FYI,
[no time now -- will investigate tomorrow if no one beats me to it]
test-getndelim2.c:141: assertion failed
/bin/bash: line 5: 1259 Abort (core dumped) EXEEXT=''
EXEEXT='
' EXEEXT='' LOCALE_FR='none' LOCALE_TR_UTF8='none' EXEEXT='' EXEEXT='' EXEEXT=''
srcdir='.' EXEEXT=''
Here's a fix for the first:
test-getaddrinfo.c: In function 'simple':
test-getaddrinfo.c:67: error: 'EAI_NODATA' undeclared (first use in this
function)
test-getaddrinfo.c:67: error: (Each undeclared identifier is reported only
once
test-getaddrinfo.c:67: error: for each function
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Jim Meyering wrote:
>> Actually this seems more appropriately run from a checked-out
>> gnulib directory, e.g., as part of a "make check" rule.
>
> I would much prefer if the gnulib/Makefile "check" rule be reserved
> to platform independent checks on the g
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Btw, the useless-if-before-free help says that
>
> Exit status:
>
> 0 no match
> 1 one or more matches
>
> but it appears to be the opposite: 0 for one or more matches (like 'grep').
Yep. Good catch.
It's fixed now.
Thanks.
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Jim Meyering wrote:
>> ... possibly get a test failure. Such a failure doesn't
>> really matter in general -- most people don't run "make syntax-check"
>> or "make distcheck" -- so a test failure here is most definitely no
>> cause for alarm
>
> I routinel
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Hi Jim,
>
>> * tests/test-vc-list-files-cvs.sh: New file.
>> * tests/test-vc-list-files-git.sh: New file.
>
> The two tests fail for me, on Linux, in a testdir created through
> gnulib-tool --create-testdir --with-tests ...
>
> ./test-vc-list-files-cvs.s
Bruno Haible clisp.org> writes:
> 3. At least one byte is read, but not stored (due to nmax), and no error.
> Return value was 0 and is now -1.
You are right that the program can usefully distinguish between -1 and 0 here
(with getdelim, there is no nmax parameter, hence getdelim can nev
Jim Meyering wrote:
> Actually this seems more appropriately run from a checked-out
> gnulib directory, e.g., as part of a "make check" rule.
I would much prefer if the gnulib/Makefile "check" rule be reserved
to platform independent checks on the gnulib code. Tests of some programs
belong in the
Hi Jim,
> * tests/test-vc-list-files-cvs.sh: New file.
> * tests/test-vc-list-files-git.sh: New file.
The two tests fail for me, on Linux, in a testdir created through
gnulib-tool --create-testdir --with-tests ...
./test-vc-list-files-cvs.sh: line 57: vc-list-files: command not found
--- expec
Jim Meyering wrote:
> ... possibly get a test failure. Such a failure doesn't
> really matter in general -- most people don't run "make syntax-check"
> or "make distcheck" -- so a test failure here is most definitely no
> cause for alarm
I routinely use "gnulib-tool --create-testdir --with-tests"
Eric Blake wrote:
> $ time (~/coreutils/src/cut -f 2-3 < data > data2)
>
> real 0m4.168s
> user 0m3.952s
> sys 0m0.109s
>
> $ time (/bin/cut -f 2-3 < data > data1)
>
> real 0m6.194s
> user 0m6.109s
> sys 0m0.108s
Impressing!!
However, I have to follow up with a fix for the reallocation
Eric Blake wrote:
> And all I did was make getndelim2 smarter to process data in blocks
> instead of bytes
I'm adding these comments. Otherwise, it is impossible to understand why
there is no buffer overrun at the end (when the terminating '\0' is stored).
*** lib/getndelim2.c.orig 2008-05
Jim Meyering <[EMAIL PROTECTED]> wrote:
> I noticed that I'd broken my useless-if-before-free script.
> It would only notice the useless tests if there were
> a cast on the argument to the free-like function.
>
> I've just pushed this correction and will add unit tests.
FYI, I've written a cheap u
Eric Blake wrote:
> It also fixes a bug where when the stream is at EOF, sometimes the
> function returned 0 and sometimes -1; the patch makes sure EOF always
> results in -1 (similar to the Posix 200x specification of getdelim always
> returning -1 at EOF).
>
> - return bytes_stored;
> + return
Simon Josefsson wrote:
> >> +-e 's|@''HAVE_DECL_INET_NTOP''@|$(HAVE_DECL_INET_NTOP)|g' \
> >> +-e 's|@''HAVE_DECL_INET_PTON''@|$(HAVE_DECL_INET_PTON)|g' \
> >> < $(srcdir)/arpa_inet.in.h; \
> >>} > [EMAIL PROTECTED]
> >>mv [EMAIL PROTECTED] $@
> >
> > This cannot be
Eric Blake wrote:
> But 'man valgrind' states:
> --suppressions= [default: $PREFIX/lib/valgrind/default.supp]
> Specifies an extra file from which to read
> descriptions of errors to suppress. You may spec-
> ify up to 10 additional suppression
16 matches
Mail list logo
