106 lines
3.0 KiB
Go
106 lines
3.0 KiB
Go
package directory
|
|
|
|
import (
|
|
"context"
|
|
"encoding/csv"
|
|
"errors"
|
|
"fmt"
|
|
"io"
|
|
"strings"
|
|
"time"
|
|
)
|
|
|
|
var requiredCSVHeaders = []string{"user_id", "name", "dept_name"}
|
|
|
|
// ImportCSV validates the complete snapshot before making one repository call.
|
|
// It consumes the tested Python sync output and accepts optional production
|
|
// enrichment columns without reimplementing the bulk DingTalk crawler in Go.
|
|
func ImportCSV(ctx context.Context, tenantKey string, input io.Reader, writer SnapshotWriter) (int, error) {
|
|
tenantKey = strings.TrimSpace(tenantKey)
|
|
if tenantKey == "" || input == nil || writer == nil {
|
|
return 0, errors.New("tenant, CSV input and writer are required")
|
|
}
|
|
r := csv.NewReader(input)
|
|
r.FieldsPerRecord = -1
|
|
header, err := r.Read()
|
|
if err != nil {
|
|
return 0, fmt.Errorf("read directory CSV header: %w", err)
|
|
}
|
|
columns := make(map[string]int, len(header))
|
|
for i, name := range header {
|
|
name = strings.TrimSpace(strings.TrimPrefix(name, "\ufeff"))
|
|
columns[name] = i
|
|
}
|
|
for _, name := range requiredCSVHeaders {
|
|
if _, ok := columns[name]; !ok {
|
|
return 0, fmt.Errorf("directory CSV missing required header %q", name)
|
|
}
|
|
}
|
|
|
|
profiles := make([]Profile, 0, 256)
|
|
line := 1
|
|
for {
|
|
record, readErr := r.Read()
|
|
if errors.Is(readErr, io.EOF) {
|
|
break
|
|
}
|
|
line++
|
|
if readErr != nil {
|
|
return 0, fmt.Errorf("read directory CSV line %d: %w", line, readErr)
|
|
}
|
|
value := func(name string) string {
|
|
idx, ok := columns[name]
|
|
if !ok || idx >= len(record) {
|
|
return ""
|
|
}
|
|
return strings.TrimSpace(record[idx])
|
|
}
|
|
staffID := value("user_id")
|
|
name := value("name")
|
|
if staffID == "" || name == "" {
|
|
return 0, fmt.Errorf("directory CSV line %d has empty user_id or name", line)
|
|
}
|
|
profiles = append(profiles, Profile{
|
|
StaffID: staffID,
|
|
UnionID: value("union_id"),
|
|
Name: name,
|
|
Nickname: value("nickname"),
|
|
JobTitle: value("title"),
|
|
DepartmentIDs: splitDepartmentIDs(value("department_ids")),
|
|
DepartmentName: value("dept_name"),
|
|
MainDepartmentName: value("main_department_name"),
|
|
Region: value("region"),
|
|
EmploymentStatus: value("employment_status"),
|
|
Source: SourceFullSync,
|
|
SyncedAt: parseSyncTime(value("sync_time")),
|
|
})
|
|
}
|
|
if len(profiles) == 0 {
|
|
return 0, errors.New("directory CSV snapshot is empty")
|
|
}
|
|
if err := writer.UpsertDirectoryProfiles(ctx, tenantKey, profiles); err != nil {
|
|
return 0, fmt.Errorf("write directory snapshot: %w", err)
|
|
}
|
|
return len(profiles), nil
|
|
}
|
|
|
|
func splitDepartmentIDs(raw string) []string {
|
|
parts := strings.FieldsFunc(raw, func(r rune) bool { return r == '/' || r == ',' || r == ';' })
|
|
out := make([]string, 0, len(parts))
|
|
for _, part := range parts {
|
|
if part = strings.TrimSpace(part); part != "" {
|
|
out = append(out, part)
|
|
}
|
|
}
|
|
return out
|
|
}
|
|
|
|
func parseSyncTime(raw string) time.Time {
|
|
for _, layout := range []string{time.RFC3339, "2006-01-02 15:04:05"} {
|
|
if parsed, err := time.Parse(layout, raw); err == nil {
|
|
return parsed
|
|
}
|
|
}
|
|
return time.Time{}
|
|
}
|