git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH v2 4/5] tests: drop here-doc check from internal chain-linter

From
Jeff King <peff@peff.net>
Date
Mar 30, 2023, 19:30 UTC
Message-ID
<20230330193047.GD27989@coredump.intra.peff.net>
In-Reply-To
<20230330192712.GA27719@coredump.intra.peff.net>

Commit 99a64e4b73c (tests: lint for run-away here-doc, 2017-03-22) tweaked the chain-lint test to catch unclosed here-docs. It works by adding an extra "echo" command after the test snippet, and checking that it is run (if it gets swallowed by a here-doc, naturally it is not run).

The downside here is that we introduced an extra $() substitution, which happens in a subshell. This has a measurable performance impact when run for many tests.

The tradeoff in safety was undoubtedly worth it when 99a64e4b73c was written. But since the external chainlint.pl learned to find these recently, we can just rely on it. By switching back to a simpler chain-lint, hyperfine reports a measurable speedup on t3070 (which has 1800 tests):

  'HEAD' ran
    1.12 ± 0.01 times faster than 'HEAD~1'
Signed-off-by: Jeff King <peff@peff.net>
---
Now with 90% less hand-waving.
 t/test-lib.sh | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/t/test-lib.sh b/t/test-lib.sh
index cfcbd899c5a..0048ec7b6f6 100644
--- a/t/test-lib.sh
+++ b/t/test-lib.sh
@@ -1101,9 +1101,10 @@ test_run_ () {
 		trace=
 		# 117 is magic because it is unlikely to match the exit
 		# code of other programs
-		if test "OK-117" != "$(test_eval_ "fail_117 && $1${LF}${LF}echo OK-\$?" 3>&1)"
+		test_eval_ "fail_117 && $1"
+		if test $? != 117
 		then
-			BUG "broken &&-chain or run-away HERE-DOC: $1"
+			BUG "broken &&-chain: $1"
 		fi
 		trace=$trace_tmp
 	fi
-- 
2.40.0.692.g7c4c956fc5c
Previous: Eric SunshineNext: Jeff King
Message 29 of 32 in “some chainlint fixes and performance improvements”
  1. 0/4 some chainlint fixes and performance improvementsJeff King, Mar 28, 2023
  2. 1/4 tests: run internal chain-linter under "make test"Jeff King, Mar 28, 2023
  3. Ævar Arnfjörð BjarmasonMar 29, 2023
  4. Junio C HamanoMar 29, 2023
  5. Jeff KingMar 29, 2023
  6. Junio C HamanoMar 30, 2023
  7. 2/4 tests: replace chainlint subshell with a functionJeff King, Mar 28, 2023
  8. Junio C HamanoMar 28, 2023
  9. 3/4 tests: drop here-doc check from internal chain-linterJeff King, Mar 28, 2023
  10. Junio C HamanoMar 28, 2023
  11. Jeff KingMar 29, 2023
  12. Jeff KingMar 29, 2023
  13. Eric SunshineMar 29, 2023
  14. Eric SunshineMar 29, 2023
  15. Eric SunshineMar 29, 2023
  16. Jeff KingMar 29, 2023
  17. Eric SunshineMar 29, 2023
  18. Eric SunshineMar 29, 2023
  19. Jeff KingMar 29, 2023
  20. 4/4 tests: skip test_eval_ in internal chain-lintJeff King, Mar 28, 2023
  21. Jeff KingMar 28, 2023
  22. Jeff KingMar 30, 2023
  23. Junio C HamanoMar 30, 2023
  24. 0/5 some chainlint fixes and performance improvementsJeff King, Mar 30, 2023
  25. 1/5 tests: run internal chain-linter under "make test"Jeff King, Mar 30, 2023
  26. 2/5 tests: replace chainlint subshell with a functionJeff King, Mar 30, 2023
  27. 3/5 tests: diagnose unclosed here-doc in chainlint.plJeff King, Mar 30, 2023
  28. Eric SunshineMar 30, 2023
  29. 4/5 tests: drop here-doc check from internal chain-linterJeff King, Mar 30, 2023
  30. 5/5 tests: skip test_eval_ in internal chain-lintJeff King, Mar 30, 2023
  31. Junio C HamanoMar 30, 2023
  32. Jeff KingMar 30, 2023

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.