From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,5c72b6fec2f83257
X-Google-Attributes: gid103376,public
From: dewar@merv.cs.nyu.edu (Robert Dewar)
Subject: Re: Help: Binary operator "="
Date: 1997/06/10
Message-ID: <dewar.865941998@merv>#1/1
X-Deja-AN: 247442445
References: <3381c4ba.quantum@quantum.pc.my> <dewar.864220244@merv>
 <33932325.6E38@lmtas.lmco.com> <dewar.865691385@merv> <339B6B90.41F@lmco.com>
Organization: New York University
Newsgroups: comp.lang.ada
Date: 1997-06-10T00:00:00+00:00
List-Id: <comp.lang.ada>


Ken says

<<Which is why I characterized it as "OK" vs. "nice" in my comments. The
(admittedly thin) distinction you could draw is as follows: Infix
operators
should be extremely obvious as to their definition (at least if written
correctly). This is why it's probably OK to use some terse notation like
"+" rather than Add. Non-infix operators are slightly less obvious, so
having some additional information (like the type of stack on which they
operate) might be useful. It's not so much being able to trace the
location
of their definition as providing useful information about the operation.
I would say that, in the second case, either leaving on the name or
taking
it off would probably be all right.>>


Which in fact leads to a different criterion entirely. Instead of having
an absolute "never-use-use" rule, you are really suggesting a rule that
says, use qualification when it is helpful, and not otherwise.

That sounds like an entirely reasonable rule to me.

For example, in the case of GNAT, we generally expect that those looking
seriously at the GNAT compiler sources get to know where things are pretty
quickly, since there is a rather systematic placement of functionality (for
example, you know that Analyze_Subprogram_Body will be in Sem_Ch6, since
analysis is to do with semantics, and subprogram bodies are in chapter 6
of the RM).

On the other hand, the runtime library is hundreds of interlinked units, and
it is definitely helpful to use qualification extensively in this context,
particularly in the specs.