Take control of your media

Shotwell is a fantastic program for organizing photos and videos. It supports publishing media to third-party sites, however that requires organizing the media again there, which is time consuming. Once the media is published, any metadata changes that are made in Shotwell will not be shown on that third-party site.

This program exports a mobile-friendly static HTML site of a Shotwell library. The generated site is fully self contained and does not require Internet access or any third-party resources for the long-term preservation of the library. The search page provides a rich, single-page user experience that allows browsing your library through various predefined views: all media over time, all events over time, year view, and nested tag view. Additionally you can search your library using 24 different fields.

A second static site is generated where the use of Javascript is minimized for the long-term preservation of the library.

You can serve these static files on a webserver such as nginx / apache2, and if desired, setup some kind of authentication (like HTTP basic auth over HTTPS). There's no complicated additional webserver plugins that need to be setup for a particular application runtime since you're just serving static content.

Note that the generated website is readonly and you cannot make any changes to your media there. If you need to make a change, then make those changes inside Shotwell, and rerun this project to incrementally update the generated website. I have automation setup so that when I backup my Shotwell library to my home server that my photo/video website is automatically regenerated.

Screenshots

See the screenshots page for more screenshots of the generated HTML site.

Photo Frame

The search page also allows you setting up a browser in kiosk mode on a small single board computer like a Raspberry Pi and use it as a photo frame that pulls from your home server as described on the photo frame page.

Keeping directory structure in sync with the library

There is a utility script utils/rename_media.py that you can run to rename all of your media files so that the directory structure on disk matches the way that you have your media organized within Shotwell. It will store the media on disk using the format year(s)/event name/original base filename.

How to create printed photo books

The search page supports exporting the current view as a CSV file. For my photo books, I export a series of CSV files for events that occurred throughout the year, and chain together all of the images into a PDF that can be sent off to the book printer. My process for creating these books is described on this page.

Usage

Ensure that you have the proper dependencies installed on your system:

# Red Hat-based systems
sudo dnf install -y exiv2 exiv2-devel ffmpeg ImageMagick boost-python3 python3 \
                    python3-dateutil python3-exiv2 python3-geojson python3-humanize \
                    python3-pillow python3-pip uglify-js

# Debian-based systems
sudo apt-get install -y exiv2 ffmpeg imagemagick libboost-python-dev libexiv2-dev python3 \
                        python3-dateutil python3-geojson python3-humanize python3-pil \
                        python3-pip python3-pkg-resources uglifyjs
pip3 install py3exiv2

Backup your Shotwell database/library and then generate a static HTML site:

shotwell_site_generator.py \
            --input-database /path/to/shotwell/sqlite/database \
            --input-media-path /path/to/shotwell/full-sized/images \
            --input-thumbs-directory /path/to/shotwell/thumbs360/directory \
            --dest-directory /path/to/generated/html/site \
            --src-assets-directory /path/to/this/repo \
            --title "My Photos"

Open the top level index.html file in your browser to view your library using the rich search experience. The file static-site/media/index.html contains the site that minimizes the use of Javascript.