🚀 go-pugleaf

In <810138845snz@soliton.demon.co.uk>, Ian Cargill <ian@soliton.demon.co.uk> writes:
>And, of course, you will have to mandate the width of tabs 
>and/or forbid the use of tabs.

Yes.  Tabs are simply unacceptable even without the use of
indentation for control structure.

The company for which I current work has had all sorts of
trouble maintaining readable code due to tabs being different
in different editors -- our shop policy is now that you cannot
use an editor in which disabling tabs is unavailable.  That
does NOT mean that you cannot use the tab KEY, and set
that for any width you want -- it just can't insert tabs.

> Otherwise, what happens when 
>you edited with tabs set to 8 and the compiler expects 4?

Its worse than that!  Languages such as COBOL -- of which
we have a few programs :-( -- requires specific columns
and tabs messes that up very nicely.  And readability in
asm files typlically does not use a fixed tab width.

As far as I see it, the only acceptable policy regarding tab
characters is DON'T!  You can use fixed or variable tabbing
all you want, just don't put it into the source file!

>I would much rather rely on layout-independant braces to define 
>my block scope.

Your choice.  But I would rather reduce typing (and since I type at
85 wpm, it is NOT because I am a lazy typist <g>), reduce code
clutter so I can read the code more easily, and reduce program
bugs.  I am used to following coding standards, a minor addition
over identation rules is NOTHING compared to the benefits.

>What happens when someone fouls up with global search and replace 
>or strange editors which changes the spacing/indentation?  I have 
>had both done to me and it would have been a nightmare having to 
>go laboriously through code re-creating the block structure.

You DO archive your source code don't you?  Including significant
temporary versions?  I have also made massive global editing
mistakes which messed up a program -- but I have never lost
more than 5 minutes of work in the last 2 decades years due to
such a mistake.

The answer to problems like this is simply professionalism in the
deveopment of code.  Professionals use archives to protect
against mistakes of this nature (and much more, of course),
amateurs and the inexperienced don't.

Michael Lee Finney