1600pr.sh is a minimalist static photoblog generator made up of a single POSIX-compliant shell script. It should work on most Unix-like systems. It passes ShellCheck and some effort has been made to ensure external tools (grep, sed, awk, etc.) are used in a portable way.
The only external dependency is ImageMagick for creation of thumbnails and alternative sizes (for responsive images), however this can be disabled if you don't want those things.
You can see it in action on minorshadows.net.
Named after Fujifilm Neopan 1600 Professional (RIP). Inspired by (but not based on) Expose.
# Download script
curl -O https://github.com/andersju/1600pr.sh/raw/master/1600pr.sh
# Make it executable
chmod +x 1600pr.sh
# Add first photo
./1600pr.sh ~/some-photo.jpgBoom, you now have a photoblog built in public/.
You probably have Python installed, so fire up a local web server and look around:
cd public/ && python3 -m http.server
rsync or scp the files in public/ to somewhere:
rsync -av -e ssh public/* user@example.com:public_html
You probably want to change a few variables (title, absolute url, menu) in
the script, though, or set environment variables (_1600PR_SITE_TITLE,
_1600PR_SITE_URL, _1600PR_EMAIL, etc).
Add photo with title:
./1600pr.sh -t "My cat." ~/my-cat.jpg
Title is only used in the RSS feed, and for the image's alt attribute. If no
title is set, current date will be used instead.
Add photo with title and custom date (should be RFC 822 datetime):
./1600pr.sh -t "My cat." -d "Sun, 31 May 2020 19:22:17 +0200" ~/my-cat.jpg
If you nuked your public/, or changed some setting (e.g. what sizes to generate),
you can rebuild everything with -b:
./1600pr.sh -b
Remove post with ID 23 from the "database" and rebuild:
./1600pr.sh -r 23
(This doesn't remove anything from public/, nor does it remove the
original photo stored in images/.)
ImageMagick is required, unless you set sizes to "" and archive_page to false,
disabling creation of alternative sizes and archive page with thumbnails.
Information about each post is stored in _1600pr.dat. It's a plaintext file
where each line represents one post, in ascending order from oldest to newest.
(To change the order, just move lines around and rebuild.)
Each line has tab-separated values: photo ID, date, filename, optional title.
The original photo is copied to the non-public folder images/. For each photo,
a number of sizes are generated (unless you've opted not to) and put in
public/images/<id>/.
One page per photo. Each photo page gets the relative URL /photo/<id>/.
Each page has "prev" and "next" links, if applicable.
/photo/ is an archive page with thumbnails of every photo.
The most recent photo page is copied to /index.html.
RSS and CSS files are also generated.
Starting with only 1600pr.sh and adding two photos (foo01.jpg and foo02.jpg),
we get the following:
.
├── _1600pr.dat
├── 1600pr.sh
├── images
│ ├── foo01.jpg
│ └── foo02.jpg
└── public
├── images
│ ├── 1
│ │ ├── 1280_foo01.jpg
│ │ ├── 1600_foo01.jpg
│ │ ├── 1920_foo01.jpg
│ │ ├── 800_foo01.jpg
│ │ └── thumb_foo01.jpg
│ └── 2
│ ├── 1280_foo02.jpg
│ ├── 1600_foo02.jpg
│ ├── 1920_foo02.jpg
│ ├── 800_foo02.jpg
│ └── thumb_foo02.jpg
├── index.html
├── index.xml
├── photo
│ ├── 1
│ │ └── index.html
│ ├── 2
│ │ └── index.html
│ └── index.html
└── style.css
A bunch of things can be set with variables at the top of the script. Otherwise just hack it to your heart's content. This is a quick hack for my own personal use: it's deliberately minimalist and I'm not looking to add major features, but perhaps it can be of use to someone else too.
To change title/url/email/sizes/etc. without modifying the script you can use environment variables. Just set them somewhere, perhaps in a wrapper script, like so:
#!/bin/sh
export _1600PR_SITE_TITLE="example.com"
export _1600PR_SITE_URL="https://example.com/"
export _1600PR_EMAIL="foo@example.com"
exec ./1600pr.sh "$@"