Re: Issue: PRETTY-PRINT-INTERFACE (version 3)

[David A. Moon <Moon@STONY-BROOK.SCRC.Symbolics.COM>]

    Date: Mon, 20 Mar 89 10:15:45 EST
    From: Dan L. Pierson <pierson@mist.encore.com>

	Date: Sat, 18 Mar 89 01:32 EST
	From: David A. Moon <Moon@STONY-BROOK.SCRC.Symbolics.COM>
    
	I could not find any specification of the units of measurement for
	*PRINT-RIGHT-MARGIN*, *DEFAULT-RIGHT-MARGIN*, *PRINT-MISER-WIDTH*,
	indentation, and tabulation.
    
    I think that all three clearly have to be in the same units; the
    writeup should be changed to specify this.  After thinking about it
    for a while, it's not clear that the exact unit matters much as long
    as it's not to small.  Pixels are too small; "n"'s or "m"'s aren't.
    Personally I'd suggest an "m" as the standard unit.

I don't see anything wrong with that, although I can't claim to be an
expert in this area.

    Other than that, I don't think the propsal has terrible problems with
    variable width (even kerned) fonts because most indentation is done
    relative to the position of the start of another word.  This common
    case can be handled with arbitrary precision; a variable width font
    implementation would presumably indent to the exact pixel position of
    the relevant character (how would that be expressed in Postscript?).

I think what you're saying is that the argument to LOGICAL-BLOCK-INDENT
(or ~I) is almost always zero?  In the examples given in version 1
of the proposal, zero is in the majority, there are also two 1's and
a 2.  I didn't find any cases where the number given to ~I seemed to
be intended to correspond to the number of characters in some word.
There might need to be a style suggestion for writers of pretty print
methods telling them to use relative indentation instead of counting
characters.  Given that, your suggestion sounds plausible to me.