Re: Bug#545422: coreutils: "tail -f -" fails

[Jim Meyering]

Giuseppe Scrivano wrote:
> what do you think about the following solution?  It avoids to revert to
> the "old" polling mechanism using "/dev/stdin" instead of "-" to
> inotify_add_watch.

I considered that and discussed the
trade-off in the comment I committed.
There have been systems and configurations with
nonexistent and unusable /dev/stdin files.

commit cdfb703c5da31a798557722df516e0d01dac828a
Author: Jim Meyering <address@hidden>
Date:   Mon Sep 7 08:37:08 2009 +0200

    tail -f: handle "-"/stdin once again

    * src/tail.c (main) [HAVE_INOTIFY]: When stdin (i.e., "-", or no args,
    but not /dev/stdin) is specified on the command line, don't use inotify.
    Reported by Bill Brelsford in <http://bugs.debian.org/545422>.
    * tests/tail-2/follow-stdin: New file.  Test for this.
    * tests/Makefile.am (TESTS): Add the test.
    * NEWS (Bug fixes): Mention it.
    This bug was introduced in coreutils-7.5 via commit ae494d4b,
    2009-06-02, "tail: use inotify if it is available".

diff --git a/src/tail.c b/src/tail.c
index e3b9529..c53df9e 100644
--- a/src/tail.c
+++ b/src/tail.c
@@ -1982,7 +1982,19 @@ main (int argc, char **argv)
   if (forever)
     {
 #if HAVE_INOTIFY
-      if (!disable_inotify)
+      /* If the user specifies stdin via a command line argument of "-",
+         or implicitly by providing no arguments, we won't use inotify.
+         Technically, on systems with a working /dev/stdin, we *could*,
+         but would it be worth it?  Verifying that it's a real device
+         and hooked up to stdin is not trivial, while reverting to
+         non-inotify-based tail_forever is easy and portable.  */
+      bool stdin_cmdline_arg = false;
+
+      for (i = 0; i < n_files; i++)
+        if (STREQ (file[i], "-"))
+          stdin_cmdline_arg = true;
+
+      if (!disable_inotify && !stdin_cmdline_arg)
         {
           int wd = inotify_init ();
           if (wd < 0)