Originally Posted by
grahaml2
two good answers, thanks
things are if not totally clear then certainly less murky
More generally, not related to this installer, any non-packaged program/suite can have completely different installations. Some allow control where the target is located, some default to /usr/local/, some default to the CWD/something-version/ ... the team that wrote the installation. I've written installations for Windows and lots of Unixen. On Windows, we never asked where anyone wanted anything. On Unix, I always asked - or showed the default location and got confirmation. Many install.sh scripts don't do that. They assume you can read the README file or INSTALL file that comes with the package and do what is explained. Sometimes that is difficult due to language barriers, sometimes by the writer.
Don't assume that a script named "install.sh" is the same as any other "install.sh" script for any other system. It might be, but it is just as likely not to be related at all. Plus, Unix systems often have an install program that copies files and sets permissions, owners, groups. There is a manpage for it on my Ubuntu system:
Code:
NAME
install - copy files and set attributes
SYNOPSIS
install [OPTION]... [-T] SOURCE DEST
install [OPTION]... SOURCE... DIRECTORY
install [OPTION]... -t DIRECTORY SOURCE...
install [OPTION]... -d DIRECTORY...
DESCRIPTION
This install program copies files (often just compiled) into destination locations you choose. If you want to download
and install a ready-to-use package on a GNU/Linux system, you should instead be using a package manager like yum(1) or
apt-get(1).
I've never used install directly, but if I wanted to make something like puppet, chef, ansible, I would.
Bookmarks