feat(backend): drain import folder and keep mtime on server-side import

The directory import now removes each source file after it is safely
ingested, so the import folder drains and re-running doesn't create
duplicates (a removal failure is reported per-file but doesn't undo the
import). It also captures the source file's mtime and passes it as a new
ContentDatetimeFallback on Upload, used for content_datetime only when
the file has no EXIF date — so non-photo files keep a meaningful date
instead of the zero value once the source is gone. Adds an integration
test covering ingest, directory skip, source removal and the mtime
fallback.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
2026-06-11 18:19:59 +03:00
parent bce79867e4
commit 1b04d67e20
2 changed files with 95 additions and 13 deletions
+63 -7
View File
@@ -50,9 +50,10 @@ const defaultAdminDSN = "host=/var/run/postgresql port=5434 user=h1k0 dbname=pos
// --------------------------------------------------------------------------- // ---------------------------------------------------------------------------
type harness struct { type harness struct {
t *testing.T t *testing.T
server *httptest.Server server *httptest.Server
client *http.Client client *http.Client
importDir string
} }
// setupSuite creates an ephemeral database, runs migrations, wires the full // setupSuite creates an ephemeral database, runs migrations, wires the full
@@ -106,6 +107,7 @@ func setupSuite(t *testing.T) *harness {
// --- Temp directories for storage ---------------------------------------- // --- Temp directories for storage ----------------------------------------
filesDir := t.TempDir() filesDir := t.TempDir()
thumbsDir := t.TempDir() thumbsDir := t.TempDir()
importDir := t.TempDir()
diskStorage, err := storage.NewDiskStorage(filesDir, thumbsDir, 160, 160, 1920, 1080) diskStorage, err := storage.NewDiskStorage(filesDir, thumbsDir, 160, 160, 1920, 1080)
require.NoError(t, err) require.NoError(t, err)
@@ -130,7 +132,7 @@ func setupSuite(t *testing.T) *harness {
tagSvc := service.NewTagService(tagRepo, tagRuleRepo, aclSvc, auditSvc, transactor) tagSvc := service.NewTagService(tagRepo, tagRuleRepo, aclSvc, auditSvc, transactor)
categorySvc := service.NewCategoryService(categoryRepo, tagRepo, aclSvc, auditSvc) categorySvc := service.NewCategoryService(categoryRepo, tagRepo, aclSvc, auditSvc)
poolSvc := service.NewPoolService(poolRepo, aclSvc, auditSvc) poolSvc := service.NewPoolService(poolRepo, aclSvc, auditSvc)
fileSvc := service.NewFileService(fileRepo, mimeRepo, diskStorage, aclSvc, auditSvc, tagSvc, transactor, filesDir) fileSvc := service.NewFileService(fileRepo, mimeRepo, diskStorage, aclSvc, auditSvc, tagSvc, transactor, importDir)
userSvc := service.NewUserService(userRepo, sessionRepo, auditSvc) userSvc := service.NewUserService(userRepo, sessionRepo, auditSvc)
// Bootstrap the admin account the suite logs in with (replaces the old // Bootstrap the admin account the suite logs in with (replaces the old
@@ -159,9 +161,10 @@ func setupSuite(t *testing.T) *harness {
t.Cleanup(srv.Close) t.Cleanup(srv.Close)
return &harness{ return &harness{
t: t, t: t,
server: srv, server: srv,
client: srv.Client(), client: srv.Client(),
importDir: importDir,
} }
} }
@@ -899,6 +902,59 @@ func TestNonOwnerAccessControl(t *testing.T) {
assert.Equal(t, http.StatusOK, resp.StatusCode, resp.String()) assert.Equal(t, http.StatusOK, resp.StatusCode, resp.String())
} }
// TestImportFromFolder verifies the admin server-side import: supported files
// are ingested, subdirectories are skipped, the source is removed from the
// import folder afterwards, and a file without EXIF takes the source's mtime as
// its content_datetime.
func TestImportFromFolder(t *testing.T) {
if testing.Short() {
t.Skip("skipping integration test in short mode")
}
h := setupSuite(t)
adminToken := h.login("admin", "admin")
// Drop a non-EXIF JPEG into the import folder with a known mtime, plus a
// subdirectory that must be skipped.
srcPath := filepath.Join(h.importDir, "scan.jpg")
require.NoError(t, os.WriteFile(srcPath, minimalJPEG(), 0o644))
mtime := time.Date(2021, 3, 4, 5, 6, 7, 0, time.UTC)
require.NoError(t, os.Chtimes(srcPath, mtime, mtime))
require.NoError(t, os.Mkdir(filepath.Join(h.importDir, "nested"), 0o755))
resp := h.doJSON("POST", "/files/import", map[string]any{}, adminToken)
require.Equal(t, http.StatusOK, resp.StatusCode, resp.String())
var res struct {
Imported int `json:"imported"`
Skipped int `json:"skipped"`
Errors []struct {
Filename string `json:"filename"`
Reason string `json:"reason"`
} `json:"errors"`
}
resp.decode(t, &res)
assert.Equal(t, 1, res.Imported, resp.String())
assert.Equal(t, 1, res.Skipped, resp.String()) // the nested directory
assert.Empty(t, res.Errors, resp.String())
// Source file is gone from the import folder after a successful import.
