WpsTokenizerDEPRECATED < Main

Downloads

release

dev builds

extras

themes

Documentation

manual

wiki

device status

Support

forums

mailing lists

IRC

Development

bugs

patches

dev guide

Wiki > Main > WpsTokenizerDEPRECATED

---+!! WPS tokenizer

%TOC%

---++Introduction

%W% This article is Work-in-progress

This page aims to explain the principles of the WPS tokenizer and give all the information needed to maintain and extend it.

---++Code structure explanation

---+++The parser

The main function for WPS loading and parsing is =wps_data_load()=. It does the following:
   * reset the WPS data structure
   * copy the WPS source to the plugin buffer
   * parses the source
   * load the bitmaps

<img src="%ATTACHURLPATH%/call_graph_parser.png" alt="call_graph_parser.png" width='684' height='533' />

---+++The renderer

<img src="%ATTACHURLPATH%/call_graph_display.png" alt="call_graph_display.png" width='1076' height='519' />

---++ How to add a WPS tag

   * Add a new token to ==enum wps_token_type== in =gwps.h=.
   * Add your token and the corresponding WPS tag to the ==all_tags== array in =wps_parser.c=. If necessary, add a special parsing function that will be called when the tag is encountered.
   * Add a case for the new token to the switch in ==get_tag()== in =gwps-common.c=.
   * Add a case for the new token to the switch in ==dump_wps_tokens()== in =wps_debug.c= (this isn't vital but helpful).

r2 - 06 Sep 2010 - 12:04:06 - JonathanGordon