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

[PATCH 11/25] t: wrap complicated expect_code users in a block

From
Jeff King <peff@peff.net>
Date
Mar 20, 2015, 10:12 UTC
Message-ID
<20150320101229.GK12543@peff.net>
In-Reply-To
<20150320100429.GA17354@peff.net>

If we are expecting a command to produce a particular exit code, we can use test_expect_code. However, some cases are more complicated, and want to accept one of a range of exit codes. For these, we end up with something like:

  cmd;
  case "$?" in
  ...

That unfortunately breaks the &&-chain and fools --chain-lint. Since these special cases are so few, we can wrap them in a block, like this:

  { cmd; ret=$?; } &&
  case "$ret" in
  ...

This accomplishes the same thing, and retains the &&-chain (the exit status fed to the && is that of the assignment, which should always be true). It's technically longer, but it is probably a good thing for unusual code like this to stand out.

Signed-off-by: Jeff King <peff@peff.net>
---
 t/t0005-signals.sh              | 4 ++--
 t/t4026-color.sh                | 6 +++---
 t/t5004-archive-corner-cases.sh | 6 ++++--
 t/t5512-ls-remote.sh            | 6 ++++--
 4 files changed, 13 insertions(+), 9 deletions(-)
diff --git a/t/t0005-signals.sh b/t/t0005-signals.sh
index aeea50c..5c5707d 100755
--- a/t/t0005-signals.sh
+++ b/t/t0005-signals.sh
@@ -10,8 +10,8 @@ one
 EOF
 
 test_expect_success 'sigchain works' '
-	test-sigchain >actual
-	case "$?" in
+	{ test-sigchain >actual; ret=$?; } &&
+	case "$ret" in
 	143) true ;; # POSIX w/ SIGTERM=15
 	271) true ;; # ksh w/ SIGTERM=15
 	  3) true ;; # Windows
diff --git a/t/t4026-color.sh b/t/t4026-color.sh
index 4d20fea..2b32c4f 100755
--- a/t/t4026-color.sh
+++ b/t/t4026-color.sh
@@ -111,9 +111,9 @@ test_expect_success 'unknown color slots are ignored (branch)' '
 '
 
 test_expect_success 'unknown color slots are ignored (status)' '
-	git config color.status.nosuchslotwilleverbedefined white || exit
-	git status
-	case $? in 0|1) : ok ;; *) false ;; esac
+	git config color.status.nosuchslotwilleverbedefined white &&
+	{ git status; ret=$?; } &&
+	case $ret in 0|1) : ok ;; *) false ;; esac
 '
 
 test_done
diff --git a/t/t5004-archive-corner-cases.sh b/t/t5004-archive-corner-cases.sh
index 305bcac..654adda 100755
--- a/t/t5004-archive-corner-cases.sh
+++ b/t/t5004-archive-corner-cases.sh
@@ -66,8 +66,10 @@ test_expect_success UNZIP 'zip archive of empty tree is empty' '
 	# handle the empty repo at all, making our later check of its exit code
 	# a no-op). But we cannot do anything reasonable except skip the test
 	# on such platforms anyway, and this is the moral equivalent.
-	"$GIT_UNZIP" "$TEST_DIRECTORY"/t5004/empty.zip
-	expect_code=$?
+	{
+		"$GIT_UNZIP" "$TEST_DIRECTORY"/t5004/empty.zip
+		expect_code=$?
+	} &&
 
 	git archive --format=zip HEAD >empty.zip &&
 	make_dir extract &&
diff --git a/t/t5512-ls-remote.sh b/t/t5512-ls-remote.sh
index 321c3e5..3bd9759 100755
--- a/t/t5512-ls-remote.sh
+++ b/t/t5512-ls-remote.sh
@@ -103,8 +103,10 @@ test_expect_success 'confuses pattern as remote when no remote specified' '
 '
 
 test_expect_success 'die with non-2 for wrong repository even with --exit-code' '
-	git ls-remote --exit-code ./no-such-repository ;# not &&
-	status=$? &&
+	{
+		git ls-remote --exit-code ./no-such-repository
+		status=$?
+	} &&
 	test $status != 2 && test $status != 0
 '
 

