	oo crm           cleanup
 //	oo mails         accounts | folders | list | get | delete
 //	oo applications  sync
+//	oo catalog       scan-contacts | scan-projects | scan-thunderbird | merge | match | apply
 //
 // Every list supports `--output/-o json|table` (table is the default).
 //

crm.go

@@ -46,9 +46,12 @@ func (c *Client) GetContact(ctx context.Context, contactID string) (map[string]a
 }
 
 // FindCompany searches for a company contact with an exact (case-insensitive)
-// displayName match. Returns nil when not found.
+// displayName match after slogan/legal-suffix normalization. Returns nil when not found.
 func (c *Client) FindCompany(ctx context.Context, name string) (map[string]any, error) {
 	needle := CompanyGroupingKey(name)
+	if needle == "" {
+		return nil, nil
+	}
 	const page = 50
 	for start := 0; ; start += page {
 		items, total, err := c.ListContacts(ctx, page, start, name)
@@ -67,6 +70,19 @@ func (c *Client) FindCompany(ctx context.Context, name string) (map[string]any,
 			break
 		}
 	}
+	// Search filter may miss longer legal names; fall back to a full company scan.
+	all, err := c.ListAllContacts(ctx)
+	if err != nil {
+		return nil, err
+	}
+	for _, co := range all {
+		if !isCompany(co) {
+			continue
+		}
+		if CompanyGroupingKey(fmt.Sprint(co["displayName"])) == needle {
+			return co, nil
+		}
+	}
 	return nil, nil
 }
 
@@ -96,6 +112,53 @@ func (c *Client) FindPerson(ctx context.Context, first, last string) (map[string
 	return nil, nil
 }
 
+// FindPersonByEmail finds a person whose primary or contact-info email matches
+// (case-insensitive). Companies are skipped. Returns nil when not found.
+func (c *Client) FindPersonByEmail(ctx context.Context, email string) (map[string]any, error) {
+	needle := strings.ToLower(strings.TrimSpace(email))
+	if needle == "" {
+		return nil, nil
+	}
+	all, err := c.ListAllContacts(ctx)
+	if err != nil {
+		return nil, err
+	}
+	for _, p := range all {
+		if isCompany(p) {
+			continue
+		}
+		for _, em := range contactEmailValues(p) {
+			if strings.ToLower(strings.TrimSpace(em)) == needle {
+				return p, nil
+			}
+		}
+	}
+	return nil, nil
+}
+
+func contactEmailValues(p map[string]any) []string {
+	var out []string
+	for _, key := range []string{"email", "primaryEmail"} {
+		if v := strings.TrimSpace(fmt.Sprint(p[key])); v != "" && v != "<nil>" {
+			out = append(out, v)
+		}
+	}
+	for _, row := range ContactInfoRows(p) {
+		t := strings.ToLower(fmt.Sprint(row["infoType"]))
+		if t != "email" {
+			continue
+		}
+		data := strings.TrimSpace(fmt.Sprint(row["data"]))
+		if data == "" || data == "<nil>" {
+			data = strings.TrimSpace(fmt.Sprint(row["value"]))
+		}
+		if data != "" {
+			out = append(out, data)
+		}
+	}
+	return out
+}
+
 func isCompany(m map[string]any) bool {
 	v, ok := m["isCompany"].(bool)
 	return ok && v

crm_normalize.go

@@ -10,6 +10,9 @@ var multiSpace = regexp.MustCompile(`\s+`)
 // sloganSeparators split a company name from a trailing tagline/slogan.
 var sloganSeparators = []string{" — ", " – ", " - ", "—", "–"}
 
+// companyLegalSuffix matches trailing legal-form tokens (GmbH, & Co. KG, …).
+var companyLegalSuffix = regexp.MustCompile(`(?i)[\s,]*(&\s*co\.?\s*kg\.?|co\.?\s*kg\.?|gmbh|ag|ltd\.?|llc\.?|inc\.?|plc\.?|s\.?l\.?|sarl|b\.?v\.?)\s*$`)
+
 // StripSloganSuffix returns the part before an em/en dash tagline, e.g.
 // "Affirm — Fraud Engineering" → "Affirm".
 func StripSloganSuffix(s string) string {
@@ -22,9 +25,23 @@ func StripSloganSuffix(s string) string {
 	return s
 }
 
