On Thu, Mar 01, 2001 at 12:33:49AM +0000, Julian Gilbey wrote: > > Makefiles are already arbitrary code. You can write a makefile rules file > > that would use shoop stuff -- which would be perfectly conformant, but would > > that make it any easier to edit (for the uninitiated)? > > But one is much less likely to do that: there may be the odd line of > code in shoop, but to actually warp the makefile into shoop would seem > like hard work.
Considering that make just runs the commands through shell, I wouldn't bet on it... > > Imagine a rules file like this: > > [...] > > How is that hard to NMU? Unless of course the developer in question doesn't > > know a thing about shell scripting. :| > > Very nice, but why? What's the point of copying a standard example > makefile and turning it into a shell script? Ask the opposite question -- what's the point of taking a sequence of commands and turning it into a makefile? Those external commands don't depend on their caller being a makefile. In fact, they could care less[1] what calls them. This is the point I'm trying to prove. The specific requirement for the rules file to be a makefile is too strict and is not necessary. > If someone is going to go to the effort of writing rules in a system other > than make, they're likely to want to do something a bit bizarre, and > that's where things get hairy. Bizarre? The next upload of maildrop will have debian/rules written in plain old shell, check it out, I don't think you'll see anything odd in there. (Feel free to file as many RC bugs you want ;) [1] aside from the fact they're computer programs and inherently have no ability to care :) -- Digital Electronic Being Intended for Assassination and Nullification
