New user, evaluating XML libraries

On Thu, Dec 21, 2006 at 09:03:21AM +0000, Boris Kolpackov
wrote:
> > ... SAX not formerly defined except for Java. On
the other hand the
> > XMLTextReader from C# is part of the ECMA C# spec,
and is a good API.
> 
> You mean SAX, which is only defined for Java, is a bad
thing but
> XMLTextReader, which is only define for C#, is good?


  API wise, definitely, yes.

> >   Also DOM *requires* UTF-16 for all strings. This
means that in general
> > 1/ you will loose time, most content around is
UTF-8
> > 2/ you will loose memory space/cache efficiency as
the converted output is
> >    way larger in average
> > 3/ you will looose CPU efficiency as breaking
cache is #1 performance
> >    problem in modern computers
> 
> I think it largely depends on what kind of data one is
handling. If there
> are a lot of non-latin characters then UTF-8 will waste
at leaset as much
> and normally a lot more (4-byte surrogate pairs) space
than UTF-16.

  all the markup is usually ASCII. And no UTF-16 XML
instances are not
common.

> >   There is no functional XSD validator. Go to the
xmlschemas-dev
> > archive at W3C, check the last 5 questions from
Michael Kay (who is
> > a Schemas implementor and one of the W3C spec
writers), they are unanswered
> > for weeks now, nobody can tell what it is supposed
to do. Trying to use
> > XSD to promote interoperability or validation of
data is kind of a joke.
> > Relax-NG on the other hand is an ISO standard, has
a formal specification
> > and can be read and understood by most programmers
in a matter of a couple
> > of days.
> 
> While what you say could be true, most XML parsers
these days support
> XML Schema while there is hardly any that support
Relax-NG. Also there

  Technically nor XSD nor RNG are handled by parsers, it's
sitting on top.

> are a lot of industry standards organizations that
managed to define
> interoperable schemas that describe fairly complex XML
vocabularies.
> Finally, one last thing, that a lot of people tend to
overlook, is that
> it is fairly straightforward (in most cases) to provide
XML data
> binding (generated types in a programming language that
represent the
> given vocabulary) from XML Schema. I do not think it
will be anywhere
> as easy or at all possible to do the same with
Relax-NG.

  When you mean data bindings, it's records, even easier in
RNG than XSD,
and RNG usually reuses XSD datatypes. I see no positive
differential for 
XSD here. Anyway the initial question was *not* about data
bindings, 
and further exchanges showed that XSD (nor RNG) was not part
of the picture,
so please could you keep this focused on what is really
needed and not 
what you would like to promote, thanks.

Daniel

-- 
Red Hat Virtualization group http://redhat.com/v
irtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
veillardredhat.com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ |
Rpmfind RPM search engine  http://rpmfind.net/
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
xmlgnome.org
http://mai
l.gnome.org/mailman/listinfo/xml

Gentlemen,

I welcome any discussion on this topic whether it extends
beyond the
scope of libxml itself or not.  As is typically the case,
the question
to be answered has morphed beyond its original scope.  The
inevitable
"what about this?", "what about that?"
questions have been asked and
people higher than me in this process have to sit down and
honestly
answer the question "What job are we really trying to
do here with XML?"
before even attempting to make a decision on which library
to apply to
it.  

Considerations involving ebXML, XSD, XSD-I, XSLT and the
like are all on
the table.  And perhaps just one library won't suffice.  I
can see two
different jobs to be done here and I won't dismiss the
possibility that
no one library is appropriate for both of them.  And from my
position,
that's the approach I'm taking, looking at it from the
standpoint of
appropriateness for the job first, assuming I know what the
job is.  The
others will have to make the tradeoffs between performance
vs. whatever,
what we know we need vs. what we think we might need,
homogeneous
solution vs. heterogeneous one.

While that is taking place, any information that people with
more
knowledge and experience on this subject than I wish to
offer can only
help the process and result in a more informed decision. 
Thank you all
for your contributions up to this point.

-will

-----Original Message-----
From: xml-bouncesgnome.org [mailtoml-bo
uncesgnome.org] On Behalf Of
Daniel Veillard
Sent: Thursday, December 21, 2006 10:38 AM
To: Boris Kolpackov
Cc: xmlgnome.org
Subject: Re: [xml] New user, evaluating XML libraries

On Thu, Dec 21, 2006 at 09:03:21AM +0000, Boris Kolpackov
wrote:
> > ... SAX not formerly defined except for Java. On
the other hand the
> > XMLTextReader from C# is part of the ECMA C# spec,
and is a good
API.
> 
> You mean SAX, which is only defined for Java, is a bad
thing but
> XMLTextReader, which is only define for C#, is good?


  API wise, definitely, yes.

> >   Also DOM *requires* UTF-16 for all strings. This
means that in
general
> > 1/ you will loose time, most content around is
UTF-8
> > 2/ you will loose memory space/cache efficiency as
the converted
output is
> >    way larger in average
> > 3/ you will looose CPU efficiency as breaking
cache is #1
performance
> >    problem in modern computers
> 
> I think it largely depends on what kind of data one is
handling. If
there
> are a lot of non-latin characters then UTF-8 will waste
at leaset as
much
> and normally a lot more (4-byte surrogate pairs) space
than UTF-16.

  all the markup is usually ASCII. And no UTF-16 XML
instances are not
common.

> >   There is no functional XSD validator. Go to the
xmlschemas-dev
> > archive at W3C, check the last 5 questions from
Michael Kay (who is
> > a Schemas implementor and one of the W3C spec
writers), they are
unanswered
> > for weeks now, nobody can tell what it is supposed
to do. Trying to
use
> > XSD to promote interoperability or validation of
data is kind of a
joke.
> > Relax-NG on the other hand is an ISO standard, has
a formal
specification
> > and can be read and understood by most programmers
in a matter of a
couple
> > of days.
> 
> While what you say could be true, most XML parsers
these days support
> XML Schema while there is hardly any that support
Relax-NG. Also there

  Technically nor XSD nor RNG are handled by parsers, it's
sitting on
top.

> are a lot of industry standards organizations that
managed to define
> interoperable schemas that describe fairly complex XML
vocabularies.
> Finally, one last thing, that a lot of people tend to
overlook, is
that
> it is fairly straightforward (in most cases) to provide
XML data
> binding (generated types in a programming language that
represent the
> given vocabulary) from XML Schema. I do not think it
will be anywhere
> as easy or at all possible to do the same with
Relax-NG.

  When you mean data bindings, it's records, even easier in
RNG than
XSD,
and RNG usually reuses XSD datatypes. I see no positive
differential for

XSD here. Anyway the initial question was *not* about data
bindings, 
and further exchanges showed that XSD (nor RNG) was not part
of the
picture,
so please could you keep this focused on what is really
needed and not 
what you would like to promote, thanks.

Daniel

-- 
Red Hat Virtualization group http://redhat.com/v
irtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
veillardredhat.com  | libxml GNOME XML XSLT toolkit
http://xmlsoft.org/
http://veillard.com/ |
Rpmfind RPM search engine  http://rpmfind.net/
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
xmlgnome.org
http://mai
l.gnome.org/mailman/listinfo/xml
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
xmlgnome.org
http://mai
l.gnome.org/mailman/listinfo/xml