Archive of UserLand's first discussion group, started October 5, 1998.

Re: XML-RPC Spell Checker

Author:Jason Levine
Posted:4/14/2000; 10:40:03 AM
Topic:XML-RPC Spell Checker
Msg #:16260 (In response to 16258)
Prev/Next:16259 / 16261

Coool. I intend to make mine (if I do one) fully XML-RPC, but there's the tricky problem of what the function will return -- does it return an HTML page that has the entire marked-up copy of text, does it return an array with two items, one the text and the other a sub-array of suggested corrections, or does it do something else?

I brought up the Manila question because it would be cool to start defining the method whereby a callback could actually require further processing before it allowed the posting to occur; if we started defining that now, then it would make development of XML-RPC functions that are intended to serve as those callbacks a lot easier.

Perhaps it could be something as simple as checking the return of the callback function; functions could return either a straight boolean true (which would mean no further processing was necessary), or they could return arrays with predefined elements (e.g., the first would be true/false, the second would be the XML-RPC hostname on which the function resides, the third would be the port, the fourth would be the function name that would be called by Manila once the data was processed, and the rest would be the data elements that had to be processed in some way).

Dunno. But it definitely matters when we're talking about how to implement something like this; a good example of why is David Adam's current spelling app, which (so far as I can tell) is bound to being run with the assumption that it generates the HTML and completely controls the functions to do the spell correction and whatnot.

/j

There are responses to this message:

This page was archived on 6/13/2001; 4:54:47 PM.

© Copyright 1998-2001 UserLand Software, Inc.