[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Genx

• To: <xml-dev@l...>
• Subject: Re: Genx
• From: "Karl Waclawek" <karl@w...>
• Date: Tue, 20 Jan 2004 10:56:58 -0500
• References: <200401201509.i0KF9v8P088186@a...> <400D4BF0.9000004@b...>

----- Original Message ----- 
From: "Joe Gregorio" <joe@b...>
> > Have you looked at the API of James Clark's Expat?
> 
> Based on your question I just took a look at
> Expat and saw a mix of calls in xmlparse.h,
> some using strings that are null terminated, 
> some that are not:
> 
>   /* s is not 0 terminated. */
>   typedef void (*XML_CharacterDataHandler)(void *userData,
> const XML_Char *s,
> int len);
> 
>   /* target and data are 0 terminated */
>   typedef void (*XML_ProcessingInstructionHandler)(void *userData,
> const XML_Char *target,
> const XML_Char *data);
> 
>   /* data is 0 terminated */
>   typedef void (*XML_CommentHandler)(void *userData, const XML_Char *data);
> 
> 
> Now this was a quick look and I could be poking
> around in the wrong file, please point me
> it the right direction if that's the case.

The right file in the current Expat would be expat.h, but that does not 
change your observation. As far as I can tell there are two reasons:

1) Due to a low frequency, the impact of copying is low for some call-backs
2) Based on its internal logic Expat needs to copy some data out anyway
   (e.g. it has to remember the name of a start tag to check for the proper
    occurrence of the end-tag). In that case Expat will pass the internal copy
    which is a null-termninated string. 

Karl

• References:
  • Re: Genx
    • From: David Tolpin <dvd@d...>
  • Re: Genx
    • From: Joe Gregorio <joe@b...>

• Prev by Date: Re: Genx
• Next by Date: Re: XML Journal System
• Previous by thread: Re: Genx
• Next by thread: Re: Genx
• Index(es):
  • Date
  • Thread