Introduction

What is Digikam-DB?

Digikam-DB is a SQLAlchemy-based Python library that enables access to the Digikam database containing Digikam’s metadata (title, tags, rating, Exif information…). You can use it to read or manipulate this data by directly accessing the Digikam database.

Digikam is an image management application using KDE. For more information on Digikam, see https://www.digikam.org/.

Digikam-DB aims at offering a convenient encapsulating the more complicated aspects of the Digikam database.

Warning

This package is still under development. Use at your own risk and make a backup of your Digikam database before writing to it.

Installation

For installation via PIP, simply type:

pip install Digikam-DB

If you are using a MySQL database, you may want to install the additional dependencies:

pip install Digikam-DB[mysql]

The source code is available on Github.

Basic Usage

In this small example, we

  • connect to the database via a Digikam object,

  • create a tag my tag if it does not exist yet,

  • add this tag to image /path/to/my/image.jpg,

  • change some properties of /path/to/my/image.jpg,

  • commit these changes to the database and

  • print file name and complete path of the image with id 42.

from digikamdb import Digikam

dk = Digikam('mysql://myuser:mypass@my.host.org/mydb')

tag = dk.tags['my tag']
if not tag:
    tag = dk.tags.add('my tag', dk.tags.root)

img = dk.images.find('/path/to/my/image.jpg')[0]
img.tags.append(tag)
img.caption = 'My Caption'
img.title = 'My Title'
img.information.rating = 3
dk.session.commit()     # Commit changes to database

img2 = dk.images[42]    # Access image by id
print(img2.name, img2.abspath)

For more information on usage, see Using Digikam-DB.

Limitations

  • Adding or deleting albumroots, albums and images is not supported.

  • Digikam-DB will probably not run under Windows or MacOS.

  • Fractions of seconds on datetime fields are ignored (this is standard on MySQL, but not on SQLite).

  • Digikam-DB only implements the Core database of Digikam, not the Thumbs, Faces, and Similarity databases.

  • There is no interface to the following tables of the Core DB:

    • DownloadHistory

    • ImageRelations

    • ImageTagProperties

    • Searches