Re: [rfa/tui] Fix bitrot in tui/tui-file.c

[Syd Polk <spolk@cygnus.com>]

Approved.

Andrew Cagney wrote:
> 
> Hello,
> 
> When the TUI code was moved (by me :-/) out of gdb/utils.c into
> gdb/tui/tui-file.c it was never compiled.  The attatched fixes that.
> 
> Ok?
>         Andrew
> 
>   --------------------------------------------------------------------------------
> Thu May 25 14:46:20 2000  Andrew Cagney  <cagney@b1.cygnus.com>
> 
>         * tui-file.c: Include "tui.h", "tuiData.h", "tuiIO.h" and
>         "tuiCommand.h".
>         (tui_file_fputs): Pass ``file'' and not ``stream'' to
>         tui_file_adjust_strbuf.
> 
> Index: tui/tui-file.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/tui/tui-file.c,v
> retrieving revision 1.1.1.2
> diff -p -r1.1.1.2 tui-file.c
> *** tui-file.c  2000/02/02 00:21:19     1.1.1.2
> --- tui-file.c  2000/05/25 06:16:13
> ***************
> *** 22,27 ****
> --- 22,34 ----
>   #include "ui-file.h"
>   #include "tui/tui-file.h"
> 
> + #ifdef TUI
> + #include "tui.h"
> + #include "tuiData.h"
> + #include "tuiIO.h"
> + #include "tuiCommand.h"
> + #endif
> +
>   #include <string.h>
> 
>   /* Called instead of fputs for all TUI_FILE output.  */
> *************** tui_file_fputs (linebuffer, file)
> *** 196,202 ****
> 
>           if (stream->ts_streamtype == astring)
>             {
> !             tui_file_adjust_strbuf (strlen (linebuffer), stream);
>               strcat (stream->ts_strbuf, linebuffer);
>             }
>           else
> --- 203,209 ----
> 
>           if (stream->ts_streamtype == astring)
>             {
> !             tui_file_adjust_strbuf (strlen (linebuffer), file);
>               strcat (stream->ts_strbuf, linebuffer);
>             }
>           else
> *************** tui_file_fputs (linebuffer, file)
> *** 215,221 ****
>           /* The normal case - just do a fputs() */
>           if (stream->ts_streamtype == astring)
>             {
> !             tui_file_adjust_strbuf (strlen (linebuffer), stream);
>               strcat (stream->ts_strbuf, linebuffer);
>             }
>           else
> --- 222,228 ----
>           /* The normal case - just do a fputs() */
>           if (stream->ts_streamtype == astring)
>             {
> !             tui_file_adjust_strbuf (strlen (linebuffer), file);
>               strcat (stream->ts_strbuf, linebuffer);
>             }
>           else
From broeker@physik.rwth-aachen.de Thu May 25 09:52:00 2000
From: Hans-Bernhard Broeker <broeker@physik.rwth-aachen.de>
To: gdb-patches@sourceware.cygnus.com
Subject: gdb5.0 on i586-pc-linux-gnulibc1: no <poll.h>, but <sys/poll.h>
Date: Thu, 25 May 2000 09:52:00 -0000
Message-id: <Pine.LNX.4.10.10005251840290.23601-100000@acp3bf>
X-SW-Source: 2000-05/msg00391.html
Content-length: 642

Hello, everyone.

System: i586-cp-linux-gnulibc1 (libc5 5.4.33, linux 2.0.32)
gdb version: gdb-5.0 (final release)

I just found a minor glitch in building gdb5.0: it did (correctly)
determine that this machine I build on does have the poll() function. But
there's no check as to where the appropriate header file might be: it's
hardwired to #include <poll.h> whenever HAVE_POLL is set. On the
admittedly outdated system Linux mentioned, it should #include
<sys/poll.h>. Changing that the brute-force way gave me a working 
gdb...



Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.