More comments interleaved below. That would be great if you could handle the explicit method calls in place of the automatic calls during Element subclassing. I'll take a look at everything else and clean it up except for the places below where we need more information or need to discuss. I think I'll also generate a similar list for the major input elements to see how we fare. I suspect we have some more straggler methods that could use cleaning up.
Hugh > assert_exists - should be private in watir-webdriver i think >element(s)_by_xpath - these methods should be private in watir-webdriver too as they are now for watir 3 I'll check on that - I generated the list using the public_methods method so maybe that wasn't close enough. Will verify manually... > browser - what is the point of that instance method? that's what I was wondering too. Jari? > cell/cells - if i'm not mistaken then these methods were available for Table in webdriver. There was some logic behind that. Same goes with row/rows. I'll see how webdriver implements it but maybe we should talk about it with Jari. Most other elements we can access directly and it seems useful to me to look for a cell without having to care about the table. It might be that you can do the same thing in which case we can just do what he's doing :) > field_set/field_sets - these methods came with watir 3 and are just there for aliases. No need to deprecate these i think, but it would be nice if watir webdriver would make these aliases too. We should talk to Jari, decide on what to do, and do the same thing in both drivers. I don't have a preference on either, just want to be consistent. >input_element(s) - aren't these methods being generated after Element is subclassed by InputElement? If that's the case then i would delete these methods and not make them private. Same goes to table_cell(s), table_row(s) and table_section(s). >I have been planning to revise that part too and make all methods creation more explicit. So if you don't have anything against that then i'd take that task for myself. Not sure when i will deal with it though exactly. Yes, that's my understanding too. I used 'make private' a little loosely but meant it shouldn't be exposed - so your fix to make the methods explicit will be great. Go for it :) >It seems to me that before we release 3.0 we need to make the watirspec tests that fail for implemented features pass (fix things like tables and xpath that we've implemented, don't worry about unimplemented features) > I'm not 100% sure that this should be a release blocker since all these xpath things are failing with < 3 version too. I'll take a look to see how hard this is - it would be nice to get working but I agree it's not an interface change and our tests on this do look OK for the most part.
_______________________________________________ Wtr-development mailing list Wtr-development@rubyforge.org http://rubyforge.org/mailman/listinfo/wtr-development
