Re: SAX: Parser Factory class

John Wilson (tug@wilson.co.uk)
Mon, 20 Apr 1998 11:20:05 +0100

This is a multi-part message in MIME format.

------=_NextPart_000_00A6_01BD6C4E.4500EF60
Content-Type: multipart/alternative;
boundary="----=_NextPart_001_00A7_01BD6C4E.4500EF60"

------=_NextPart_001_00A7_01BD6C4E.4500EF60
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

The Java Crypto Extensions allow you to request a generic implementation =
of an algorithm or a specific implementation. This is bases on having a =
set of "Providers" and a set of engine classes (essentially class =
factories). When calling getInstance() on the appropriate engine class =
you can specify either that you want any implementation of the algorithm =
or that you want a specific implementation from a specific provider. =
Providers are installed by editing a configuration file or by adding =
them dynamically using Security.addProvider(). Providers are ordered so =
a request for a generic implementation will be serviced by the first =
provider found which offers that algorithm.

It might be possible to do such a thing for SAX parsers, but you would =
really have to have an agreed set of functions that parsers did or did =
not implement. If you would then be able to say:

"give me an instance of a parser that at minimum does this, this and =
this"

or

"give me an instance of the IBM parser"

you should also be able to say

"give me an instance of the IBM parser that does this, that and the =
other"

This raises two questions:

1/ Could everybody agree on the set of functions?

2/ Is there a real need for a general solution to the problem of working =
with several parsers?
-----Original Message-----
From: Andrew n marshall <amarshal@usc.edu>
To: Xml-Dev List <xml-dev@ic.ac.uk>
Date: 16 April 1998 16:15
Subject: SAX: Parser Factory class
=20
=20
While it is great that there are so many SAX parsers available, the =
thing that bothers me is that my application has to know what Parser to =
load in just to use SAX. And this means that even is a SAX parser is =
available on the users machine, they may still have to download the one =
I use to to run my application.
=20
If there is already a solution for this, then please let me know.

------=_NextPart_001_00A7_01BD6C4E.4500EF60
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">

The Java Crypto Extensions allow you = to request=20 a generic implementation of an algorithm or a specific implementation. = This is=20 bases on having a set of "Providers" and a set of engine = classes=20 (essentially class factories). When calling getInstance() on the = appropriate=20 engine class you can specify either that you want any implementation of = the=20 algorithm or that you want a specific implementation from a specific = provider.=20 Providers are installed by editing a configuration file or by adding = them=20 dynamically using Security.addProvider(). Providers are ordered so a = request for=20 a generic implementation will be serviced by the first provider found = which=20 offers that algorithm.
 
It might be possible to do such a = thing for SAX=20 parsers, but you would really have to have an agreed set of functions = that=20 parsers did or did not implement. If you would then be able to = say:
 
    "give me an = instance of=20 a parser that at minimum does this, this and this"
 
or
 
    "give me an = instance of=20 the IBM parser"
 
you should also be able to say
 
    "give me an = instance of=20 the IBM parser that does this, that and the other"
 
This raises two questions:
 
1/ Could everybody agree on the set of = functions?
 
2/ Is there a real need for a general solution to = the problem=20 of working with several parsers?
-----Original = Message-----
From:=20 Andrew n marshall <amarshal@usc.edu>
To:=20 Xml-Dev List <xml-dev@ic.ac.uk>
Date: = 16=20 April 1998 16:15
Subject: SAX: Parser Factory=20 class

While it is great that there are so many SAX parsers = available, the=20 thing that bothers me is that my application has to know what Parser = to load=20 in just to use SAX.  And  this means that even is a SAX = parser is=20 available on the users machine, they may still have to download the = one I=20 use to to run my application.
 
If=20 there is already a solution for this, then please let me=20 know.
 
------=_NextPart_001_00A7_01BD6C4E.4500EF60-- ------=_NextPart_000_00A6_01BD6C4E.4500EF60 Content-Type: text/x-vcard; name="John Wilson.vcf" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="John Wilson.vcf" BEGIN:VCARD VERSION:2.1 N:Wilson;John;G. FN:John Wilson ORG:The Wilson Partnership TEL;WORK;VOICE:+44 1296 641072 TEL;CELL;VOICE:+44 976 611010 TEL;WORK;FAX:+44 1296 641874 ADR;WORK;ENCODING=3DQUOTED-PRINTABLE:;;5 Market = Hill=3D0D=3D0AWhitchurch;Aylesbury;Buckinghamshire;HP22 4JB;UK LABEL;WORK;ENCODING=3DQUOTED-PRINTABLE:5 Market = Hill=3D0D=3D0AWhitchurch=3D0D=3D0AAylesbury, Buckinghamshire HP22 = 4JB=3D0D=3D =3D0AUK URL: URL:http://www.wilson.co.uk/ EMAIL;PREF;INTERNET:tug@wilson.co.uk REV:19980420T102005Z END:VCARD ------=_NextPart_000_00A6_01BD6C4E.4500EF60-- xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ To (un)subscribe, mailto:majordomo@ic.ac.uk the following message; (un)subscribe xml-dev To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message; subscribe xml-dev-digest List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)