Re: PR/46935 CVS commit: src/lib/libedit

[christos%zoulas.com@localhost (Christos Zoulas)]

The following reply was made to PR lib/46935; it has been noted by GNATS.

From: christos%zoulas.com@localhost (Christos Zoulas)
To: Steffen "Daode" Nurpmeso <sdaoden%gmail.com@localhost>, 
gnats-bugs%NetBSD.org@localhost
Cc: netbsd-bugs%netbsd.org@localhost, lib-bug-people%netbsd.org@localhost, 
        gnats-admin%netbsd.org@localhost
Subject: Re: PR/46935 CVS commit: src/lib/libedit
Date: Tue, 11 Sep 2012 21:11:00 -0400

 On Sep 11,  4:48pm, sdaoden%gmail.com@localhost (Steffen "Daode" Nurpmeso) 
wrote:
 -- Subject: Re: PR/46935 CVS commit: src/lib/libedit
 
 | I have this version running (on FreeBSD..) at the moment, and it
 | seems to work (i'm having READ_RESTART set here - any chance on
 | that?).
 | 
 | First el_getc() now *always* sets el_errno so that read_getcmd()
 | can definitely avoid doing so.
 | 
 | I think v1.69 still has the problem that the loop requires a new
 | stop condition (*num* is looked at in the switch, though the value
 | is effectively somewhat meaningless).
 | So i've introduced a new command state enum which should get
 | rid of any ambiguities that may arise from missing the complete
 | picture :), and which is used as a break-off command in the loop.
 | 
 | Note that i'm not sure at all about the CC_EOF case; it yet looked
 | at num==-1, but how can that happen there?  Anyway - no more.
 | 
 | This patch applies to v1.69 only with fuzziness, but it is a bit
 | chaotic here at the moment..
 | 
 | What do you think about that?
 
 I like the enum clarity, but I am not sure about removing the EINTR
 code. I also have to check the EOF logic too. Test it some more, and
 I need to find some time here to test it myself. I bit busy since I
 just came back from vacation...
 
 christos