-// CompanyGroupingKey normalizes a company name for dedupe (ignores slogans).
+// StripCompanyLegalSuffix removes trailing GmbH / & Co. KG / Ltd / … tokens.
+func StripCompanyLegalSuffix(s string) string {
+	s = strings.TrimSpace(s)
+	s = strings.TrimRight(s, ",; ")
+	for {
+		next := companyLegalSuffix.ReplaceAllString(s, "")
+		next = strings.TrimRight(strings.TrimSpace(next), ",; ")
+		if next == s || next == "" {
+			return s
+		}
+		s = next
+	}
+}
+
+// CompanyGroupingKey normalizes a company name for dedupe (ignores slogans and legal suffixes).
 func CompanyGroupingKey(s string) string {
-	return NormalizeCompanyName(StripSloganSuffix(s))
+	return NormalizeCompanyName(StripCompanyLegalSuffix(StripSloganSuffix(s)))
 }
 
 // NormalizeCompanyName lowercases and collapses whitespace for grouping.

crm_normalize_test.go

@@ -65,6 +65,10 @@ func TestCompanyGroupingKey(t *testing.T) {
 	if a != b {
 		t.Fatalf("%q != %q", a, b)
 	}
+	if CompanyGroupingKey("WhereGroup") != CompanyGroupingKey("WhereGroup GmbH & Co. KG,") {
+		t.Fatalf("legal suffix should not change grouping key: %q vs %q",
+			CompanyGroupingKey("WhereGroup"), CompanyGroupingKey("WhereGroup GmbH & Co. KG,"))
+	}
 }
 
 func TestOpportunityTitlesMatchSlogan(t *testing.T) {

go.mod

@@ -9,12 +9,15 @@ require (
 	github.com/charmbracelet/glamour v0.8.0
 	github.com/charmbracelet/lipgloss v0.12.1
 	github.com/charmbracelet/x/ansi v0.1.4
+	github.com/emersion/go-vcard v0.0.0-20260618161152-d854b7e0e2d3
 	github.com/eslider/go-xls/v2 v2.1.0
 	github.com/google/go-querystring v1.2.0
 	github.com/joho/godotenv v1.5.1
 	github.com/mattn/go-runewidth v0.0.15
 	github.com/muesli/termenv v0.16.0
 	github.com/spf13/cobra v1.10.2
+	gopkg.in/yaml.v3 v3.0.1
+	modernc.org/sqlite v1.54.0
 )
 
 require (
@@ -25,6 +28,8 @@ require (
 	github.com/aymerick/douceur v0.2.0 // indirect
 	github.com/containerd/console v1.0.4-0.20230313162750-1ae8d489ac81 // indirect
 	github.com/dlclark/regexp2 v1.11.0 // indirect
+	github.com/dustin/go-humanize v1.0.1 // indirect
+	github.com/google/uuid v1.6.0 // indirect
 	github.com/gorilla/css v1.0.1 // indirect
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
 	github.com/lucasb-eyer/go-colorful v1.4.0 // indirect
@@ -34,13 +39,18 @@ require (
 	github.com/muesli/ansi v0.0.0-20211018074035-2e021307bc4b // indirect
 	github.com/muesli/cancelreader v0.2.2 // indirect
 	github.com/muesli/reflow v0.3.0 // indirect
+	github.com/ncruces/go-strftime v1.0.0 // indirect
+	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
 	github.com/rivo/uniseg v0.4.7 // indirect
 	github.com/spf13/pflag v1.0.9 // indirect
 	github.com/yuin/goldmark v1.8.2 // indirect
 	github.com/yuin/goldmark-emoji v1.0.3 // indirect
 	golang.org/x/net v0.55.0 // indirect
-	golang.org/x/sync v0.20.0 // indirect
-	golang.org/x/sys v0.45.0 // indirect
+	golang.org/x/sync v0.21.0 // indirect
+	golang.org/x/sys v0.46.0 // indirect
 	golang.org/x/term v0.43.0 // indirect
 	golang.org/x/text v0.37.0 // indirect
+	modernc.org/libc v1.74.1 // indirect
+	modernc.org/mathutil v1.7.1 // indirect
+	modernc.org/memory v1.11.0 // indirect
 )

