Importing Google Photos
Smriti can migrate a Google Photos library directly from Google Takeout ZIP files. You do not need to unzip, combine or rearrange the export first.
Before importing
In Google Takeout, export Google Photos as ZIP files. Large exports are split into numbered ZIPs. Download every part from the same export; Smriti reads them as one set because a media file and its supplemental JSON metadata can be placed in different parts.
Keep enough free space in the destination library for the uncompressed unique originals. Smriti checks this before writing.
Start the import
From Welcome:
- Choose Import Google Photos, or drop all Takeout ZIPs together.
- Select every ZIP from the export.
- Choose the folder or drive that will become the Smriti library.
From an open library:
- Open Settings → Library → Google Photos.
- Choose Import Takeout ZIPs.
- Select every ZIP from the export.
That is the whole migration. Progress remains visible while you browse the app.
What Smriti preserves
- Each byte-identical original is stored once, even when Takeout repeats it in year and album folders.
- Distinct edited versions remain distinct files.
- Google Photos capture dates, including corrected dates in supplemental metadata.
- GPS coordinates and offline place names when GeoNames is installed.
- Favorites.
- User-created Google Photos albums.
- Embedded EXIF camera, lens and exposure details through Smriti’s normal metadata reader.
Imported originals are ordinary files under:
<library>/Imported from Google Photos/<year>/
Takeout ZIPs are never modified or deleted.
Resume and repeat
The import is restart-safe. Completed originals are atomically placed before being recorded in the library-local Takeout ledger. If the app closes, the drive disconnects, or you cancel the job, select the same ZIP files again. Smriti verifies and reuses completed files instead of creating duplicates.
Re-importing the same export is also safe. A later export can restore additional album or metadata records for an already imported original.
Warnings and unsupported data
Smriti imports photo and video formats supported by the normal scanner. It reports unsupported files, very small invalid media, malformed sidecars and corrupt ZIP entries at completion. A missing sidecar does not block its media file; embedded EXIF and filename dates are used instead.
Only ZIP Takeout exports are supported. TGZ archives must be requested again from Takeout as ZIP. Google comments, sharing permissions and Google’s people labels are not turned into Smriti features. Smriti performs its own local face recognition.
Safety
Archive paths are validated before extraction, files are streamed instead of expanded into a second temporary tree, and a file is moved into the library only after it has been fully written and hashed. Existing files are never overwritten.