tr: fix various issues with case conversion

This valid translation spec aborted:
  LC_ALL=en_US.iso-8859-1 tr '[:upper:]- ' '[:lower:]_'
This invalid translation spec aborted:
  LC_ALL=en_US.iso-8859-1 tr '[:upper:] '  '[:lower:]'
This was caused by commit 6efd1046, 05-01-2008,
"Avoid tr case-conversion failure in some locales"

This misaligned conversion spec was allowed:
  LC_ALL=C tr 'A-Y[:lower:]' 'a-z[:upper:]'
This was caused by commit af5d0c36, 21-10-2007,
"tr: do not reject an unmatched [:lower:] or [:upper:] in SET1"

This misaligned spec was allowed by extending the class:
  LC_ALL=C tr '[:upper:] ' '[:lower:]'

* src/tr.c (validate_case_classes): A new function to check
alignment of case conversion classes.  Also it adjusts the
length of the sets so that locales with different numbers of
upper and lower case characters, don't cause issues.
(string2_extend): Disallow extending the case conversion
class as in the above example.  That is locale dependent
and most likely not what the user wants.
(validate): Do the simple test for "restricted" char classes
earlier, so we don't redundantly do more expensive validation.
(main): Remove the case class validation, and simplify.
* tests/misc/tr-case-class: A new test to test the various
alignment and locale issues, associated with case conversion.
* tests/misc/tr: Move case conversion tests to new tr-case-class.
* tests/Makefile.am: Reference the new test.
* NEWS: Mention the fixes.

1 files changed, 112 insertions, 0 deletions

diff --git a/tests/misc/tr-case-class b/tests/misc/tr-case-class
new file mode 100755
index 000000000..d81c676c3
--- /dev/null
+++ b/tests/misc/tr-case-class
@@ -0,0 +1,112 @@
+#!/bin/sh
+# Test case conversion classes
+
+# Copyright (C) 2010 Free Software Foundation, Inc.
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+. "${srcdir=.}/init.sh"; path_prepend_ ../src
+
+# Ensure we support translation of case classes with extension
+echo '01234567899999999999999999' > exp
+echo 'abcdefghijklmnopqrstuvwxyz' |
+tr '[:lower:]' '0-9' > out || fail=1
+compare out exp || fail=1
+echo 'abcdefghijklmnopqrstuvwxyz' |
+tr '[:lower:][:lower:]' '[:upper:]0-9' > out || fail=1
+compare out exp || fail=1
+
+# Validate the alignment of case classes
+tr 'A-Z[:lower:]' 'a-y[:upper:]' < /dev/null && fail=1
+tr '[:upper:][:lower:]' 'a-y[:upper:]' < /dev/null && fail=1
+tr 'A-Y[:lower:]' 'a-z[:upper:]' < /dev/null && fail=1
+tr 'A-Z[:lower:]' '[:lower:][:upper:]' < /dev/null && fail=1
+tr 'A-Z[:lower:]' '[:lower:]A-Z' < /dev/null && fail=1
+tr '[:upper:][:lower:]' 'a-z[:upper:]' < /dev/null || fail=1
+tr '[:upper:][:lower:]' '[:upper:]a-z' < /dev/null || fail=1
+
+# Before coreutils 8.6 the trailing space in string1
+# caused the case class in string2 to be extended.
+# However that was not portable, dependent on locale
+# and in contravention of POSIX.
+tr '[:upper:] ' '[:lower:]' < /dev/null 2>out && fail=1
+echo 'tr: when translating with string1 longer than string2,
+the latter string must not end with a character class' > exp
+compare out exp || fail=1
+
+# Up to coreutils-6.9, tr rejected an unmatched [:lower:] or [:upper:] in SET1.
+echo '#$%123abcABC' | tr '[:lower:]' '[.*]' > out || fail=1
+echo '#$%123...ABC' > exp
+compare out exp || fail=1
+echo '#$%123abcABC' | tr '[:upper:]' '[.*]' > out || fail=1
+echo '#$%123abc...' > exp
+compare out exp || fail=1
+
+# When doing a case-converting translation with something after the
+# [:upper:] and [:lower:] elements, ensure that tr honors the following byte.
+echo 'abc.' | tr '[:lower:].' '[:upper:]x' > out || fail=1
+echo 'ABCx' > exp
+compare out exp || fail=1
+
+# Before coreutils 8.6 the disparate number of upper and lower
+# characters in some locales, triggered abort()s and invalid behavior
+export LC_ALL=en_US.ISO-8859-1
+
+if test "$(locale charmap 2>/dev/null)" = ISO-8859-1; then
+  # Up to coreutils-6.9.91, this would fail with the diagnostic:
+  # tr: misaligned [:upper:] and/or [:lower:] construct
+  # with LC_CTYPE=en_US.ISO-8859-1.
+  tr '[:upper:]' '[:lower:]' < /dev/null || fail=1
+
+  tr '[:upper:] ' '[:lower:]' < /dev/null 2>out && fail=1
+  echo 'tr: when translating with string1 longer than string2,
+the latter string must not end with a character class' > exp
+  compare out exp || fail=1
+
+  # Ensure when there are a different number of elements
+  # in each string, we validate the case mapping correctly
+  echo 'abc.xyz' |
+  tr 'ab[:lower:]' '0-1[:upper:]' > out || fail=1
+  echo 'ABC.XYZ' > exp
+  compare out exp || fail=1
+
+  # Ensure we extend string2 appropriately
+  echo 'ABC- XYZ' |
+  tr '[:upper:]- ' '[:lower:]_' > out || fail=1
+  echo 'abc__xyz' > exp
+  compare out exp || fail=1
+
+  # Ensure the size of the case classes are accounted
+  # for as a unit.
+  echo 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' |
+  tr '[:upper:]A-B' '[:lower:]0' >out || fail=1
+  echo '00cdefghijklmnopqrstuvwxyz' > exp
+  compare out exp || fail=1
+
+  # Ensure the size of the case classes are accounted
+  # for as a unit.
+  echo 'a' |
+  tr -t '[:lower:]a' '[:upper:]0' >out || fail=1
+  echo '0' > exp
+  compare out exp || fail=1
+
+  # Ensure the size of the case classes are accounted
+  # for as a unit.
+  echo 'a' |
+  tr -t '[:lower:][:lower:]a' '[:lower:][:upper:]0' >out || fail=1
+  echo '0' > exp
+  compare out exp || fail=1
+fi
+
+Exit $fail