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

Re: XML parsers use what computational power?

• From: Michael Sokolov <msokolov@safaribooksonline.com>
• To: "Simon St.Laurent" <simonstl@simonstl.com>
• Date: Mon, 08 Apr 2013 09:14:16 -0400

On 04/08/2013 08:50 AM, Simon St.Laurent wrote:
> On 4/8/13 8:10 AM, Costello, Roger L. wrote:
>> Hi Folks,
>>
>> Are there any XML parsers that are more powerful than a deterministic
>> pushdown automata?
>>
>> Why are XML parsers written as libraries rather than as parser
>> generators?
>
> I have seen at least one XML parser generator, though I think it was 
> at XTech long long ago.  They've also come up in conversation about 
> ASN.1, but I suspect that was more on the ASN.1 side than the XML side.
>
> Thanks,
I (and I think some others) wrote MicroXML parsers using parser 
generators, like JavaCC.  I haven't tried that with XML, so I'm not 
sure, but I think that the DTD and entity handling is difficult to do 
with classic parser generators.  Also you will tend to get better 
performance from a hand-written parser.

-- 
Michael Sokolov
Senior Architect
Safari Books Online

• References:
  • XML parsers use what computational power?
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: XML parsers use what computational power?
    • From: "Simon St.Laurent" <simonstl@simonstl.com>