go.sum

@@ -33,14 +33,24 @@ github.com/containerd/console v1.0.4-0.20230313162750-1ae8d489ac81/go.mod h1:Yyn
 github.com/cpuguy83/go-md2man/v2 v2.0.6/go.mod h1:oOW0eioCTA6cOiMLiUPZOpcVxMig6NIQQ7OS05n1F4g=
 github.com/dlclark/regexp2 v1.11.0 h1:G/nrcoOa7ZXlpoa/91N3X7mM3r8eIlMBBJZvsz/mxKI=
 github.com/dlclark/regexp2 v1.11.0/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
+github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkpeCY=
+github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
+github.com/emersion/go-vcard v0.0.0-20260618161152-d854b7e0e2d3 h1:B9YK+Tck5mTccyDhtxBzWyqGYcFxLyB6+noMNW4/VgI=
+github.com/emersion/go-vcard v0.0.0-20260618161152-d854b7e0e2d3/go.mod h1:HMJKR5wlh/ziNp+sHEDV2ltblO4JD2+IdDOWtGcQBTM=
 github.com/eslider/go-xls/v2 v2.1.0 h1:HszWKqYQbXxACmAXXWdMsfNl1NDBfGVBnJUPtyUHQ7A=
 github.com/eslider/go-xls/v2 v2.1.0/go.mod h1:xgxO6JrfuBr9jGUB+0z5l/yDmFFZ5diGk0ATGihxlMU=
 github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
 github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 github.com/google/go-querystring v1.2.0 h1:yhqkPbu2/OH+V9BfpCVPZkNmUXhb2gBxJArfhIxNtP0=
 github.com/google/go-querystring v1.2.0/go.mod h1:8IFJqpSRITyJ8QhQ13bmbeMBDfmeEJZD5A0egEOmkqU=
+github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e h1:ijClszYn+mADRFY17kjQEVQ1XRhq2/JR1M3sGqeJoxs=
+github.com/google/pprof v0.0.0-20250317173921-a4b03ec1a45e/go.mod h1:boTsfXsheKC2y+lKOCMpSfarhxDeIzfZG1jqGcPl3cA=
+github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
+github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/gorilla/css v1.0.1 h1:ntNaBIghp6JmvWnxbZKANoLyuXTPZ4cAMlo6RyhlbO8=
 github.com/gorilla/css v1.0.1/go.mod h1:BvnYkspnSzMmwRK+b8/xgNPLiIuNZr6vbZBTPQ2A3b0=
+github.com/hashicorp/golang-lru/v2 v2.0.7 h1:a+bsQ5rvGLjzHuww6tVxozPZFVghXaHOwFs4luLUK2k=
+github.com/hashicorp/golang-lru/v2 v2.0.7/go.mod h1:QeFd9opnmA6QUJc5vARoKUSoFhyfM2/ZepoAG6RGpeM=
 github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
 github.com/hexops/gotextdiff v1.0.3/go.mod h1:pSWU5MAI3yDq+fZBTazCSJysOMbxWL1BSow5/V2vxeg=
 github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
@@ -66,8 +76,12 @@ github.com/muesli/reflow v0.3.0 h1:IFsN6K9NfGtjeggFP+68I4chLZV2yIKsXJFNZ+eWh6s=
 github.com/muesli/reflow v0.3.0/go.mod h1:pbwTDkVPibjO2kyvBQRBxTWEEGDGq0FlB1BIKtnHY/8=
 github.com/muesli/termenv v0.16.0 h1:S5AlUN9dENB57rsbnkPyfdGuWIlkmzJjbFf0Tf5FWUc=
