sort -u data loss... who would have guessed? This is not your typical coreutils bug. Actually, there were *two* sort -u bugs. How many more are lurking?
Anyhow, here's a snapshot. Please give this a try, and report any problems. I will probably release coreutils-8.19 tomorrow. coreutils snapshot: http://meyering.net/cu/coreutils-ss.tar.xz 4.9 MB http://meyering.net/cu/coreutils-ss.tar.xz.sig http://meyering.net/cu/coreutils-8.18.10-46444.tar.xz There are .sig files here, too: http://people.redhat.com/meyering/cu/coreutils-ss.tar.xz http://people.redhat.com/meyering/cu/coreutils-8.18.10-46444.tar.xz ======================================================== NEWS ** Bug fixes df now fails when the list of mounted file systems (/etc/mtab) cannot be read, yet the file system type information is needed to process certain options like -a, -l, -t and -x. [This bug was present in "the beginning".] sort -u could fail to output one or more result lines. For example, this command would fail to print "1": (yes 7 | head -11; echo 1) | sort --p=1 -S32b -u [bug introduced in coreutils-8.6] sort -u could read freed memory. For example, this evokes a read from freed memory: perl -le 'print "a\n"."0"x900'|valgrind sort --p=1 -S32b -u>/dev/null [bug introduced in coreutils-8.6] ======================================================== Changes in coreutils since v8.18: Bernhard Voelker (1): df: fail when the mount list is required but cannot be read Jim Meyering (7): maint: post-release administrivia maint: fix comment grammar to placate make syntax-check maint: correct a stale comment in sort.c tests: reverse args in Coreutils.pm-invoked diff, for consistency sort: sort --unique (-u) could cause data loss tests: wrap the valgrind-requiring assertion in a function tests: trigger the sort -u free-memory-read bug Krzysztof Goj (1): rm: new option --dir (-d) to remove empty directories Paul Eggert (1): sort: simpler fix for sort -u data-loss bug, and for a FMR bug ======================================================== Changes in gnulib since v8.18: * gnulib 39cedf6...bc33a8a (9): > test-parse-datetime: avoid glibc leap-second glitch > autoupdate > gnulib-tool: Fix indentation of generated gnulib-comp.m4 file. > ldexp: relax license > update from texinfo > gnulib-tool: Fix persistence of --witness-c-macro option. > count-leading-zeros: use a lookup table on non-gcc compilers > count-leading-zeros: new module > maintainer-makefile: Fix syntax error with dash.
