“fgallery” is a static photo gallery generator with no frills that has a stylish, minimalist look. “fgallery” shows your photos, and nothing else.
There is no server-side processing, only static generation. The resulting gallery can be uploaded anywhere without additional requirements and works with any modern browser.
An example gallery can be found here.
Latest 5 available releases of fgallery (most recent first):
Full archive in releases/.
|NEWS:||Summary of changes between releases.|
Release announcements are made through Gitlab.
All the relevant source/developer information can be found on Gitlab:
Generate all the static files with ./fgallery:
./fgallery photo-dir my-gallery
Upload “my-gallery” somewhere.
To test/preview the gallery locally using Firefox, you can just open the file my-gallery/index.html. On other browsers you need a web server (due to AJAX same-origin restrictions). If you have python installed, a quick way to test the gallery locally is to run:
cd my-gallery python3 -m http.server
and then navigate to http://localhost:8000.
Pre-built packages for fgallery (and facedetect) are available:
sudo apt-get install fgallery facedetect
sudo pacman -S fgallery
sudo layman -a robert7k sudo emerge www-apps/fgallery
sudo nix-env -i fgallery
pkg install fgallery
You can also try the latest fgallery bundled with facedetect in a Docker container using the following Dockerfile provided by Stavros Korokithakis:
The images as shown by the viewer are scaled/compressed using the specified quality to reduce viewing lag. They are also stripped of any EXIF tag. However, the pictures in the generated zip album are preserved unchanged.
Lossless auto-rotation is applied so that images can be opened with a browser directly. JPEG and PNG files are also re-optimized (losslessy) before being archived to furthermore save space.
Image captions are read from simple text files or directly from EXIF metadata. Captions can be controlled by the user using the “bubble” icon or by pressing the “c” keyboard shortcut, which cycles between normal/always hidden/always shown visualization modes.
Preview and thumbnail images are converted to the sRGB color-space by default, which provides better results on normal displays and browsers without color management support.
All images can be included to be viewed individually at full resolution in the gallery by using the -i flag. Panoramas are automatically detected and the original image is included in full-size by default, as often the image preview alone doesn’t give it justice.
For best results when shooting with multiple cameras (or friends), synchronize the camera clocks before starting to take pictures. Just pick one camera’s time as the reference. By doing this the album is automatically shown in logical shooting order instead of file-name order.
Never use the -s or -d flags. Let your friends and viewers download the raw album at full resolution, not the downscaled crap. Don’t make me angry.
The sizes of the thumbnails and the main image can be customized on the command line with the appropriate flags. Two settings are available for the thumbnail sizes: minimum (150x112) and maximum (267x200). Thumbnails will always be as big as the minimum size, but they can be enlarged up to the specified maximum depending on the screen orientation. The default settings are tuned for a mostly-landscape gallery, but they can be changed as needed.
Images having a different aspect ratio (like panoramas) are cut and centered instead of being scaled-to-fit, so that the thumbnail shows the central subject of the image instead of a thin, unwatchable strip. When this happens, the viewer shows a sign on the thumbnail along the cut edges (this effect can be seen in the demo gallery).
To simply favor photos shot in portrait format, invert the width/height of the thumbnail sizes:
./fgallery --min-thumb 112x150 --max-thumb 200x267 ...
This will force the thumbnails to always fit vertically, at the expense of a higher horizontal thumbnail strip.
If your photos are mixed and can contain people, faces or portraits, you can enable face detection by using the -f flag and installing facedetect.
Face detection will ensure that the thumbnails, especially when cut, will be centered on the face of the subject. If face detection is enabled, there’s generally no need to increase the thumbnail size.
Since every camera is different, and every monitor is different, some color transformation is necessary to reproduce the colors on your monitor as originally captured by the camera. Color management is an umbrella term for all the techniques required to perform this task.
Most image-viewing software support color management to some degree, but it’s rarely configured properly on most systems except for Safari on Mac OSX. No other browser, unfortunately, supports decent color management.
This causes the familiar effect of looking at the same picture from your laptop and your tablet, and noticing that the blue of the sky is just slightly off, or that colors look much more contrasty on one screen as opposed to the other. Often the image has the information required for a more balanced color reproduction, but the browser is just ignoring it.
We’re writing this down because Firefox has built-in color-management support, but it’s disabled by default on all platforms. It’s also ranking very low on the list of improvements to make, with some bugs being open for years. In an attempt to raise awareness, please complain/contribute to any of the existing bug reports, citing the Technical details on this web page.
On Firefox, the installation of the following “Color Management” add-on is recommended:
When installed, in the add-on configuration, you’ll need to enable color management for “All images” and restart the browser. Also, if you have a multi-monitor setup, it’s advisable to manually set the “Display profile” to the external/calibrated screen, since FF won’t automatically select the color profile for the current monitor, and just default to the primary. Firefox has also known bugs with LUT profiles, though the more common Matrix profiles seem to work fine.
We understand that CM has a considerable impact on image rendering performance, but strictly speaking CM doesn’t need to be enabled on all images by default. It would be perfectly fine to have an additional attribute on the image tag to request CM. The current method of enabling CM only on images with an ICC profile is clearly not adequate, since images without a profile should be assumed to be in sRGB color-space already.
Because of the general lack of color management, fgallery transforms the preview and thumbnail images from the built-in color profile to the sRGB color-space by default. On most devices this will result in images appearing to be closer to true colors with only minimal lack of absolute color depth. As usual, no transformation is done on the original downloadable files.
Frontend/viewer: none (static html/js/css)
ImageMagick (imagemagick, http://www.imagemagick.org)
LittleCMS2 utilities (liblcms2-utils, http://www.littlecms.com/).
perl >= 5.14 (threading support enabled), with the following required modules:
and the following additional recommended modules:
Several other tools are supported, but are only used when installed. Therefore it’s also helpful to install:
On Debian/Ubuntu, you can install all the required dependencies with:
sudo apt-get install imagemagick exiftran zip liblcms2-utils sudo apt-get install libimage-exiftool-perl libcpanel-json-xs-perl
To save more space in the generated galleries, we recommend installing also the optional dependencies:
sudo apt-get install jpegoptim pngcrush 7zip
fcaption is written in Python and requires either PyQt6 or PyQt5. You can install the required packages with:
sudo apt-get install python3-pyqt6
For face detection support, simply follow the facedetect installation instructions.
On a Mac, we recommend installing the dependencies using MacPorts. After installing MacPorts, type:
sudo port install imagemagick lcms2 jpeg jpegoptim pngcrush sudo port install p5-image-exiftool p5-cpanel-json-xs
Installation is currently optional. fgallery can be run from the source directory directly. A simple Makefile is provided to install the required data files into the appropriate prefix:
sudo make install install_fcaption PREFIX=/usr/local
“fgallery” is composed of a backend (the “fgallery” script) and a viewer (contained in the “view” directory). Both are distributed as one package, but they are designed to be used also independently.
“fgallery” just cares about generating the image previews and the album data. All the presentation logic however is inside the viewer.
It’s relatively easy to generate the album data dynamically and just use the viewer. This was my aim when I started to develop “fgallery”, as it’s much easier to just modify an existing CMS instead of trying to reinvent the wheel. All a backend has to do is provide a valid “data.json” at some prefixed address. A plugin for a CMS such as Gallery should be very easy to implement.