 github.com/muesli/termenv v0.16.0/go.mod h1:ZRfOIKPFDYQoDFF4Olj7/QJbW60Ol/kL1pU3VfY/Cnk=
+github.com/ncruces/go-strftime v1.0.0 h1:HMFp8mLCTPp341M/ZnA4qaf7ZlsbTc+miZjCLOFAw7w=
+github.com/ncruces/go-strftime v1.0.0/go.mod h1:Fwc5htZGVVkseilnfgOVb9mKy6w1naJmn9CehxcKcls=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec h1:W09IVJc94icq4NjY3clb7Lk8O1qJ8BdBEF8z0ibU0rE=
+github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec/go.mod h1:qqbHyh8v60DhA7CoWK5oRCqLrMHRGoxYCSS9EjAz6Eo=
 github.com/rivo/uniseg v0.1.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
 github.com/rivo/uniseg v0.2.0/go.mod h1:J6wj4VEh+S6ZtnVlnTBMWIodfgj8LQOQFoIToxlJtxc=
 github.com/rivo/uniseg v0.4.7 h1:WUdvkW8uEhrYfLC4ZzdpI2ztxP1I582+49Oc5Mq64VQ=
@@ -87,15 +101,50 @@ github.com/yuin/goldmark v1.8.2/go.mod h1:ip/1k0VRfGynBgxOz0yCqHrbZXhcjxyuS66Brc
 github.com/yuin/goldmark-emoji v1.0.3 h1:aLRkLHOuBR2czCY4R8olwMjID+tENfhyFDMCRhbIQY4=
 github.com/yuin/goldmark-emoji v1.0.3/go.mod h1:tTkZEbwu5wkPmgTcitqddVxY9osFZiavD+r4AzQrh1U=
 go.yaml.in/yaml/v3 v3.0.4/go.mod h1:DhzuOOF2ATzADvBadXxruRBLzYTpT36CKvDb3+aBEFg=
+golang.org/x/mod v0.37.0 h1:vF1DjpVEshcIqoEaauuHebaLk1O1forxjxBaVn884JQ=
+golang.org/x/mod v0.37.0/go.mod h1:m8S8VeM9r4dzDwjrKO0a1sZP3YjeMamRRlD+fmR2Q/0=
 golang.org/x/net v0.55.0 h1:bcvxaJn3e1U6InsFWt1JUq1aSjnRxLzT2rtD2KfkDF8=
 golang.org/x/net v0.55.0/go.mod h1:L5U2KuzuOe1lY7Z+aWVIKK6qEeJXnXV9yzGA+WCHJww=
-golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
-golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
+golang.org/x/sync v0.21.0 h1:HLII4xRRTtCRkxYp4HNFF0Js/Og6q2i++KXbg0gHCwM=
+golang.org/x/sync v0.21.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
 golang.org/x/sys v0.1.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.45.0 h1:dO4czNzziLiiXplLQgBCEpCvXQ3dnkn0SdaZSYdQ+FY=
-golang.org/x/sys v0.45.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
+golang.org/x/sys v0.46.0 h1:noSf2Fq6F8DBgS+LysIkx7rIExoNHJsxOAtPp4rthXw=
+golang.org/x/sys v0.46.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.43.0 h1:S4RLU2sB31O/NCl+zFN9Aru9A/Cq2aqKpTZJ6B+DwT4=
 golang.org/x/term v0.43.0/go.mod h1:lrhlHNdQJHO+1qVYiHfFKVuVioJIheAc3fBSMFYEIsk=
 golang.org/x/text v0.37.0 h1:Cqjiwd9eSg8e0QAkyCaQTNHFIIzWtidPahFWR83rTrc=
 golang.org/x/text v0.37.0/go.mod h1:a5sjxXGs9hsn/AJVwuElvCAo9v8QYLzvavO5z2PiM38=
+golang.org/x/tools v0.47.0 h1:7Kn5x/d1svx/PzryTsqeoZN4TZwqeH5pGWjefhLi/1Q=
