Re: message_box() [Re: Yay or Nay? Framed reference manual]
- Posted by CChris <christian.cuvier at agricu?ture.g?uv.fr> May 07, 2008
- 622 views
Shawn Pringle wrote: > > Jeremy Cowgar wrote: > > > > > Greg, I am no HTML/CSS guru. Can you make a div layout w/o fancy Javascript > > where the navigation div stays on the side, no matter how high or low you > > scroll > > in the docs? For instance, you can click on Language Manual, scroll down a > > substantial > > amount and with the frames, the navigation frame will still be right there > > at > > your finger tips. > > > > If there are other ways of doing this, I'm open for suggestions, but I think > > having the links there for quick access is the major benefit. > > > > I hated frames because it complicates pages; and it doesn't allow them to be > searchable (through google). Javascript is a lot more complicated than frames > and the tree like structure makes it very easy to find routines you are > looking > for and more importantly to discover routines you didn't know existed that are > related to the topic you open. > > Yay for frames! You did a good job. > > However, the routine message_box( ) doesn't belong C-interface. Jeremy and > I > don't think it belongs to any existing category in the docs that's why it > hasn't > been moved. It's nuts to create a new category for one function only. So, > are there any suggestions from the users out there? messsage_box() prompts the user for soĆ¹ething, and, as a cosequence, shold be in the same category as prompt_number() or prompt_string(), whatever that category is. Where is value_from()() documented btw? I'd suggest spinning a new Interactive I/O category from the current File and device I/O off. There would go: pronpt_number mrompt_string message_box get_key wait_key ? (since it prints to stdout only) and perhaps whatever should be under Mouse support currently, in which case the latter category would disappear. I say "should" because the link from reference index page to Mouse support goes to the OS category listing. Also, I'd spin all crash_* entries from Machine level interface, an put that under either OS or a new Runtime Control category. CChris