[Octave-bug-tracker] [bug #60742] test failure in hgsave.m with Octave 6.2.90 (on Windows)

[Hartmut]

Follow-up Comment #3, bug #60742 (project octave):

Oops, I've only now realised that most of my initial comment was cut off by
Savannah. Here are some answers:

* reproducible? Yes, "test hgsave" gives me a reproducible failure on my Win10
laptop. This means I can run this test 5 times in a row, and they all fail. A
I have not ever seen a passing test with Octave 6.2.90 on this pc so far.

* older versions? I have virtually the last 10 Octave versions on the same pc
installed. Here is how they behave on "test hgsave":
** 6.2.0: FAILs(3 times -> reproducibly)
** 6.1.0: produces the debug message "DEBUG: FC_WEIGHT didn't match". But
PASSes. (3 times -> reproducible)
** 5.2.0: same debug message as in 6.1.0, also PASSes.
** 5.1.0: same debug message as in 6.1.0, also PASSes.
** 4.4.1: PASSes (no debug messages), runs only 4 tests (later versions to 6
tests)
** 4.2.2: hundreds of warnings "warning: could not match any font:
*-normal-normal-14". PASSes all 4 tests

** I use a dual monitor setup, a small laptop monitor plus an additional large
external monitor. This gives a funny (i.e. non-rectangular) screen size. I am
not sure if this is relevant.

Can I do something helpful to figure out if this failing test is a real
problem or only a curiosity?

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?60742>

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