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

[PATCH v2] t/Makefile: add a rule to re-run previously-failed tests

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Aug 29, 2016, 13:46 UTC
Message-ID
<0dfa96b17edfe84ba19c7e57fe0b017c77943e0c.1472478285.git.johannes.schindelin@gmx.de>
In-Reply-To
<b2d016e44fa04e8a318967c43762d6933faf7956.1467183740.git.johannes.schindelin@gmx.de>

While developing patch series, it is a good practice to run the test suite from time to time, just to make sure that obvious bugs are caught early. With complex patch series, it is common to run `make -j15 -k test`, i.e. run the tests in parallel and *not* stop at the first failing test but continue. This has the advantage of identifying possibly multiple problems in one big test run.

It is particularly important to reduce the turn-around time thusly on Windows, where the test suite spends 45 minutes on the computer on which this patch was developed.

It is the most convenient way to determine which tests failed after running the entire test suite, in parallel, to look for left-over "trash directory.t*" subdirectories in the t/ subdirectory. However, as was pointed out by Jeff King, those directories might live outside t/ when overridden using the --root=<directory> option, to which the Makefile has no access. The next best method is to grep explicitly for failed tests in the test-results/ directory, which the Makefile *can* access.

This patch automates the process of determinig which tests failed previously and re-running them.

Note that we need to be careful to inspect only the *newest* entries in test-results/: this directory contains files of the form tNNNN-<name>-<pid>.counts and is only removed wholesale when running the *entire* test suite, not when running individual tests. We ensure that with a little sed magic on `ls -t`'s output that simply skips lines when the file name was seen earlier.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
	The patch is unfortunately no longer as trivial as before, but it
	now also works with --root=..., i.e. when the user overrode the
	location of the trash directories.
Published-As: https://github.com/dscho/git/releases/tag/failing-tests-v2
Fetch-It-Via: git fetch https://github.com/dscho/git failing-tests-v2
Interdiff vs v1:
 diff --git a/t/Makefile b/t/Makefile
 index c402a9ec..8aa6a72 100644
 --- a/t/Makefile
 +++ b/t/Makefile
 @@ -35,7 +35,12 @@ all: $(DEFAULT_TEST_TARGET)
  test: pre-clean $(TEST_LINT)
  	$(MAKE) aggregate-results-and-cleanup
  
 -failed: $(patsubst trash,,$(patsubst directory.%,%.sh,$(wildcard trash\ directory.t[0-9]*)))
 +failed:
 +	@failed=$$(cd '$(TEST_RESULTS_DIRECTORY_SQ)' && \
 +		grep -l '^failed [1-9]' $$(ls -t *.counts | \
 +			sed 'G;h;/^\(t[^.]*\)-[0-9]*\..*\n\1-[0-9]*\./d;P;d') | \
 +		sed -n 's/-[0-9]*\.counts$$/.sh/p') && \
 +	test -z "$$failed" || $(MAKE) $$failed
  
  prove: pre-clean $(TEST_LINT)
  	@echo "*** prove ***"; $(PROVE) --exec '$(SHELL_PATH_SQ)' $(GIT_PROVE_OPTS) $(T) :: $(GIT_TEST_OPTS)
 t/Makefile | 7 +++++++
 1 file changed, 7 insertions(+)
diff --git a/t/Makefile b/t/Makefile
index d613935..8aa6a72 100644
--- a/t/Makefile
+++ b/t/Makefile
@@ -35,6 +35,13 @@ all: $(DEFAULT_TEST_TARGET)
 test: pre-clean $(TEST_LINT)
 	$(MAKE) aggregate-results-and-cleanup
 
+failed:
+	@failed=$$(cd '$(TEST_RESULTS_DIRECTORY_SQ)' && \
+		grep -l '^failed [1-9]' $$(ls -t *.counts | \
+			sed 'G;h;/^\(t[^.]*\)-[0-9]*\..*\n\1-[0-9]*\./d;P;d') | \
+		sed -n 's/-[0-9]*\.counts$$/.sh/p') && \
+	test -z "$$failed" || $(MAKE) $$failed
+
 prove: pre-clean $(TEST_LINT)
 	@echo "*** prove ***"; $(PROVE) --exec '$(SHELL_PATH_SQ)' $(GIT_PROVE_OPTS) $(T) :: $(GIT_TEST_OPTS)
 	$(MAKE) clean-except-prove-cache
-- 
2.10.0.rc1.114.g2bd6b38

base-commit: d5cb9cbd64165153a318e1049f8bf14b09a16b11
Previous: Johannes SchindelinNext: Jeff King
Message 6 of 32 in “t/Makefile: add a rule to re-run previously-failed tests”
  1. t/Makefile: add a rule to re-run previously-failed testsJohannes Schindelin, Jun 29, 2016
  2. Junio C HamanoJun 29, 2016
  3. Johannes SchindelinJul 1, 2016
  4. Jeff KingJun 30, 2016
  5. Johannes SchindelinJul 1, 2016
  6. t/Makefile: add a rule to re-run previously-failed testsJohannes Schindelin, Aug 29, 2016
  7. Jeff KingAug 30, 2016
  8. Junio C HamanoAug 30, 2016
  9. Johannes SchindelinAug 31, 2016
  10. Sverre RabbelierSep 1, 2016
  11. Johannes SchindelinSep 1, 2016
  12. Sverre RabbelierSep 1, 2016
  13. Junio C HamanoSep 1, 2016
  14. Johannes SchindelinSep 2, 2016
  15. Junio C HamanoSep 8, 2016
  16. Ævar Arnfjörð BjarmasonAug 30, 2016
  17. Jeff KingAug 30, 2016
  18. Ævar Arnfjörð BjarmasonAug 30, 2016
  19. Johannes SchindelinAug 31, 2016
  20. Ævar Arnfjörð BjarmasonAug 31, 2016
  21. Johannes SchindelinAug 31, 2016
  22. Ævar Arnfjörð BjarmasonSep 2, 2016
  23. Johannes SchindelinSep 2, 2016
  24. Matthieu MoySep 2, 2016
  25. Johannes SchindelinSep 4, 2016
  26. Matthieu MoySep 4, 2016
  27. t/Makefile: add a rule to re-run previously-failed testsJohannes Schindelin, Jan 27, 2017
  28. Jeff KingJan 27, 2017
  29. Johannes SchindelinJan 27, 2017
  30. t/Makefile: add a rule to re-run previously-failed testsJohannes Schindelin, Jan 27, 2017
  31. Junio C HamanoJan 27, 2017
  32. Johannes SchindelinJan 30, 2017

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.