+golang.org/x/tools v0.47.0/go.mod h1:dFHnyTvFWY212G+h7ZY4Vsp/K3U4/7W9TyVaAul8uCA=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
+gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
+modernc.org/cc/v4 v4.29.0 h1:CXgwL8cvxmyzBQZzbSl/6xFtMCryb6u8IOqDci39cgc=
+modernc.org/cc/v4 v4.29.0/go.mod h1:OnovgIhbbMXMu1aISnJ0wvVD1KnW+cAUJkIrAWh+kVI=
+modernc.org/ccgo/v4 v4.34.6 h1:sBgfIwyN0TQ9C5hwIeuqyeAKyMWnbvj2fvpF4L11uzU=
+modernc.org/ccgo/v4 v4.34.6/go.mod h1:SZ8YcN9NG7XVsQYdm6jYBvi8PQP1qi+kqB6OhjqI3Fk=
+modernc.org/fileutil v1.4.0 h1:j6ZzNTftVS054gi281TyLjHPp6CPHr2KCxEXjEbD6SM=
+modernc.org/fileutil v1.4.0/go.mod h1:EqdKFDxiByqxLk8ozOxObDSfcVOv/54xDs/DUHdvCUU=
+modernc.org/gc/v2 v2.6.5 h1:nyqdV8q46KvTpZlsw66kWqwXRHdjIlJOhG6kxiV/9xI=
+modernc.org/gc/v2 v2.6.5/go.mod h1:YgIahr1ypgfe7chRuJi2gD7DBQiKSLMPgBQe9oIiito=
+modernc.org/gc/v3 v3.1.4 h1:2g65LGVSmFQrXeITAw97x7hCRvZFcyE1uDP+7Vng7JI=
+modernc.org/gc/v3 v3.1.4/go.mod h1:HFK/6AGESC7Ex+EZJhJ2Gni6cTaYpSMmU/cT9RmlfYY=
+modernc.org/goabi0 v0.2.0 h1:HvEowk7LxcPd0eq6mVOAEMai46V+i7Jrj13t4AzuNks=
+modernc.org/goabi0 v0.2.0/go.mod h1:CEFRnnJhKvWT1c1JTI3Avm+tgOWbkOu5oPA8eH8LnMI=
+modernc.org/libc v1.74.1 h1:bdR4VTKFMC4966QSNZ05XLGI/VwzVa2kTUX51Dm0riQ=
+modernc.org/libc v1.74.1/go.mod h1:uH4t5bOx3G3g9Xcmj10YKlTcVISlRDwv8VoQJG9n8Os=
+modernc.org/mathutil v1.7.1 h1:GCZVGXdaN8gTqB1Mf/usp1Y/hSqgI2vAGGP4jZMCxOU=
+modernc.org/mathutil v1.7.1/go.mod h1:4p5IwJITfppl0G4sUEDtCr4DthTaT47/N3aT6MhfgJg=
+modernc.org/memory v1.11.0 h1:o4QC8aMQzmcwCK3t3Ux/ZHmwFPzE6hf2Y5LbkRs+hbI=
+modernc.org/memory v1.11.0/go.mod h1:/JP4VbVC+K5sU2wZi9bHoq2MAkCnrt2r98UGeSK7Mjw=
+modernc.org/opt v0.2.0 h1:tGyef5ApycA7FSEOMraay9SaTk5zmbx7Tu+cJs4QKZg=
+modernc.org/opt v0.2.0/go.mod h1:03fq9lsNfvkYSfxrfUhZCWPk1lm4cq4N+Bh//bEtgns=
+modernc.org/sortutil v1.2.1 h1:+xyoGf15mM3NMlPDnFqrteY07klSFxLElE2PVuWIJ7w=
+modernc.org/sortutil v1.2.1/go.mod h1:7ZI3a3REbai7gzCLcotuw9AC4VZVpYMjDzETGsSMqJE=
+modernc.org/sqlite v1.54.0 h1:JCxR4qwkJvOaqAoYcgDoO25Nc+ROg6EJ2LfBVzdrgog=
+modernc.org/sqlite v1.54.0/go.mod h1:4ntCLuNmnH8+GNqjka1wNg7KJd5/Hi5FYp8K+XQ7GZw=
+modernc.org/strutil v1.2.1 h1:UneZBkQA+DX2Rp35KcM69cSsNES9ly8mQWD71HKlOA0=
+modernc.org/strutil v1.2.1/go.mod h1:EHkiggD70koQxjVdSBM3JKM7k6L0FbGE5eymy9i3B9A=
+modernc.org/token v1.1.0 h1:Xl7Ap9dKaEs5kLoOQeQmPWevfnk/DM5qcLcYlA8ys6Y=
+modernc.org/token v1.1.0/go.mod h1:UGzOrNV1mAFSEB63lOFHIpNRUVMvYTc6yu1SMY/XTDM=