"Page, Bill" <[EMAIL PROTECTED]> writes: | On Monday, March 27, 2006 9:57 PM Gaby wrote: | > ... | > | | > | As discussed in section "5.4.8 for Loops" of the Axiom book, you | > | may follow the keyword 'in' only by an integer segment, stream or | > | a list, but try this: | > | | > | V:=vector [1,2,3] | > | [i for i in entries V] | > | | > | The function 'entries' returns the elements of a vector as a | > | List. | > | > That should be seen as a "defect". for...in should be made to | > work with sequence (whether built-in list or vectors). | > | | What is a "sequence"?
Sequence is my general term for list, vector or any "thingy" walked through by a generator. | As far as I know this is not defined in Axiom. yes, I know. That is precisely why I said it "should be" seen as a defect. | Do mean that some types should be automatically coercible | to type List? No. I'm saying that the "iterator constructor" should be made to traverse vectors too, not just list; and no I don't consider convolutions like "entries V" acceptable :-). | Note that: | | V:=vector [1,2,3] | [i for i in V::List Integer] | | works in the Axiom interpreter but it is not clear to me exactly | where this conversion is defined. coercion to List should not be needed, nor required. Hey, have a look at what you have around in Java, C#, Perl, etc. (admittely, Perl is convoluted). -- Gaby _______________________________________________ Axiom-developer mailing list Axiom-developer@nongnu.org http://lists.nongnu.org/mailman/listinfo/axiom-developer
