This is (most of) the curl.se website contents. It mostly builds static HTML files that are preprocessed.
The website is a on old custom made setup that mostly builds static HTML
files from a set of source files using (GNU) make. The sources files are
preprocessed with what is basically a souped-up C preprocessor called fcpp
and a set of perl scripts. The manpages get converted to HTML with
roffit.
Markdown is converted to HTML with pandoc.
Make sure the following tools are in your $PATH.
- curl
- GNU date
- GNU enscript
- fcpp
- GNU make
- pandoc
- perl (with CPAN packages: CGI, HTML::Entities)
- roffit
- zip
Once you have cloned the Git repo the first time, invoke sh bootstrap.sh once
to get a symlink and some some initial local files setup, and then you can
build the website locally by invoking make in the source root tree.
Note that this does not make you a complete website mirror, as some scripts and files are only available on the real actual site, but should give you enough to let you load most HTML pages locally.
To run a local copy of the curl website, have a local Apache or python
to serve curl.local on 127.0.0.1. Add this line to /etc/hosts:
127.0.0.1 curl.local
A config file for apache2 to run a virtual server for curl.local on your
local machine might look like this:
<VirtualHost *:80>
ServerName curl.local
ServerAdmin [my email address]
DocumentRoot [full path to the curl-www build]
ErrorLog ${APACHE_LOG_DIR}/curllocal-error.log
CustomLog ${APACHE_LOG_DIR}/curllocal-access.log combined
</VirtualHost>
<Directory [full path to the curl-www build]>
Options Indexes Includes FollowSymLinks ExecCGI
AllowOverride All
AddHandler cgi-script .cgi
Require all granted
</Directory>
From the directory containing the website, run:
python3 -m http.server --cgi -b curl.local 8000