t/t0028-working-tree-encoding.sh

   1 #!/bin/sh
   2
   3 test_description='working-tree-encoding conversion via gitattributes'
   4
   5 GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME=main
   6 export GIT_TEST_DEFAULT_INITIAL_BRANCH_NAME
   7
   8 . ./test-lib.sh
   9
  10 GIT_TRACE_WORKING_TREE_ENCODING=1 && export GIT_TRACE_WORKING_TREE_ENCODING
  11
  12 test_lazy_prereq NO_UTF16_BOM '
  13         test $(printf abc | iconv -f UTF-8 -t UTF-16 | wc -c) = 6
  14 '
  15
  16 test_lazy_prereq NO_UTF32_BOM '
  17         test $(printf abc | iconv -f UTF-8 -t UTF-32 | wc -c) = 12
  18 '
  19
  20 write_utf16 () {
  21         if test_have_prereq NO_UTF16_BOM
  22         then
  23                 printf '\376\377'
  24         fi &&
  25         iconv -f UTF-8 -t UTF-16
  26 }
  27
  28 write_utf32 () {
  29         if test_have_prereq NO_UTF32_BOM
  30         then
  31                 printf '\0\0\376\377'
  32         fi &&
  33         iconv -f UTF-8 -t UTF-32
  34 }
  35
  36 test_expect_success 'setup test files' '
  37         git config core.eol lf &&
  38
  39         text="hallo there!\ncan you read me?" &&
  40         echo "*.utf16 text working-tree-encoding=utf-16" >.gitattributes &&
  41         echo "*.utf16lebom text working-tree-encoding=UTF-16LE-BOM" >>.gitattributes &&
  42         printf "$text" >test.utf8.raw &&
  43         printf "$text" | write_utf16 >test.utf16.raw &&
  44         printf "$text" | write_utf32 >test.utf32.raw &&
  45         printf "\377\376"                         >test.utf16lebom.raw &&
  46         printf "$text" | iconv -f UTF-8 -t UTF-16LE >>test.utf16lebom.raw &&
  47
  48         # Line ending tests
  49         printf "one\ntwo\nthree\n" >lf.utf8.raw &&
  50         printf "one\r\ntwo\r\nthree\r\n" >crlf.utf8.raw &&
  51
  52         # BOM tests
  53         printf "\0a\0b\0c"                         >nobom.utf16be.raw &&
  54         printf "a\0b\0c\0"                         >nobom.utf16le.raw &&
  55         printf "\376\377\0a\0b\0c"                 >bebom.utf16be.raw &&
  56         printf "\377\376a\0b\0c\0"                 >lebom.utf16le.raw &&
  57         printf "\0\0\0a\0\0\0b\0\0\0c"             >nobom.utf32be.raw &&
  58         printf "a\0\0\0b\0\0\0c\0\0\0"             >nobom.utf32le.raw &&
  59         printf "\0\0\376\377\0\0\0a\0\0\0b\0\0\0c" >bebom.utf32be.raw &&
  60         printf "\377\376\0\0a\0\0\0b\0\0\0c\0\0\0" >lebom.utf32le.raw &&
  61
  62         # Add only UTF-16 file, we will add the UTF-32 file later
  63         cp test.utf16.raw test.utf16 &&
  64         cp test.utf32.raw test.utf32 &&
  65         cp test.utf16lebom.raw test.utf16lebom &&
  66         git add .gitattributes test.utf16 test.utf16lebom &&
  67         git commit -m initial
  68 '
  69
  70 test_expect_success 'ensure UTF-8 is stored in Git' '
  71         test_when_finished "rm -f test.utf16.git" &&
  72
  73         git cat-file -p :test.utf16 >test.utf16.git &&
  74         test_cmp_bin test.utf8.raw test.utf16.git
  75 '
  76
  77 test_expect_success 're-encode to UTF-16 on checkout' '
  78         test_when_finished "rm -f test.utf16.raw" &&
  79
  80         rm test.utf16 &&
  81         git checkout test.utf16 &&
  82         test_cmp_bin test.utf16.raw test.utf16
  83 '
  84
  85 test_expect_success 're-encode to UTF-16-LE-BOM on checkout' '
  86         rm test.utf16lebom &&
  87         git checkout test.utf16lebom &&
  88         test_cmp_bin test.utf16lebom.raw test.utf16lebom
  89 '
  90
  91 test_expect_success 'check $GIT_DIR/info/attributes support' '
  92         test_when_finished "rm -f test.utf32.git" &&
  93         test_when_finished "git reset --hard HEAD" &&
  94
  95         echo "*.utf32 text working-tree-encoding=utf-32" >.git/info/attributes &&
  96         git add test.utf32 &&
  97
  98         git cat-file -p :test.utf32 >test.utf32.git &&
  99         test_cmp_bin test.utf8.raw test.utf32.git
 100 '
 101
 102 for i in 16 32
 103 do
 104         test_expect_success "check prohibited UTF-${i} BOM" '
 105                 test_when_finished "git reset --hard HEAD" &&
 106
 107                 echo "*.utf${i}be text working-tree-encoding=utf-${i}be" >>.gitattributes &&
 108                 echo "*.utf${i}le text working-tree-encoding=utf-${i}LE" >>.gitattributes &&
 109
 110                 # Here we add a UTF-16 (resp. UTF-32) files with BOM (big/little-endian)
 111                 # but we tell Git to treat it as UTF-16BE/UTF-16LE (resp. UTF-32).
