The Pedigree Project
0.1
|
Pedigree differs from traditional UNIX systems in a few important ways. This page describes these and also clarifies why.
In Pedigree, the filesystem is referenced using a custom model where mount points are disambiguated and then paths within those mounts are specified.
This looks like the following:
/path/to/file
(this loads a file on the same mount as the current working directory)mount»/path/to/file
(this loads a file from the mount
mount point)There are some well-defined mount points that exist by default:
root»
is the root filesystem, the default filesystem on which users begin and from which configuration is read for the Registry.dev»
provides device access, like /dev.raw»
provides access to raw disk devices and partitions.scratch»
is an in-RAM filesystem, much like /tmp.runtime»
is an in-RAM filesystem for runfiles (ala /run), where files can only be modified by the owning Process.unix»
provides a location for named UNIX sockets to reside.To enter the »
seperator character, use ALT-SHIFT-.
.
Be aware that if $PATH does not use fully-qualified paths, switching the current working directory to a different mount point can have unexpected results as the $PATH lookups will be done on the local mount point, rather than the location on which the files actually reside.
For applications that depend upon the first character in a path being /
to be recognised as an absolute path, the path prefix /@/
can be added.
For example, the two paths root»/applications/bash
and /@/root»/applications/bash
point to the same file, but the latter is clearly an absolute path whereas the former could be interpreted by some applications as a relative path.