Is it worth having shell-agnostic ./run and ./finish?
 
I'm asking because I went through some trouble to make my project 
somewhat shell-agnostic.  The intent is I can have separate support for 
/bin/sh, execline, or some other scripting environment, should it be 
desired.  But it's only worth the effort if I can get off of the 
sh-as-launcher dependency completely.  I can see how execline can be 
used to make run scripts, but I haven't tried making a finish (nor have 
I had a real need for ./finish scripts in shell).  So the question is: 
how difficult would it be to write a ./finish script in execline?  The 
answer will determine if it's worth the effort to avoid hard 
dependencies on /bin/sh.  If it's easy enough, then I go ahead with this 
plan.  If there are difficulties...well...
At  the moment, all of the ./run scripts for a service definition are 
simply symlinks to another symlink, which points to a script. Example:
/etc/sv/daemon/run >points to> /etc/sv/.run/run >points to> 
/etc/sv/.run/run.sh
Change the middle symlink and you get, for example:
/etc/sv/daemon/run >points to> /etc/sv/.run/run >points to> 
/etc/sv/.run/run.execline
That way you don't have to change all of the ./run symlinks in hundreds 
of definitions, you just change the one.
Received on Fri Mar 20 2015 - 22:05:51 UTC
This archive was generated by hypermail 2.3.0
: Sun May 09 2021 - 19:44:19 UTC