Side comment: ± should probably be allowed as a unary operator. On Wed, Jun 22, 2016 at 3:39 PM, Kaj Wiik <[email protected]> wrote:
> To wrap this item up, it would be good to know where the distinction > (parsing) is done in sources? > > Thanks, > Kaj > > > On Wednesday, June 22, 2016 at 5:03:09 PM UTC+3, Steven G. Johnson wrote: >> >> >> >> On Wednesday, June 22, 2016 at 2:34:00 PM UTC+1, Kaj Wiik wrote: >>> >>> Yes, it is probably a binary operator but searches fail to find it. >>> >> >> Certain symbols are parsed as operators, and others are parsed as >> identifiers. This is independent of whether that symbol is defined in Base. >> >> For example α is parsed as an identifier (which is not defined in Base). >> ± and ⊕, on the other hand, are parsed as binary operators (which are also >> not defined in Base). >> >> For something parsed as a binary operator, if you want to give it a >> definition you should define it as a two-argument function, e.g. ⊕(x,y) = x >> .+ 2y .... then e.g. a⊕b will correspond to the function call ⊕(a,b). >> >
