
www.Usenet.com
| <-- __Chronological__ --> | <-- __Thread__ --> |
"Fu, Ren-Li" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > Was there any idea on how to process the post-text of a name? I.e. 'an' in > 'an orc', 'an apple', but what about 'a yumi'? How does the game know the > a/an difference? I can think of making a g/name/s structure.. "a/yumi/s" and > "an/apple/s". But this wasn't done for rogue or nethack. How do they know > what to put on the screen? thanks. OK, for the use of a/an, you should be quite capable of getting by using the simple rule of an before vowels, a before consonants. The way to deal with the exceptions is to either put in code dealing with them or avoid using them. Avoiding using them is most likely going to be your best bet. I think that I could personally avoid ever needing to use 'euphoric' or 'hourglass' in a generated text message. As far as pluralization, I'd use a table lookup for the nonstandard ones, and I'd put the responsibility for that on whatever is converting from the object or struct to a string. YMMV, void where prohibited. Paul
| <-- __Chronological__ --> | <-- __Thread__ --> |