-- 
2.3.3.520.g3cfbb5d
Previous: Jeff KingNext: Jeff King
Message 14 of 68 in “detecting &&-chain breakage”
  1. 0/25 detecting &&-chain breakageJeff King, Mar 20, 2015
  2. 01/25 t/test-lib: introduce --chain-lint optionJeff King, Mar 20, 2015
  3. SZEDER GáborMar 25, 2015
  4. Jeff KingMar 25, 2015
  5. 02/25 t: fix severe &&-chain breakageJeff King, Mar 20, 2015
  6. 03/25 t: fix moderate &&-chain breakageJeff King, Mar 20, 2015
  7. 04/25 t: fix trivial &&-chain breakageJeff King, Mar 20, 2015
  8. 05/25 t: assume test_cmp produces verbose outputJeff King, Mar 20, 2015
  9. 06/25 t: use verbose instead of hand-rolled errorsJeff King, Mar 20, 2015
  10. 07/25 t: use test_must_fail instead of hand-rolled blocksJeff King, Mar 20, 2015
  11. 08/25 t: fix &&-chaining issues around setup which might failJeff King, Mar 20, 2015
  12. 09/25 t: use test_might_fail for diff and grepJeff King, Mar 20, 2015
  13. 10/25 t: use test_expect_code instead of hand-rolled comparisonJeff King, Mar 20, 2015
  14. 11/25 t: wrap complicated expect_code users in a blockJeff King, Mar 20, 2015
  15. 12/25 t: avoid using ":" for commentsJeff King, Mar 20, 2015
  16. 13/25 t3600: fix &&-chain breakage for setup commandsJeff King, Mar 20, 2015
  17. 14/25 t7201: fix &&-chain breakageJeff King, Mar 20, 2015
  18. 15/25 t9502: fix &&-chain breakageJeff King, Mar 20, 2015
  19. Johannes SixtMar 20, 2015
  20. Jeff KingMar 20, 2015
  21. 16/25 t6030: use modern test_* helpersJeff King, Mar 20, 2015
  22. 17/25 t0020: use modern test_* helpersJeff King, Mar 20, 2015
  23. SZEDER GáborMar 25, 2015
  24. Jeff KingMar 25, 2015
  25. 0/8 more &&-chaining test fixupsJeff King, Mar 25, 2015
  26. 1/8 perf-lib: fix ignored exit code inside loopJeff King, Mar 25, 2015
  27. 2/8 t0020: fix ignored exit code inside loopsJeff King, Mar 25, 2015
  28. 3/8 t3305: fix ignored exit code inside loopJeff King, Mar 25, 2015
  29. Johan HerlandMar 25, 2015
  30. 4/8 t7701: fix ignored exit code inside loopJeff King, Mar 25, 2015
  31. 5/8 t: fix some trivial cases of ignored exit codes in loopsJeff King, Mar 25, 2015
  32. 6/8 t: simplify loop exit-code status variablesJeff King, Mar 25, 2015
  33. Junio C HamanoMar 25, 2015
  34. Jeff KingMar 25, 2015
  35. 7/8 t0020: use test_* helpers instead of hand-rolled messagesJeff King, Mar 25, 2015
  36. 8/8 t9001: drop save_confirm helperJeff King, Mar 25, 2015
  37. Junio C HamanoMar 25, 2015
  38. 18/25 t1301: use modern test_* helpersJeff King, Mar 20, 2015
  39. SZEDER GáborMar 24, 2015
  40. Jeff KingMar 25, 2015
  41. 19/25 t6034: use modern test_* helpersJeff King, Mar 20, 2015
  42. SZEDER GáborMar 24, 2015
  43. 20/25 t4117: use modern test_* helpersJeff King, Mar 20, 2015
  44. 21/25 t9001: use test_when_finishedJeff King, Mar 20, 2015
  45. SZEDER GáborMar 25, 2015
  46. Jeff KingMar 25, 2015
  47. 22/25 t0050: appease --chain-lintJeff King, Mar 20, 2015
  48. 23/25 t7004: fix embedded single-quotesJeff King, Mar 20, 2015
  49. 24/25 t0005: fix broken &&-chainsJeff King, Mar 20, 2015
  50. 25/25 t4104: drop hand-rolled error reportingJeff King, Mar 20, 2015
  51. Jeff KingMar 20, 2015
  52. Michael J GruberMar 20, 2015
  53. 26/27 t/*svn*: fix moderate &&-chain breakageMichael J Gruber, Mar 20, 2015
  54. Junio C HamanoMar 20, 2015
  55. Junio C HamanoMar 20, 2015
  56. Jeff KingMar 20, 2015
  57. Jeff KingMar 20, 2015
  58. Michael J GruberMar 23, 2015
  59. 27/27 t9104: fix test for following larger parentsMichael J Gruber, Mar 20, 2015
  60. Jeff KingMar 20, 2015
  61. Junio C HamanoMar 20, 2015
  62. Junio C HamanoMar 20, 2015
  63. Jeff KingMar 20, 2015
  64. Junio C HamanoMar 20, 2015
  65. Eric SunshineMar 20, 2015
  66. Jeff KingMar 21, 2015
  67. Junio C HamanoMar 21, 2015
  68. Jeff KingMar 21, 2015

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.