iconvdata/tst-table.sh

   1 #!/bin/sh
   2 # Copyright (C) 2000-2024 Free Software Foundation, Inc.
   3 # This file is part of the GNU C Library.
   4 #
   5
   6 # The GNU C Library is free software; you can redistribute it and/or
   7 # modify it under the terms of the GNU Lesser General Public
   8 # License as published by the Free Software Foundation; either
   9 # version 2.1 of the License, or (at your option) any later version.
  10
  11 # The GNU C Library is distributed in the hope that it will be useful,
  12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14 # Lesser General Public License for more details.
  15
  16 # You should have received a copy of the GNU Lesser General Public
  17 # License along with the GNU C Library; if not, see
  18 # <https://www.gnu.org/licenses/>.
  19
  20 # Checks that the iconv() implementation (in both directions) for a
  21 # stateless encoding agrees with the charmap table.
  22
  23 common_objpfx=$1
  24 objpfx=$2
  25 test_program_prefix=$3
  26 charset=$4
  27 charmap=$5
  28
  29 # sort is used on the build system.
  30 LC_ALL=C
  31 export LC_ALL
  32
  33 set -e
  34
  35 # Get the charmap.
  36 ./tst-table-charmap.sh \
  37   < ../localedata/charmaps/${charmap:-$charset} \
  38   > ${objpfx}tst-${charset}.charmap.table
  39 # When the charset is GB18030, truncate this table because for this encoding,
  40 # the charmap contains ranges (<Unnnn>..<Ummmm> notation), which the
  41 # tst-table-charmap.sh script does not grok.
  42 if test ${charset} = GB18030; then
  43   grep '0x....$' < ${objpfx}tst-${charset}.charmap.table \
  44     > ${objpfx}tst-${charset}.truncated.table
  45   mv ${objpfx}tst-${charset}.truncated.table ${objpfx}tst-${charset}.charmap.table
  46 fi
  47
  48 # Precomputed expected differences between the charmap and iconv forward.
  49 precomposed=${charset}.precomposed
  50
  51 # Precompute expected differences between the charmap and iconv backward.
  52 if test ${charset} = EUC-TW; then
  53   irreversible=${objpfx}tst-${charset}.irreversible
  54   (grep '^0x8EA1' ${objpfx}tst-${charset}.charmap.table
  55    cat ${charset}.irreversible
  56   ) > ${irreversible}
  57 else
  58   irreversible=${charset}.irreversible
  59 fi
  60
  61 # iconv in one direction.
  62 ${test_program_prefix} \
  63 ${objpfx}tst-table-from ${charset} \
  64   > ${objpfx}tst-${charset}.table
  65
  66 # iconv in the other direction.
  67 ${test_program_prefix} \
  68 ${objpfx}tst-table-to ${charset} | sort \
  69   > ${objpfx}tst-${charset}.inverse.table
  70
  71 # Difference between the charmap and iconv backward.
  72 diff ${objpfx}tst-${charset}.charmap.table ${objpfx}tst-${charset}.inverse.table | \
  73   grep '^[<>]' | sed -e 's,^. ,,' > ${objpfx}tst-${charset}.irreversible.table
  74
  75 # Check 1: charmap and iconv forward should be identical, except for
  76 # precomposed characters.
  77 { if test -f ${precomposed}; then
  78     cat ${objpfx}tst-${charset}.table ${precomposed} | sort | uniq -u
  79   else
  80     cat ${objpfx}tst-${charset}.table
  81   fi
  82 } | { if test ${charset} = GB18030; then grep '0x....$'; else cat; fi; } \
  83   > ${objpfx}tst-${charset}.tmp1.table
  84 cmp -s ${objpfx}tst-${charset}.charmap.table ${objpfx}tst-${charset}.tmp1.table ||
  85   exit 1
  86
  87 # Check 2: the difference between the charmap and iconv backward.
  88 { if test -f ${irreversible}; then
  89     cat ${objpfx}tst-${charset}.charmap.table ${irreversible} | sort | uniq -u
  90   else
  91     cat ${objpfx}tst-${charset}.charmap.table
  92   fi
  93 } | { if test ${charset} = GB18030; then grep '0x....$'; else cat; fi; } \
  94   > ${objpfx}tst-${charset}.tmp2c.table
  95 cat ${objpfx}tst-${charset}.inverse.table \
  96   | { if test ${charset} = GB18030; then grep '0x....$'; else cat; fi; } \
  97   > ${objpfx}tst-${charset}.tmp2i.table
  98 cmp -s ${objpfx}tst-${charset}.tmp2c.table ${objpfx}tst-${charset}.tmp2i.table ||
  99   exit 1
 100
 101 # Check 3: the difference between iconv forward and iconv backward. This is
 102 # necessary only for GB18030, because ${objpfx}tst-${charset}.charmap.table
 103 # is truncated for this encoding (see above).
 104 if test ${charset} = GB18030; then
 105   { if test -f ${irreversible}; then
 106       cat ${objpfx}tst-${charset}.table ${irreversible} | sort | uniq -u
 107     else
 108       cat ${objpfx}tst-${charset}.table
 109     fi
 110   } > ${objpfx}tst-${charset}.tmp3.table
 111   cmp -s ${objpfx}tst-${charset}.tmp3.table ${objpfx}tst-${charset}.inverse.table ||
 112     exit 1
 113 fi
 114
 115 exit 0