[Octave-bug-tracker] [bug #65406] pkg test statistics fails include testing shadowed functions with core equivalents?

[Markus Mützel]

Follow-up Comment #3, bug #65406 (group octave):

Ah, I see. Thanks for the explanation.

It seems to work correctly when I add a `keyboard` command to one of the first
tests in that file, run the tests, and then do nothing but continue with
`dbcont`.
If I add `which mad` to the test (and remove the `keyboard` command), it shows
the path to the file in Octave core.

It calls the function from the statistics package with the following change:

diff -r fe201b17eb86 scripts/testfun/__run_test_suite__.m
--- a/scripts/testfun/__run_test_suite__.m      Fri Mar 01 18:16:19 2024 +0100
+++ b/scripts/testfun/__run_test_suite__.m      Sun Mar 03 16:11:40 2024 +0100
@@ -199,6 +199,7 @@
       [dnm, fnm] = fileparts (d);
       if (fnm(1) != "@")
         cd (d);
+        rehash ();
       endif
       for i = 1:length (lst)
         nm = lst(i).name;


It looks like the load path isn't properly rehashed on a directory change
(unless the Octave prompt is displayed?).

We probably shouldn't need to manually rehash the load path after a directory
change in a script or function. Or do we?
Was that rehashing removed on purpose (maybe for performance)? Or should the
load path be rehashed automatically each time the directory changes? (I don't
remember if that was discussed before.)




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?65406>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/