Hi Ralf, > -----Original Message----- > From: Ralf Wildenhues [mailto:ralf.wildenh...@gmx.de] > > Thank you for the bug report. I think this is a coreutils bug or two. > configure is working as intended in substituting a multiline value > (Autoconf supports multiline values since a couple of years), but the > configure.ac bits from coreutils somehow fail to flatten the value of > $MANS. > > coreutils/configure.ac contains something like this: > [...] > > which should remove newlines from $MAN altogether. There is more > munging of $optional_bin_progs earlier, and through the macros in > m4/include-exclude-prog.m4. I don't quite see yet where this code > should fail. Likely another 'tr' incompatibility we are not aware > of yet? Can you try to look into this? I cannot reproduce it on > Tru64/OSF 4.0D. ($PATH would be interesting, too, to see which tr > you are using).
The PATH just gets me /usr/bin/tr. I've instrumented the configure script a bit... ----begin hacked-up configure snippet---- echo "optional_bin_progs = >>$optional_bin_progs<<" MAN=`echo "$optional_bin_progs "|sed 's/ /.1 /g;s/ $//'|tr -d '\\015\\012'` echo "MAN (1) = >>$MAN<<" # Change ginstall.1 to "install.h" in $MAN. MAN=`for m in $MAN; do test $m = ginstall.1 && m=install.1; echo $m; done \ | tr '\015\012' ' '; echo` echo "MAN (2) = >>$MAN<<" # Remove [.1, since writing a portable rule for it in man/Makefile.am # is not practical. The sed LHS below uses the autoconf quadrigraph # representing '['. MAN=`echo "$MAN"|sed 's/\[\.1//'` echo "MAN (3) = >>$MAN<<" ----end hacked-up configure snippet---- ...and obtained some interesting results: ----begin configure output excerpt---- checking for getspnam... no checking for library containing crypt... none required optional_bin_progs = >>uname chroot hostid nice who users pinky uptime stty df [ chcon chgrp chown chmod cp dd dircolors du \ ginstall link ln dir vdir ls mkdir \ mkfifo mknod mktemp \ mv nohup readlink rm rmdir shred stat sync touch unlink \ cat cksum comm csplit cut expand fmt fold head join groups md5sum \ nl od paste pr ptx sha1sum sha224sum sha256sum sha384sum sha512sum \ shuf sort split sum tac tail tr tsort unexpand uniq wc \ basename date dirname echo env expr factor false \ id kill logname pathchk printenv printf pwd \ runcon seq sleep tee \ test true tty whoami yes \ base64<< MAN (1) = >>uname.1 chroot.1 hostid.1 nice.1 who.1 users.1 pinky.1 uptime.1 stty.1 df.1 [.1 chcon.1 chgrp.1 chown.1 chmod.1 cp.1 dd.1 dircolors.1 du.1 \.1 ginstall.1 link.1 ln.1 dir.1 vdir.1 ls.1 mkdir.1 \.1 mkfifo.1 mknod.1 mktemp.1 \.1 mv.1 nohup.1 readlink.1 rm.1 rmdir.1 shred.1 stat.1 sync.1 touch.1 unlink.1 \.1 cat.1 cksum.1 comm.1 csplit.1 cut.1 expand.1 fmt.1 fold.1 head.1 join.1 groups.1 md5sum.1 \.1 nl.1 od.1 paste.1 pr.1 ptx.1 sha1sum.1 sha224sum.1 sha256sum.1 sha384sum.1 sha512sum.1 \.1 shuf.1 sort.1 split.1 sum.1 tac.1 tail.1 tr.1 tsort.1 unexpand.1 uniq.1 wc.1 \.1 basename.1 date.1 dirname.1 echo.1 env.1 expr.1 factor.1 false.1 \.1 id.1 kill.1 logname.1 pathchk.1 printenv.1 printf.1 pwd.1 \.1 runcon.1 seq.1 sleep.1 tee.1 \.1 test.1 true.1 tty.1 whoami.1 yes.1 \.1 base64.1<< MAN (2) = >>uname.1 chroot.1 hostid.1 nice.1 who.1 users.1 pinky.1 uptime.1 stty.1 df.1 [.1 chcon.1 chgrp.1 chown.1 chmod.1 cp.1 dd.1 dircolors.1 du.1 \.1 install.1 link.1 ln.1 [...] tty.1 whoami.1 yes.1 \.1 base64.1<< MAN (3) = >>uname.1 chroot.1 hostid.1 nice.1 who.1 users.1 pinky.1 uptime.1 stty.1 df.1 chcon.1 chgrp.1 chown.1 chmod.1 cp.1 dd.1 dircolors.1 du.1 \.1 install.1 link.1 ln.1 [...] tty.1 whoami.1 yes.1 \.1 base64.1<< checking whether NLS is requested... no checking for msgfmt... no ----end configure output excerpt---- Here is the real problem---not gawk(1), but indeed tr(1): % ( echo foo ; echo bar ) | tr '\015\012' ' ' foo bar % ( echo foo ; echo bar ) | tr '\015\012' ' ' <---- note the two-space string foo bar % The expected "SET2 is extended to length of SET1 by repeating its last character as necessary" thing is not happening! --Daniel _______________________________________________ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
