help-emacs-windows
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [h-e-w] c++ editing , variable name completion


From: Robert Mecklenburg
Subject: Re: [h-e-w] c++ editing , variable name completion
Date: Tue, 30 Jul 2002 07:56:41 -0600

RK> Richard Y. Kim writes:
RK> I've been using tags based completion for many years.

PK> My impression is that Emacs tags-based completion blindly provides any
PK> strings in the tags file that match the symbol at point, regardless of
PK> whether the strings correspond to any method or field defined by the
PK> class of the object whose method or field is sought. If this is the
PK> case, tags-based completion falls far short of what people who are
PK> seeking an intelligent completion facility like those found in
PK> commercial C++ and Java IDEs are looking for in Emacs.

Paul, I understand your point, but you fail to mention in your post
the effects of performance on useability.  I have used dabbrev-expand
(which also scans tags tables and buffers) for completion since 1986
(typically with C/C++).  I've been using JDE since 2000 (thank you,
Paul, for this outstanding package), however I still typically use
dabbrev-expand in jde mode buffers because the performance of
"intelligent" completion makes it less useful than dabbrev-expand.
The dabbrevs package cycles through various possible completions at
sub-second speeds while a single execution of jde-complete takes
roughly five seconds (30 seconds the first time it is executed).  This
is on an 800MHz/P4/256MB Windows 2000 machine.  I can almost always
find a method or symbol name faster with dabbrevs than with
jde-complete.  I may be getting a faster machine soon, so I still have
great hopes for jde-complete, since it is the Right Thing (tm), but
sometimes "worse is better".

Cheers,
-- 
Robert




reply via email to

[Prev in Thread] Current Thread [Next in Thread]