Hello Tanu, > > patch 4 implements a new .ifenv CLI meta command I found useful: executes commands > > only when an environment variable exists (optionally with a certain value) > > Can you elaborate, what problem you solve with environment variables? Is > there no reasonable alternative? Here is someone else's attempt at > explaining why environment variables are not nice: > http://peterlyons.com/problog/2010/02/environment-variables-considered-harmful thanks for taking a look :) I'm sure there are different ways of doing things... I found it useful to conditionally load either module A or module B (or the same module with different parameters) in a sequence of modules that get loaded in a custom script.pa I could have split up my script.pa into scriptA.pa and script.B and specify using pulseaudio's -F parameter, and use .include's to avoid duplication; or I could have scripted the PA commands elsewhere I have no strong opinion about that added feature; I guess one cannot make the point that it is absolutely essential -- but the same thing can be said about PA's script in general are you OK with the cleanups/fixes in patches 1 and 2? regarding the links provided, I think that the usefulness of environment variable has passed the test of time and they are fundamental to many unix tools for a good reason; esp. sourcing a key/value file into env. vars. is really useful; I can manipulate an env. var. easily in a shell script, writing a particular config file format -- not so much regards, p. -- Peter Meerwald +43-664-2444418 (mobile)