_, statErr := os.Stat(srcPath)
assert.True(t, os.IsNotExist(statErr), "source should be removed after import")
// The imported file took the mtime as content_datetime (no EXIF present).
listResp := h.doJSON("GET", "/files?limit=10", nil, adminToken)
require.Equal(t, http.StatusOK, listResp.StatusCode, listResp.String())
var list struct {
Items []struct {
ContentDatetime string `json:"content_datetime"`
} `json:"items"`
}
listResp.decode(t, &list)
require.Len(t, list.Items, 1, listResp.String())
ct, err := time.Parse(time.RFC3339, list.Items[0].ContentDatetime)
require.NoError(t, err)
assert.True(t, ct.Equal(mtime), "content_datetime %v should equal mtime %v", ct, mtime)
}
// TestBlockRevokesActiveSessions verifies that blocking a user immediately // TestBlockRevokesActiveSessions verifies that blocking a user immediately
// invalidates their outstanding access tokens. // invalidates their outstanding access tokens.
func TestBlockRevokesActiveSessions(t *testing.T) { func TestBlockRevokesActiveSessions(t *testing.T) {
+32 -6
View File
@@ -33,8 +33,12 @@ type UploadParams struct {
Notes *string Notes *string
Metadata json.RawMessage Metadata json.RawMessage
ContentDatetime *time.Time ContentDatetime *time.Time
IsPublic bool // ContentDatetimeFallback is used for content_datetime only when neither an
TagIDs []uuid.UUID // explicit ContentDatetime nor an EXIF date is available (e.g. the source
// file's mtime on a server-side import).
ContentDatetimeFallback *time.Time
IsPublic bool
TagIDs []uuid.UUID
} }
// UpdateParams holds the parameters for updating file metadata. // UpdateParams holds the parameters for updating file metadata.
@@ -128,12 +132,14 @@ func (s *FileService) Upload(ctx context.Context, p UploadParams) (*domain.File,
// Extract EXIF metadata (best-effort; non-image files will error silently). // Extract EXIF metadata (best-effort; non-image files will error silently).
exifData, exifDatetime := extractEXIFWithDatetime(data) exifData, exifDatetime := extractEXIFWithDatetime(data)
// Resolve content datetime: explicit > EXIF > zero value. // Resolve content datetime: explicit > EXIF > fallback (e.g. import mtime) > zero.
var contentDatetime time.Time var contentDatetime time.Time
if p.ContentDatetime != nil { if p.ContentDatetime != nil {
contentDatetime = *p.ContentDatetime contentDatetime = *p.ContentDatetime
} else if exifDatetime != nil { } else if exifDatetime != nil {
contentDatetime = *exifDatetime contentDatetime = *exifDatetime
} else if p.ContentDatetimeFallback != nil {
contentDatetime = *p.ContentDatetimeFallback
} }
// Assign UUID v7 so CreatedAt can be derived from it later. // Assign UUID v7 so CreatedAt can be derived from it later.
@@ -584,11 +590,21 @@ func (s *FileService) Import(ctx context.Context, path string) (*ImportResult, e
continue continue
} }
// Preserve the file's mtime as a content_datetime fallback (used only when
// the file has no EXIF date) — once the source is removed below it's the
// only date left for non-photo files.
var mtime *time.Time
if info, statErr := entry.Info(); statErr == nil {
t := info.ModTime()
mtime = &t
}
name := entry.Name() name := entry.Name()
_, uploadErr := s.Upload(ctx, UploadParams{ _, uploadErr := s.Upload(ctx, UploadParams{
Reader: f, Reader: f,
MIMEType: mimeStr, MIMEType: mimeStr,
OriginalName: &name, OriginalName: &name,
ContentDatetimeFallback: mtime,
}) })
f.Close() f.Close()
@@ -600,6 +616,16 @@ func (s *FileService) Import(ctx context.Context, path string) (*ImportResult, e
continue continue
} }
result.Imported++ result.Imported++
// Remove the source on success so the import folder drains and re-running
// doesn't duplicate. The file is already safely copied into storage; a
// removal failure is reported but doesn't undo the import.
if rmErr := os.Remove(fullPath); rmErr != nil {
result.Errors = append(result.Errors, ImportFileError{
Filename: entry.Name(),
Reason: fmt.Sprintf("imported, but failed to remove source: %s", rmErr),
})
}
} }
return result, nil return result, nil