wswsh is a static website script using shell. It means [w]eb [s]ucks [w]ithout [sh]ell. Simple name for a simple script. It has many advantages:
- Lightweight
- Only requires a shell + UNIX utilities
- Compatible with ahrf
- Easily "hackable" with external scripts / interpreters
- Less than 140 LOC (without external layouts)
- Human readable configuration
You can read another howto with examples here (might be a good intro).
Create a directory including the following files:
includes/layout
wswsh
wswsh.conf.default
You'll need a config file. Copy the file wswsh.conf.default to
wswsh.conf. Edit it according to your needs. The comments explain
almost everything.
A typical hierarchy contains a src directory, with your website inside
it.
.
├── includes/
│ └── layout
├── src/
│ ├── css/
│ │ └── style.css
│ ├── blog/
│ │ └── my_post.txt
│ ├── me/
│ │ └── john_doe.txt
│ └── foo/
│ └── baz/
│ └── this_is_a_test.txt
├── wswsh
└── wswsh.conf
Each directory in src will be reproduced in a new directory dest.
There is no default interpreter, only cat is called. It involves posts
written in HTML.
wswsh also supports ahrf.
When you're ready, launch ./wswsh "$PWD". Using the previous example,
we now have:
.
├── includes/
│ └── layout
├── dest/
│ ├── css/
│ │ └── style.css
│ ├── blog/
│ │ └── my_post.html
│ ├── me/
│ │ └── john_doe.html
│ └── foo/
│ └── baz/
│ └── this_is_a_test.html
├── src/
│ ├── css/
│ │ └── style.css
│ ├── blog/
│ │ └── my_post.txt
│ ├── me/
│ │ └── john_doe.txt
│ └── foo/
│ └── baz/
│ └── this_is_a_test.txt
├── wswsh
└── wswsh.conf
dest is your generated website. You can upload it anywhere.
If you want to have the same wswsh executable for all your blogs
(assuming you have several websites), it's possible to put wswsh in
your PATH, instead of having a "redundant" file, in every directory.
The default behavior allows you to modify wswsh for your websites. So,
it's possible to write custom modifications per site.
The true "power" of wswsh resides in hooks launched at many different
steps. Those hooks are sourced inside wswsh env allowing you to use
inherited variables to launch all kind of specific actions.
An "interpreter" can be run if it's placed ouside your PATH. Write the
full path to the executable, within wswsh.conf:
WSH_INTERP="/home/foo/my_exec"
Few months ago, I still wanted to write a sh compliant version but I
decided to drop that idea. At the moment, we are less than 5 people who
use it. Creating a second version would be a waste of time, especially
when the users already switched to mksh.
Maintaining two redundant versions isn't easy and I do not want to work
for nothing. If you're still interested, you're free to adapt to sh.
It shouldn't be complicated.
When I added awk regexes, one of my goal was to support nawk, mawk
and gawk. gawk is very common among Linux distributions, so I didn't
have the choice. The regexes were "created" on mawk. I tested the
compatibility and it worked flawlessly with the required implementations.
So, you can gain some precious seconds if you're brave enough to use
nawk or mawk.
Code created by Ypnose, under BSD (3-Clause) License.
Powering:
You decided to adopt wswsh for your website(s)? Please contact me. I would be happy to add it in the README.