Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Tom Tromey via Gdb-patches <gdb-patches@sourceware.org>
To: gdb-patches@sourceware.org
Cc: Tom Tromey <tromey@adacore.com>
Subject: [PATCH 11/14] TUI stdout buffering cleanup
Date: Fri, 12 Aug 2022 18:54:39 -0600	[thread overview]
Message-ID: <20220813005442.4163512-12-tromey@adacore.com> (raw)
In-Reply-To: <20220813005442.4163512-1-tromey@adacore.com>

The TUI checks against gdb_stdout to decide when to buffer.  It seems
much cleaner to me to simply record this as an attribute of the stream
itself.
---
 gdb/tui/tui-file.c | 23 +++--------------------
 gdb/tui/tui-file.h | 10 +++++++++-
 gdb/tui/tui-io.c   |  4 ++--
 3 files changed, 14 insertions(+), 23 deletions(-)

diff --git a/gdb/tui/tui-file.c b/gdb/tui/tui-file.c
index 49dafce112c..3cf3dbc171a 100644
--- a/gdb/tui/tui-file.c
+++ b/gdb/tui/tui-file.c
@@ -22,24 +22,11 @@
 #include "tui/tui-command.h"
 #include "tui.h"
 
-tui_file::tui_file (FILE *stream)
-  : stdio_file (stream)
-{}
-
-/* All TUI I/O sent to the *_filtered and *_unfiltered functions
-   eventually ends up here.  The fputs_unfiltered_hook is primarily
-   used by GUIs to collect all output and send it to the GUI, instead
-   of the controlling terminal.  Only output to gdb_stdout and
-   gdb_stderr are sent to the hook.  Everything else is sent on to
-   fputs to allow file I/O to be handled appropriately.  */
-
 void
 tui_file::puts (const char *linebuffer)
 {
   tui_puts (linebuffer);
-  /* gdb_stdout is buffered, and the caller must gdb_flush it at
-     appropriate times.  Other streams are not so buffered.  */
-  if (this != gdb_stdout)
+  if (!m_buffered)
     tui_refresh_cmd_win ();
 }
 
@@ -47,18 +34,14 @@ void
 tui_file::write (const char *buf, long length_buf)
 {
   tui_write (buf, length_buf);
-  /* gdb_stdout is buffered, and the caller must gdb_flush it at
-     appropriate times.  Other streams are not so buffered.  */
-  if (this != gdb_stdout)
+  if (!m_buffered)
     tui_refresh_cmd_win ();
 }
 
 void
 tui_file::flush ()
 {
-  /* gdb_stdout is buffered.  Other files are always flushed on
-     every write.  */
-  if (this == gdb_stdout)
+  if (m_buffered)
     tui_refresh_cmd_win ();
   stdio_file::flush ();
 }
diff --git a/gdb/tui/tui-file.h b/gdb/tui/tui-file.h
index 1b28780c51b..ff60ded71c1 100644
--- a/gdb/tui/tui-file.h
+++ b/gdb/tui/tui-file.h
@@ -26,11 +26,19 @@
 class tui_file : public stdio_file
 {
 public:
-  explicit tui_file (FILE *stream);
+  tui_file (FILE *stream, bool buffered)
+    : stdio_file (stream),
+      m_buffered (buffered)
+  {}
 
   void write (const char *buf, long length_buf) override;
   void puts (const char *) override;
   void flush () override;
+
+private:
+
+  /* True if this stream is buffered.  */
+  bool m_buffered;
 };
 
 #endif /* TUI_TUI_FILE_H */
diff --git a/gdb/tui/tui-io.c b/gdb/tui/tui-io.c
index 9f27f8bcc01..0efaf69335c 100644
--- a/gdb/tui/tui-io.c
+++ b/gdb/tui/tui-io.c
@@ -904,8 +904,8 @@ tui_initialize_io (void)
 #endif
 
   /* Create tui output streams.  */
-  tui_stdout = new pager_file (new tui_file (stdout));
-  tui_stderr = new tui_file (stderr);
+  tui_stdout = new pager_file (new tui_file (stdout, true));
+  tui_stderr = new tui_file (stderr, false);
   tui_stdlog = new timestamped_file (tui_stderr);
   tui_out = new tui_ui_out (tui_stdout);
 
-- 
2.34.1


  parent reply	other threads:[~2022-08-13  0:55 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-13  0:54 [PATCH 00/14] Minor ui / interp cleanups Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 01/14] Remove some dead code Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 02/14] Free ui::line_buffer Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 03/14] Use ui_out_redirect_pop in more places Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 04/14] Remove the "for moment" comments Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 05/14] Remove obsolete filtering comment Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 06/14] Remove two unused members from mi_interp Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 07/14] Use member initialization in 'struct ui' Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 08/14] Use scoped_restore in safe_parse_type Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 09/14] Remove tui_out_new Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 10/14] Remove a ui-related memory leak Tom Tromey via Gdb-patches
2022-08-13  0:54 ` Tom Tromey via Gdb-patches [this message]
2022-08-13  0:54 ` [PATCH 12/14] Remove a call to clear_interpreter_hooks Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 13/14] Fix "source" with interpreter-exec Tom Tromey via Gdb-patches
2022-08-13  1:58   ` Enze Li via Gdb-patches
2022-08-15 17:28     ` Tom Tromey via Gdb-patches
2022-08-13  0:54 ` [PATCH 14/14] Fix interpreter-exec crash Tom Tromey via Gdb-patches
2022-08-31 17:13 ` [PATCH 00/14] Minor ui / interp cleanups Tom Tromey

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220813005442.4163512-12-tromey@adacore.com \
    --to=gdb-patches@sourceware.org \
    --cc=tromey@adacore.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox