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