• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: ANNOUNCE: XML Tools 2.3.2
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ANNOUNCE: XML Tools 2.3.2


  • Subject: Re: ANNOUNCE: XML Tools 2.3.2
  • From: Brennan <email@hidden>
  • Date: Wed, 21 Nov 2001 22:49:54 +0100

Arthur J Knapp <email@hidden> wrote:

> > Damn it, No!
>
> Wow.
>
>
> > ... These folks need to fix their documents. XML is
> > deliberately draconian about error handling. This is a feature, not a
> > bug. It ensures interoperability between different tools. You're
> > encouraging the development of a class of incompatible documents that
> > will break when passed to different parsers.
>
> Which would be their fault, not Mark's.

I must side with the dammit Nos.

XML must be strictly followed. It's not as if it's difficult to have documents with leading spaces dealt with by some other more specialised pre-process.

You want each custom parser to check for leading whitespace characters? Any other special cases they should look out for? Most don't do this stuff right now and it would take valuable time away from more important tasks for developers to implement that. Besides, it's explicitly *not* part of the XML standard.

> Perhaps there could be an optional "strict" parameter of XML Tools,
> that would allow a user to define the level of "loose" parsing?

If it must be such, then 'strict' should be the default, and it should not be called 'loose', it should be called 'leading whitespaces allowed'. (I wonder what other special cases of xml-like documents might come under the category of 'loose'?)

> It is a document's adherence to the standard itself that defines
> interoperability, not the degree of "looseness" that a given parser
> allows.

Damn it, No!

If I see a file called "whatever.xml", I want to be able to rely on the contents being formatted in a way that I (and my code) expects. Special cases means wasted man hours. People can make whatever loose intepretations of XML they like, and parsers can have special flags for dealing with them, but these document types should not be called XML, and the default provided by any given XML parser should be strict.

The strictness is not an essential part of XML because of over-zealous potty training, its because people are tired of wasting time, and their bosses are tired of paying for them to do it.

(Perhaps you can detect some venom in this post, but I just spent an extra week on a job, compromising a deadline because of poorly formed documents where several small groups of a dozen or so each had their own intelligent-and-carefully-considered-but-underdocumented quirks. Arrgh!).

--
_____________

Brennan Young

Artist, Composer and Multimedia programmer

mailto:email@hidden

In software, the chain isn't as strong as the weakest link; it's as weak as all the weak links multiplied together.
-Steve McConnell


  • Prev by Date: Re: Entourage: Changing current folder viewed
  • Next by Date: Re: ANNOUNCE: XML Tools 2.3.2
  • Previous by thread: Re: ANNOUNCE: XML Tools 2.3.2
  • Next by thread: Re: ANNOUNCE: XML Tools 2.3.2
  • Index(es):
    • Date
    • Thread