On 1/2/07, <b class="gmail_sendername">aslak hellesoy</b> <<a href="mailto:aslak.hellesoy@gmail.com">aslak.hellesoy@gmail.com</a>> wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On 1/2/07, Nick Sieger <<a href="mailto:nicksieger@gmail.com">nicksieger@gmail.com</a>> wrote:<br>> Hi, I wrote an article [1] outlining some tricks I've developed adding<br>> custom extensions into RSpec. But they all hinge upon
<br>> Spec::Runner::Context#before_context_eval and the<br>> @context_eval_module ivar. So I was wondering whether this mechanism can be<br>> relied on in the future, or whether this is even the best way to wire in
<br>> such extensions. If you have a recommended way I'd love to hear about it so<br>> I can incorporate it into my article and make sure I don't highlight the<br>> wrong way of doing things.<br>><br>
<br>Regarding the stability of the API - I'm a little worried about having<br>extensions use private variables. </blockquote><div><br>Agreed. I was still motivated to post, but I suspected you would think that about the monkeypatching and the direct use of the ivar. I think I'll put a small disclaimer at the top of the post, and keep it up to date to follow whatever you guys think would be appropriate for an extension point API.
<br><br>/Nick<br></div><br></div>