t/t0028-working-tree-encoding.sh

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