Re: Return value docs of empty()

On 9 January 2018 13:21:59 GMT+00:00, Peter Cowburn <petercowburn@gmail.com> wrote: >On 9 January 2018 at 12:55, Christoph M. Becker <cmbecker69@gmx.de> >wrote: > >> Hi! >> >> I would suggest to apply the following patch: >> >> Index: empty.xml >> =================================================================== >> --- empty.xml (revision 343793) >> +++ empty.xml (working copy) >> @@ -50,7 +50,7 @@ >> Otherwise returns &true;. >> </para> >> <para> >> - The following things are considered to be empty: >> + The following values are considered to be empty: >> <simplelist> >> <member><literal>""</literal> (an empty >> string)</member> >> <member><literal>0</literal> (0 as an integer)</member> >> @@ -59,7 +59,6 @@ >> <member>&null;</member> >> <member>&false;</member> >> <member><literal>array()</literal> (an empty array)</member> >> - <member><literal>$var;</literal> (a variable declared, but >without >> a value)</member> >> </simplelist> >> </para> >> </refsect1> >> >> Any objections? >> > >Not from me. The variable-without-a-value is covered by the null case, >and >the fact that a variable-without-a-value is null is covered elsewhere. I don't know if there are other notes on the page that make it redundant, but the handling of undefined variables is one of the most important things about empty() - as a language construct, it has a unique ability to accept them without generating a notice. Regards, Rowan -- Rowan Collins [IMSoP]