 112                 # In these cases the BOM is prohibited.
 113                 cp bebom.utf${i}be.raw bebom.utf${i}be &&
 114                 test_must_fail git add bebom.utf${i}be 2>err.out &&
 115                 test_i18ngrep "fatal: BOM is prohibited .* utf-${i}be" err.out &&
 116                 test_i18ngrep "use UTF-${i} as working-tree-encoding" err.out &&
 117
 118                 cp lebom.utf${i}le.raw lebom.utf${i}be &&
 119                 test_must_fail git add lebom.utf${i}be 2>err.out &&
 120                 test_i18ngrep "fatal: BOM is prohibited .* utf-${i}be" err.out &&
 121                 test_i18ngrep "use UTF-${i} as working-tree-encoding" err.out &&
 122
 123                 cp bebom.utf${i}be.raw bebom.utf${i}le &&
 124                 test_must_fail git add bebom.utf${i}le 2>err.out &&
 125                 test_i18ngrep "fatal: BOM is prohibited .* utf-${i}LE" err.out &&
 126                 test_i18ngrep "use UTF-${i} as working-tree-encoding" err.out &&
 127
 128                 cp lebom.utf${i}le.raw lebom.utf${i}le &&
 129                 test_must_fail git add lebom.utf${i}le 2>err.out &&
 130                 test_i18ngrep "fatal: BOM is prohibited .* utf-${i}LE" err.out &&
 131                 test_i18ngrep "use UTF-${i} as working-tree-encoding" err.out
 132         '
 133
 134         test_expect_success "check required UTF-${i} BOM" '
 135                 test_when_finished "git reset --hard HEAD" &&
 136
 137                 echo "*.utf${i} text working-tree-encoding=utf-${i}" >>.gitattributes &&
 138
 139                 cp nobom.utf${i}be.raw nobom.utf${i} &&
 140                 test_must_fail git add nobom.utf${i} 2>err.out &&
 141                 test_i18ngrep "fatal: BOM is required .* utf-${i}" err.out &&
 142                 test_i18ngrep "use UTF-${i}BE or UTF-${i}LE" err.out &&
 143
 144                 cp nobom.utf${i}le.raw nobom.utf${i} &&
 145                 test_must_fail git add nobom.utf${i} 2>err.out &&
 146                 test_i18ngrep "fatal: BOM is required .* utf-${i}" err.out &&
 147                 test_i18ngrep "use UTF-${i}BE or UTF-${i}LE" err.out
 148         '
 149
 150         test_expect_success "eol conversion for UTF-${i} encoded files on checkout" '
 151                 test_when_finished "rm -f crlf.utf${i}.raw lf.utf${i}.raw" &&
 152                 test_when_finished "git reset --hard HEAD^" &&
 153
 154                 cat lf.utf8.raw | write_utf${i} >lf.utf${i}.raw &&
 155                 cat crlf.utf8.raw | write_utf${i} >crlf.utf${i}.raw &&
 156                 cp crlf.utf${i}.raw eol.utf${i} &&
 157
 158                 cat >expectIndexLF <<-EOF &&
 159                         i/lf    w/-text attr/text               eol.utf${i}
 160                 EOF
 161
 162                 git add eol.utf${i} &&
 163                 git commit -m eol &&
 164
 165                 # UTF-${i} with CRLF (Windows line endings)
 166                 rm eol.utf${i} &&
 167                 git -c core.eol=crlf checkout eol.utf${i} &&
 168                 test_cmp_bin crlf.utf${i}.raw eol.utf${i} &&
 169
 170                 # Although the file has CRLF in the working tree,
 171                 # ensure LF in the index
 172                 git ls-files --eol eol.utf${i} >actual &&
 173                 test_cmp expectIndexLF actual &&
 174
 175                 # UTF-${i} with LF (Unix line endings)
 176                 rm eol.utf${i} &&
 177                 git -c core.eol=lf checkout eol.utf${i} &&
 178                 test_cmp_bin lf.utf${i}.raw eol.utf${i} &&
 179
 180                 # The file LF in the working tree, ensure LF in the index
