Hi Jacek,
2014-03-14 22:22 GMT+08:00 Jacek Caban jacek@codeweavers.com:
Sorry for the delay. I'm not sure that's the best strategy for having results that will help users in general. Having all functions implemented is a nice thing, but many of them don't cause problems in real world application and web pages. I think I would prefer testing random web pages and implementing things that they need. This, however, would be tricky to set goals required for GSoC application. Maybe you could come up with a mix of both? Start with selected area (say, IHTMLTable*, not necessarily whole) and move to random APIs required by pages as the project progresses?
I think work on both of them is a great idea, And I'm glad to do such job. As you said, it is not easy to set up goals. I have nearly 12 weeks, so my previous plan is to split them into several parts: IHTMLTableCell 3 weeks IHTMLTableCol & IHTMLTableRow 3 weeks IHTMLTableSection 2 weeks IHTMLTable 4 weeks (test-cases included)
I don't know if this meets the need of GSoC. Also, is it necessary to add "fix real work webpage/app problem" in this schedule?
Thank you very much.
Cheers, Jacek