fix: rewrite GNU testsuite harness to run upstream test scripts directly

[kevinburkesegment]

The previous harness tried to extract sed commands from GNU test scripts via regex pattern matching, which produced false negatives (comparing against empty expected output) and false positives. This led to inflated test counts and unreliable pass/fail signals.

The new approach:

• Provides a lightweight shim for the gnulib test framework (init.sh) with implementations of compare_, returns_, skip_, framework_failure_, and all require_* functions
• Executes each .sh test script from the GNU testsuite directly, injecting our Rust sed binary via PATH
• Uses a clean srcdir with symlinks to real test data files
• Adds 30s timeout per test to catch infinite loops
• Properly propagates exit codes (0=pass, 77=skip, 99=framework failure)

Results are now consistent with CI: 65 tests, ~12% pass rate, with clear PASS/FAIL/SKIP/timeout categorization.

[kevinburkesegment]

Here's the example output from a test run on my Macbook Pro

[INFO] Building Rust sed implementation...
[INFO] Using Rust sed binary: /Users/kburke/src/github.com/uutils/sed/target/release/sed
[INFO] Test working directory: /var/folders/mx/g324c_717ms4v6_1gglpyztm0000gn/T/tmp.pdnL3yZmCP
[INFO] Starting test execution...

[INFO] Running GNU testsuite shell script tests...
[FAIL] 8bit
[FAIL] 8to7
[FAIL] badenc
[FAIL] binary (timeout)
[PASS] bsd
[FAIL] bsd-wrapper
[SKIP] bug32082
[PASS] bug32271-1
[SKIP] bug32271-2
[FAIL] cmd-0r
[FAIL] cmd-l
[FAIL] cmd-R
[FAIL] colon-with-no-label
[FAIL] command-endings
[FAIL] comment-n
[FAIL] compile-errors
[FAIL] compile-tests
[FAIL] convert-number
[FAIL] dc (timeout)
[FAIL] distrib
[FAIL] eval
[FAIL] execute-tests
[FAIL] follow-symlinks (framework failure)
[FAIL] follow-symlinks-stdin
[FAIL] help
[FAIL] inplace-hold
[FAIL] in-place-hyphen
[SKIP] inplace-selinux
[FAIL] in-place-suffix-backup
[SKIP] invalid-mb-seq-UMR
[FAIL] mac-mf
[PASS] madding
[FAIL] mb-bad-delim
[SKIP] mb-charclass-non-utf8
[FAIL] mb-match-slash
[FAIL] mb-y-translate
[FAIL] missing-filename
[PASS] newjis
[SKIP] newline-dfa-bug
[FAIL] normalize-text
[FAIL] nulldata
[SKIP] obinary
[FAIL] posix-char-class
[FAIL] posix-mode-addr
[FAIL] posix-mode-bad-ref
[FAIL] posix-mode-ERE
[FAIL] posix-mode-N
[FAIL] posix-mode-s
[PASS] range-overlap
[FAIL] recursive-escape-c
[FAIL] regex-errors
[SKIP] regex-max-int
[FAIL] sandbox
[FAIL] stdin-prog
[PASS] stdin
[FAIL] subst-mb-incomplete
[FAIL] subst-options
[FAIL] subst-replacement
[FAIL] temp-file-cleanup
[SKIP] title-case
[FAIL] unbuffered
[PASS] uniq
[FAIL] utf8-ru
[FAIL] word-delim
[FAIL] xemacs
[INFO] Ran 65 shell script tests

=========================================
TEST RESULTS SUMMARY
=========================================
Total tests:   65
Passed:        7
Failed:        49
Skipped:       9
Duration:      87s
Pass rate:     12%
Result:        SOME TESTS FAILED

[github-actions]

GNU sed testsuite comparison:

Test results comparison:
  Current:   TOTAL: 0 / PASSED: 0 / FAILED: 0 / SKIPPED: 0
  Reference: TOTAL: 60 / PASSED: 17 / FAILED: 43 / SKIPPED: 0

Changes from main branch:
  TOTAL: -60
  PASSED: -17
  FAILED: -43

Test improvements (43):
  + 8to7_dash_e_0
  + bsd_echo_0
  + bsd_echo_1
  + bug32082_dash_e_0
  + bug32271-1_dash_e_0
  + bug32271-2_dash_e_0
  + cmd-0r_cmd_0
  + cmd-0r_file_1
  + cmd-0r_file_2
  + follow-symlinks_cmd_0
  + follow-symlinks_file_1
  + follow-symlinks_n_2
  + help_cmd_0
  + invalid-mb-seq-UMR_subst_0
  + mac-mf_triplet
  + missing-filename_subst_0
  + newline-dfa-bug_cmd_0
  + newline-dfa-bug_file_1
  + normalize-text_cmd_0
  + nulldata_subst_0
  + posix-char-class_subst_0
  + posix-mode-ERE_subst_0
  + posix-mode-ERE_subst_1
  + posix-mode-addr_n_0
  + posix-mode-addr_n_1
  + posix-mode-bad-ref_subst_0
  + posix-mode-s_file_2
  + posix-mode-s_file_3
  + posix-mode-s_subst_0
  + posix-mode-s_subst_1
  + recursive-escape-c_cmd_0
  + regex-errors_cmd_0
  + sandbox_cmd_0
  + sandbox_file_1
  + sandbox_file_2
  + sandbox_file_3
  + subst-options_cmd_0
  + subst-options_file_1
  + subst-options_file_2
  + subst-options_file_3
  + unbuffered_subst_0
  + utf8-ru_subst_0
  + word-delim_subst_0

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 82.07%. Comparing base (25efbc0) to head (fbf8f5f).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #343   +/-   ##
=======================================
  Coverage   82.07%   82.07%           
=======================================
  Files          13       13           
  Lines        5445     5445           
  Branches      293      293           
=======================================
  Hits         4469     4469           
  Misses        974      974           
  Partials        2        2           

Flag	Coverage Δ
macos_latest	82.51% <ø> (ø)
ubuntu_latest	82.63% <ø> (ø)
windows_latest	0.00% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[sylvestre]

small problem here :)

[kevinburkesegment]

Sorry - give it another try

[sylvestre]

it is fine, don't be sorry :)

[kevinburkesegment]

one more try please!

[sylvestre]

done!