 181                 git ls-files --eol eol.utf${i} >actual &&
 182                 test_cmp expectIndexLF actual
 183         '
 184 done
 185
 186 test_expect_success 'check unsupported encodings' '
 187         test_when_finished "git reset --hard HEAD" &&
 188
 189         echo "*.set text working-tree-encoding" >.gitattributes &&
 190         printf "set" >t.set &&
 191         test_must_fail git add t.set 2>err.out &&
 192         test_i18ngrep "true/false are no valid working-tree-encodings" err.out &&
 193
 194         echo "*.unset text -working-tree-encoding" >.gitattributes &&
 195         printf "unset" >t.unset &&
 196         git add t.unset &&
 197
 198         echo "*.empty text working-tree-encoding=" >.gitattributes &&
 199         printf "empty" >t.empty &&
 200         git add t.empty &&
 201
 202         echo "*.garbage text working-tree-encoding=garbage" >.gitattributes &&
 203         printf "garbage" >t.garbage &&
 204         test_must_fail git add t.garbage 2>err.out &&
 205         test_i18ngrep "failed to encode" err.out
 206 '
 207
 208 test_expect_success 'error if encoding round trip is not the same during refresh' '
 209         BEFORE_STATE=$(git rev-parse HEAD) &&
 210         test_when_finished "git reset --hard $BEFORE_STATE" &&
 211
 212         # Add and commit a UTF-16 file but skip the "working-tree-encoding"
 213         # filter. Consequently, the in-repo representation is UTF-16 and not
 214         # UTF-8. This simulates a Git version that has no working tree encoding
 215         # support.
 216         echo "*.utf16le text working-tree-encoding=utf-16le" >.gitattributes &&
 217         echo "hallo" >nonsense.utf16le &&
 218         TEST_HASH=$(git hash-object --no-filters -w nonsense.utf16le) &&
 219         git update-index --add --cacheinfo 100644 $TEST_HASH nonsense.utf16le &&
 220         COMMIT=$(git commit-tree -p $(git rev-parse HEAD) -m "plain commit" $(git write-tree)) &&
 221         git update-ref refs/heads/main $COMMIT &&
 222
 223         test_must_fail git checkout HEAD^ 2>err.out &&
 224         test_i18ngrep "error: .* overwritten by checkout:" err.out
 225 '
 226
 227 test_expect_success 'error if encoding garbage is already in Git' '
 228         BEFORE_STATE=$(git rev-parse HEAD) &&
 229         test_when_finished "git reset --hard $BEFORE_STATE" &&
 230
 231         # Skip the UTF-16 filter for the added file
 232         # This simulates a Git version that has no checkoutEncoding support
 233         cp nobom.utf16be.raw nonsense.utf16 &&
 234         TEST_HASH=$(git hash-object --no-filters -w nonsense.utf16) &&
 235         git update-index --add --cacheinfo 100644 $TEST_HASH nonsense.utf16 &&
 236         COMMIT=$(git commit-tree -p $(git rev-parse HEAD) -m "plain commit" $(git write-tree)) &&
 237         git update-ref refs/heads/main $COMMIT &&
 238
 239         git diff 2>err.out &&
 240         test_i18ngrep "error: BOM is required" err.out
 241 '
 242
 243 test_lazy_prereq ICONV_SHIFT_JIS '
 244         iconv -f UTF-8 -t SHIFT-JIS </dev/null
 245 '
 246
 247 test_expect_success ICONV_SHIFT_JIS 'check roundtrip encoding' '
 248         test_when_finished "rm -f roundtrip.shift roundtrip.utf16" &&
 249         test_when_finished "git reset --hard HEAD" &&
 250
 251         text="hallo there!\nroundtrip test here!" &&
 252         printf "$text" | iconv -f UTF-8 -t SHIFT-JIS >roundtrip.shift &&
 253         printf "$text" | write_utf16 >roundtrip.utf16 &&
 254         echo "*.shift text working-tree-encoding=SHIFT-JIS" >>.gitattributes &&
 255
 256         # SHIFT-JIS encoded files are round-trip checked by default...
