sed tests

sed tests

[Skip Collins]

Recently (last two or three weeks), when I run make test on the latest
git source (currently release_9.1.6-395-g8ecc4c) on my mac with Emacs
26.0.90 (compiled with brew on my Mac), I get the following results:

Ran 775 tests, 771 results as expected, 4 unexpected (2018-01-23 11:37:59-0500)
11 expected failures

4 unexpected results:
   FAILED  ob-sed-test/cmd-line-header-argument
   FAILED  ob-sed-test/in-file-header-argument
   FAILED  ob-sed-test/simple-execution-of-script
   FAILED  test-ob-python/session-multiline

I would appreciate any suggestions for how to address the problem.
This is not urgent, as I don't use ob-sed and org seems to be
functioning normally otherwise. But OSX command line tools sometimes
have quirks that show up as failed tests. Of course it is possible
that I broke my emacs/org in some way.

The backtrace from the first failed test is:
executing Sed code block...
executing sed source code block
Babel evaluation exited with code 1
Code block produced no output.
Test ob-sed-test/cmd-line-header-argument backtrace:
  signal(ert-test-failed (((should (string= "A tested again file.\n" (
  ert-fail(((should (string= "A tested again file.\n" (with-temp-buffe
  (if (unwind-protect (setq value-1104 (apply fn-1102 args-1103)) (set
  (let (form-description-1106) (if (unwind-protect (setq value-1104 (a
  (let ((value-1104 'ert-form-evaluation-aborted-1105)) (let (form-des
  (let* ((fn-1102 (function string=)) (args-1103 (condition-case err (
  (let ((default-directory temporary-file-directory)) (let ((temp-buff
  (save-restriction (let ((default-directory temporary-file-directory)
  (progn (org-id-goto "E3C6A8BA-39FF-4840-BA8E-90D5C4365AB1") (setq to
  (unwind-protect (progn (org-id-goto "E3C6A8BA-39FF-4840-BA8E-90D5C43
  (let ((save-match-data-internal (match-data))) (unwind-protect (prog
  (progn (let ((save-match-data-internal (match-data))) (unwind-protec
  (unwind-protect (progn (let ((save-match-data-internal (match-data))
  (let ((wconfig (current-window-configuration))) (unwind-protect (pro
  (unwind-protect (let ((wconfig (current-window-configuration))) (unw
  (let* ((id-location (org-id-find "E3C6A8BA-39FF-4840-BA8E-90D5C4365A
  (lambda nil (let* ((id-location (org-id-find "E3C6A8BA-39FF-4840-BA8
  ert--run-test-internal(#s(ert--test-execution-info :test #s(ert-test
  ert-run-test(#s(ert-test :name ob-sed-test/cmd-line-header-argument
  ert-run-or-rerun-test(#s(ert--stats :selector "\\(org\\|ob\\)" :test
  ert-run-tests("\\(org\\|ob\\)" #f(compiled-function (event-type &res
  ert-run-tests-batch("\\(org\\|ob\\)")
  ert-run-tests-batch-and-exit("\\(org\\|ob\\)")
  (let ((org-id-track-globally t) (org-test-selector (if org-test-sele
  org-test-run-batch-tests("\\(org\\|ob\\)")
  eval((org-test-run-batch-tests org-test-select-re))
  command-line-1(("--eval" "(setq vc-handled-backends nil org-startup-
  command-line()
  normal-top-level()
Test ob-sed-test/cmd-line-header-argument condition:
    (ert-test-failed
     ((should
       (string= "A tested again file.
"
(with-temp-buffer ... ...)))
      :form
      (string= "A tested again file.
" "A test file.
")
      :value nil))
   FAILED  109/775  ob-sed-test/cmd-line-header-argument

Re: sed tests

[Marco Wahl]

Hi Skip,

Collins <skip.collins@gmail.com> writes:

> Recently (last two or three weeks), when I run make test on the latest
> git source (currently release_9.1.6-395-g8ecc4c) on my mac with Emacs
> 26.0.90 (compiled with brew on my Mac), I get the following results:
>
> Ran 775 tests, 771 results as expected, 4 unexpected (2018-01-23 11:37:59-0500)
> 11 expected failures
>
> 4 unexpected results:
>    FAILED  ob-sed-test/cmd-line-header-argument
>    FAILED  ob-sed-test/in-file-header-argument
>    FAILED  ob-sed-test/simple-execution-of-script
>    FAILED  test-ob-python/session-multiline
>
> I would appreciate any suggestions for how to address the problem.
> This is not urgent, as I don't use ob-sed and org seems to be
> functioning normally otherwise. But OSX command line tools sometimes
> have quirks that show up as failed tests. Of course it is possible
> that I broke my emacs/org in some way.

Some days ago I accidentially saw that the sed tests were disabled and
that I could easily activate them.  And I did and thought that was a
wonderful idea.  And also the sed tests pass on my linux box.

This

> Test ob-sed-test/cmd-line-header-argument condition:
>     (ert-test-failed
>      ((should
>        (string= "A tested again file.
> "
> (with-temp-buffer ... ...)))
>       :form
>       (string= "A tested again file.
> "              "A test file.
> ")
>       :value nil))

shows that sed did not do the substitution of "test" by "tested again"
as defined in the respective sed babel block.

Maybe you could try to execute sed babel blocks directly.  Possibly you
get a better understanding of the behaviour.


HTH
   Marco