lynx-dev
[Top][All Lists]
Advanced

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

Re: lynx-dev [patch] more SH_EX tweak


From: Klaus Weide
Subject: Re: lynx-dev [patch] more SH_EX tweak
Date: Sun, 17 Oct 1999 09:47:03 -0500 (CDT)

On Sun, 17 Oct 1999, Hiroyuki Senshu wrote:

> Hello, lynx-dev.

Hello..

> Klaus Weide wrote:
> Of course, SH_EX is not Windows exclusive use.
> A code is being written on the assumption that 
> even the UNIX like OS is used.
> 
> I am making LYNX with the following environment.
> 
>   o Borland C++ compiler
>   o Microsoft Visual C++ compiler
>   o Cygnus Cygwin (egcs, gcc compiler)
> 
> In other words, _WINDOWS doesn't sometimes come into 
> being on Cygwin.
> 
> Because of this, a very complex condition judgment is done.
> Of course, there may be some coming off omissions.
> It asks for an opinion about the point.

Some changes seem to be just different personal preferences,
is that right?

For example I just ran across this in LYForms.c:

#ifdef SH_EX
/* ASATAKU emacskey hack 1997/08/26 (Tue) 09:19:23 */
        if (emacs_keys &&
            (EditBinding(ch) == LYE_FORWW || EditBinding(ch) == LYE_BACKW))
            goto breakfor;
/* ASATAKU emacskey hack */
#endif

(Quite old change date)

I do not know whether that is is personal customization or whether it
is meant as a general improvement / bug fix.  If it's the second,
then it should get un-ifdef'd.

> -------- * ------ * ------- * -------
> 
> As for present dev.12, some modification is done.
> /* !!! */ A movement was unstable under the condition 
> that a part wasn't changed.
> 
> Teach me the reason why this part was removed.

It caused memory leaks.  See

   http://www.flora.org/lynx-dev/html/month1099/msg00154.html
   http://www.flora.org/lynx-dev/html/month1099/msg00164.html
   http://www.flora.org/lynx-dev/html/month1099/msg00169.html

also Tom's summary in CHANGES:

* undo spurious change from win32 integration in dev.5 to HTSprint0() which
  causes a memory leak (reported by John Bley, analysis KW) -TD

If something bad happens because of that change, then something must
have used HTSprintf0 the wrong way.  That code should be fixed.

(I did not understand the description "a movement was unstable".)

   Klaus


reply via email to

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