> I tend to agree that the line width spotless uses is too narrow for my taste, 
> but I guess it is not configurable, and I still prefer some standard over the 
> mishmash we had before.

It's one of the key reasons I think this is an advantage - you can't
have 100 people agree on a consistent format. Not possible. Here you
just have
to accept the way it's formatted and live with it. Sometimes you can
reshape your code to make it more aligned with your expectations
though.

For example:

-········byte[]·b·=·new·byte[offset·+·length·*·Float.BYTES·-·TestUtil.nextInt(random(),·1,·Float.BYTES)];
+········byte[]·b·=
+············new·byte[offset·+·length·*·Float.BYTES·-·TestUtil.nextInt(random(),·1,·Float.BYTES)];

This line is broken because it's too long and gjf tries to keep
statements together (or vertically aligned, as you'll see in many
other cases). If this
bothers you then extract something to make the code read (and format) better:

int length = 
offset·+·length·*·Float.BYTES·-·TestUtil.nextInt(random(),·1,·Float.BYTES);
byte[] b = new byte[length]

Is it more verbose? Sure. Is this a nicer read? Maybe. I've seen diffs
where the nested expressions were humongous - very complicated
to grasp as one-liners. I found them much more intuitive the way gcj
formatted them (over multiple lines sometimes, indenting arithmetic
and quotes for clarity).

It's definitely biased towards shorter expressions than long ones
involving multiple statements.

Dawid

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to