[bug #68530] error([]) handling

--===============6944288425951604386==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: base64
Content-Disposition: inline

UGxlYXNlIHVzZSB0aGUgYnVnIHRyYWNrZXIgdG8gcG9zdCB1cGRhdGVzIHRvIGEgYnVnIHJlcG9y
dC4gIFRoZSBtYWlsaW5nIGxpc3QgaXMgaW50ZW5kZWQgYXMgYSByZWFkLW9ubHkgbm90aWZpY2F0
aW9uIHN0cmVhbS4gIEluZm8gcG9zdGVkIHRvIHRoaXMgbWFpbGluZyBsaXN0IGFkZHJlc3Mgd29u
J3QgYXBwZWFyIGluIHRoZSB0cmFja2VyIGRhdGFiYXNlIHdoZXJlIGl0IGlzIG1vc3QgdXNlZnVs
LgoK

--===============6944288425951604386==
Content-Type: MULTIPART/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; BOUNDARY="8323329-424238335-1784221358=:1792116"
Content-Disposition: inline

--8323329-424238335-1784221358=:1792116
Content-Type: TEXT/plain; CHARSET=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Content-Disposition: inline

Follow-up Comment #1, bug #68530 (group octave):=0A=0AThanks for the bug re=
port.=0A=0AYou mean that Matlab silently accepts "error ([])" to be no erro=
r and Octave=0Adoes the same but along the way first hickups when convertin=
g the empty=0Anumeric matrix to a string, while it should rather have detec=
ted beforehand=0Athat it was an empty value anyway and act more directly.=
=0A=0AYes, looks like a, well, cosmetic incompatibility.  The warning is ou=
t of=0Aplace and unneeded.=0A=0A=0A    ____________________________________=
___________________=0A=0AReply to this item at:=0A=0A  <https://savannah.gn=
u.org/bugs/?68530>=0A=0A_______________________________________________=0AM=
essage sent via Savannah=0Ahttps://savannah.gnu.org/=0A
--8323329-424238335-1784221358=:1792116
Content-Type: APPLICATION/pgp-signature; name=signature.asc

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQQk97aszIMMAvLLwm6qLAuaBUf3TgUCalkOrgAKCRCqLAuaBUf3
TjrSAP9MOu6Pi73sQ7TQkYQD2IXy3HlWnzqUv25Bx6lJdgq5nQD/bq3uWPB9LhCG
op3woVOcTdXSs11YJs23EBLzrIs5pQM=
=EByb
-----END PGP SIGNATURE-----

--8323329-424238335-1784221358=:1792116--



--===============6944288425951604386==--