m4-patches
[Top][All Lists]
Advanced

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

Re: seekable stdin test failure on OS X


From: Gary V . Vaughan
Subject: Re: seekable stdin test failure on OS X
Date: Tue, 10 Apr 2007 21:19:45 +0100

Hi Eric,

On 10 Apr 2007, at 04:44, Eric Blake wrote:
According to Eric Blake on 4/9/2007 8:47 PM:
However, for now I will omit the lseek (and thus the
dependence on the unistd module), check in the modified patch (attached),
and see how the module fares in testing.

And here's the followup for m4 to use it.  Gary, if you could, please
rerun 'gnulib-tool --import' (or bootstrap), and see if it helps solve
your test failure?

Nope, sorry :-(

#                             -*- compilation -*-
68. others.at:458: testing ...
../../tests/others.at:467: m4 -b -d  < in.m4
--- -   2007-04-10 16:20:25.000000000 +0100
+++ /Users/gary/Devo/Source/m4--devo--0/+build/tests/testsuite.dir/at- stdout
2007-04-10 16:20:25.000000000 +0100
@@ -1,2 +1 @@
-m4exit(15)
68. others.at:458: 68. stdin seekable (others.at:458): FAILED (others.at:467)

And rerunning the test by hand yields:

] cd tests/testsuite.dir/068/
] ls
total 12
4 in.m4  4 run  4 testsuite.log
] M4MODPATH=/Users/gary/Devo/Source/m4--devo--0/+build/ modules ../../../src/m4 -b -d - < in.m4
] echo $?
0

cpp turns fflush.c into something like this:

int
rpl_fflush (FILE *stream)
{
  int e1;
  int e2;
  int result;





  e1 = (*__error());
  result = fflush (stream);
  if (! stream || result == 0 || (*__error()) != 9)
    return result;


  e2 = (*__error());
  if (fseeko (stream, 0, 1) != 0)
    {
      (*__error()) = e2;
      return (-1);
    }




  (*__error()) = e1;
  result = fpurge (stream);
# 95 "../gnu/fflush.c"
  return result;
}


2007-04-09  Eric Blake  <address@hidden>

        Address failure of 'seekable stdin' on Mac OS X.
        * ltdl/m4/gnulib-cache.m4: Augment with 'gnulib-tool --import
        fflush.

Unfortunately, I can't even fix today's tree by hardcoding an fpurge
as I did when I reported the bug earlier.  I'll have time to roll
back commits and figure out where the problem occurred after the weekend.

In the meantime, I'll be happy to run tests and report results if
you'd like to debug it remotely...

Cheers,
        Gary
--
  ())_.              Email me: address@hidden
  ( '/           Read my blog: http://blog.azazil.net
  / )=         ...and my book: http://sources.redhat.com/autobook
`(_~)_ Join my AGLOCO Network: http://www.agloco.com/r/BBBS7912




Attachment: PGP.sig
Description: This is a digitally signed message part


reply via email to

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