Re: RFI: Trailing blanks

[Paul Eggert]

> From: "Gary V. Vaughan" <address@hidden>
> Date: Sat, 18 Aug 2001 15:36:38 +0100
> 
> I vote for dropping line smashing altogether.

I don't fully understand all the issues involved with line smashing,
but I tend to agree that it's relatively important to remove multiple
blank lines from the output.  If that gets in the way of more
important goals, we should leave the blank lines alone.

> If I'm irritated by huge blocks of whitespace when reading the
> generated code, I do like I do when cpp macros go awry:
> 
>   gcc -E -DHAVE_CONFIG -I. -I.. hello.c | sed 's/^[   ]*$//' | less -s

I have the LESS environment variable set to "cins", so I have -s
turned off by default.  Perhaps that is partly why I don't mind extra
blank lines.  (I use "less" a lot.)

(Surely there is an Emacs equivalent.  :-)

However, even if we don't need a quadrigraph for the empty string for
this, it still would be useful in some other cases.


> From: address@hidden
> Date: Sat, 18 Aug 2001 11:24:38 +0200

> So we are looking for specially non special characters. @address@hidden

Any repeated character will make the string more likely.

How about @&address@hidden  That would be easy to remember as expanding to the
empty string, as the mnemonic would be "amp tee".  (Here's where I
duck to avoid being hit by a tomato....)