 257         GIT_TRACE=1 git add .gitattributes roundtrip.shift 2>&1 |
 258                 grep "Checking roundtrip encoding for SHIFT-JIS" &&
 259         git reset &&
 260
 261         # ... unless we overwrite the Git config!
 262         ! GIT_TRACE=1 git -c core.checkRoundtripEncoding=garbage \
 263                 add .gitattributes roundtrip.shift 2>&1 |
 264                 grep "Checking roundtrip encoding for SHIFT-JIS" &&
 265         git reset &&
 266
 267         # UTF-16 encoded files should not be round-trip checked by default...
 268         ! GIT_TRACE=1 git add roundtrip.utf16 2>&1 |
 269                 grep "Checking roundtrip encoding for UTF-16" &&
 270         git reset &&
 271
 272         # ... unless we tell Git to check it!
 273         GIT_TRACE=1 git -c core.checkRoundtripEncoding="UTF-16, UTF-32" \
 274                 add roundtrip.utf16 2>&1 |
 275                 grep "Checking roundtrip encoding for utf-16" &&
 276         git reset &&
 277
 278         # ... unless we tell Git to check it!
 279         # (here we also check that the casing of the encoding is irrelevant)
 280         GIT_TRACE=1 git -c core.checkRoundtripEncoding="UTF-32, utf-16" \
 281                 add roundtrip.utf16 2>&1 |
 282                 grep "Checking roundtrip encoding for utf-16" &&
 283         git reset
 284 '
 285
 286 # $1: checkout encoding
 287 # $2: test string
 288 # $3: binary test string in checkout encoding
 289 test_commit_utf8_checkout_other () {
 290         encoding="$1"
 291         orig_string="$2"
 292         expect_bytes="$3"
 293
 294         test_expect_success "Commit UTF-8, checkout $encoding" '
 295                 test_when_finished "git checkout HEAD -- .gitattributes" &&
 296
 297                 test_ext="commit_utf8_checkout_$encoding" &&
 298                 test_file="test.$test_ext" &&
 299
 300                 # Commit as UTF-8
 301                 echo "*.$test_ext text working-tree-encoding=UTF-8" >.gitattributes &&
 302                 printf "$orig_string" >$test_file &&
 303                 git add $test_file &&
 304                 git commit -m "Test data" &&
 305
 306                 # Checkout in tested encoding
 307                 rm $test_file &&
 308                 echo "*.$test_ext text working-tree-encoding=$encoding" >.gitattributes &&
 309                 git checkout HEAD -- $test_file &&
 310
 311                 # Test
 312                 printf $expect_bytes >$test_file.raw &&
 313                 test_cmp_bin $test_file.raw $test_file
 314         '
 315 }
 316
 317 test_commit_utf8_checkout_other "UTF-8"        "Test Тест" "\124\145\163\164\040\320\242\320\265\321\201\321\202"
 318 test_commit_utf8_checkout_other "UTF-16LE"     "Test Тест" "\124\000\145\000\163\000\164\000\040\000\042\004\065\004\101\004\102\004"
 319 test_commit_utf8_checkout_other "UTF-16BE"     "Test Тест" "\000\124\000\145\000\163\000\164\000\040\004\042\004\065\004\101\004\102"
 320 test_commit_utf8_checkout_other "UTF-16LE-BOM" "Test Тест" "\377\376\124\000\145\000\163\000\164\000\040\000\042\004\065\004\101\004\102\004"
 321 test_commit_utf8_checkout_other "UTF-16BE-BOM" "Test Тест" "\376\377\000\124\000\145\000\163\000\164\000\040\004\042\004\065\004\101\004\102"
 322 test_commit_utf8_checkout_other "UTF-32LE"     "Test Тест" "\124\000\000\000\145\000\000\000\163\000\000\000\164\000\000\000\040\000\000\000\042\004\000\000\065\004\000\000\101\004\000\000\102\004\000\000"
 323 test_commit_utf8_checkout_other "UTF-32BE"     "Test Тест" "\000\000\000\124\000\000\000\145\000\000\000\163\000\000\000\164\000\000\000\040\000\000\004\042\000\000\004\065\000\000\004\101\000\000\004\102"
 324
 325 test_done