@

Tuesday, March 12, 2002

StimpSoft has closed its doors. What a shame. Everything about StimpSoft is cool. Cool apps. Cool website. Cool outlook. And everything was written in REALbasic. It looks like StimpSoft's owner is on to bigger and better things. I wish him well. As a side note, I actually tried to recruit him. He didn't take me seriously. But I was serious. I think he has a real eye for UI. comment ()  11:46:11 PM  #   I think I was wrong about whitespace and XML. Apparently there has been a change. Whitespace is now preserved. All the tabs, spaces,and such, are preserverd. And I just fixed my parser so that it handles text elements. Actuallly it wasn't that bad and I think it is still simple enough to explain it. comment ()  11:13:15 PM  #   I've run into a conceptual snag with my XML parser. I need to be able to store user templates in it. But, in order to simplify it, I've decided not to implement text elements. That's a big omission, on the face of it. But in fact I can do everything with tag attributes that I can do with text elements. There's only one problem. Runs of spaces in a tag attribute are collapsed to one space. Furthermore, all whitespace (tabs, carriage returns, etc.) are turned into a space character (ASCII 32). If I put a template in there, then I have no guarantee that some other XML parser won't screw it up. The only solution I can see right now is to entity encode all of those characters. So all spaces and carriage returns and such will become entities. comment ()  9:31:12 PM  #   The Document Object Model is a platform- and language-neutral interface that will allow programs and scripts to dynamically access and update the content, structure and style of documents. [w3c.org] I always thought DOM was a philosophy about parsing XML. That philosophy being that the parser constructs an entire parsed tree of the XML before any processing takes place. As opposed to the philosophy of SAX, which says to fire off events to the user wheneverfor every XML element encountered. A SAX parser is free to rip through the XML as fast as it can, calling back to the user as it goes. It is up to the user to handle the parser's event. So a DOM parser is data-centric and a SAX parser is event-centric. But it turns out that there is actually a DOM API. I didn't know that. A compliant DOM implementation must implement the DOM API. This is interesting from the standpoint of the little XML parser I'm writing in REALbasic. I think that trying to implement even a subset of the DOM API is going to be too complicated for the very simple parser I am trying to write. But it would be cool if I could do it. comment ()  11:58:16 AM  #

© Copyright 2002 Will Leshner. Last update: 3/12/02; 11:58:22 AM.