Comments for page "Code Generation & Visual Smog in Code (part II)"

(no title)

Sun, 27 Apr 2014 11:59:30 +0000

The idea was to minimise possible clashes with Ruby.

The assumption was that a dot at the beginning of the line is very rare i Ruby. A dot following a whitespace may be less rare:

 foo .bar() .baz()

The sequence of multiple dots should be actually viable. Still, I feel like anchoring the generated code to the left is good for readability, i.e. the indentation of the text you see (within a single function) is the same as the one in the output. Kind of wysiwyg principle.

That would not apply, if mutliple dots are allowed:

 ........a ....b

Would produce:

a b

(no title)

Sun, 27 Apr 2014 11:44:22 +0000

Why don't you allow identation *before* the dot for generated code lines? Or alternatively allow any number of dots?

So this:

 .foo = bar; if(a == 0) .++foo; if(b == 0) .if (foo == 3) . bar = 0; if(c == 0) .++bar; end .foo = 0; end end

could become this:

 .foo = bar; if(a == 0) .++foo; if(b == 0) .if (foo == 3) . bar = 0; if(c == 0) .++bar; end .foo = 0; end end

or this:

 .foo = bar; if(a == 0) .....++foo; if(b == 0) .........if (foo == 3) ......... bar = 0; if(c == 0) .............++bar; end .........foo = 0; end end