Martin Klang | 4 Jun 20:16 2003
Picon

hosting issues


as you may have noticed, there's been some problems with the o:XML web
site, mailing list and CVS over the last week or so.

This was at first due to our DNS service provider going down and -
repeatedly! - loosing our DNS configuration. These problems were
eventually resolved by me setting up a new DNS server.

Then the whole subnet with the o:XML servers went down! One of the servers
was hacked, and the machine was used in a DoS attack against a
neighbouring subnet, so the hosting provider shut us down completely. The
vulnerability (a PHP application exploit) was fixed and the machines put
back online.
Unfortunately Easynet forgot to remove a bandwidth throttle, which
resulted in download rates of sometimes just a few hundred bytes/s! This
problem wasn't fixed until yesteday, but finally we're back at full
capacity.

Now, there might be some more downtime in the near future as o:XML may be
changing hosting providers. There shouldn't be more than a days loss of
service, and I will let you know what the schedule is when a deal has been
agreed.

sorry about the downtime, and excuses in advance if the site disappears
for a day or so next week, but after that we should be running faster and
more reliably than ever - regards,

/m

Martin Klang
(Continue reading)

Yves Moisan | 6 Jun 14:26 2003

Water vs o:XML

Hi,

Just bumped across a site (http://www.waterlanguage.org/) that presents a
language called Water that is deemed to be the "the only [language] that
actually executes XML"  How does that compare with o:XML ?
Martin Klang | 6 Jun 15:16 2003
Picon

Re: Water vs o:XML


I remember a rather braindead thread on xml-dev regarding Concise-XML, the
XML-'compatible' format they use ->
http://lists.xml.org/archives/xml-dev/200301/threads.html#00322

o:XML is built on the idea that there are significant advantages to
representing programming logic in an interchangeable, processable format
such as XML.
It also tries to take advantage of the data formats that are native to XML
- strings and trees - and XPath. The point is that XML should be enabling
and empowering rather than restrictive.

Water seems to go the opposite way in fitting XML around all the
traditional programming data formats and constraints, and lands us with a
solution where the actual XML seems extrinsic to the processing model.

At least that's my impression. I think the attitude is reflected in many
places in the article 'the trouble with XML' [1], eg:
"The most common and most important kind of computer data is code."
/snip/
"Lack of by-position arguments and the lack of elements as attribute
values makes XML a lousy syntax for the most important kind of data on a
computer."
It really sounds like someone with a rather narrow programmers perspective
on XML in particular and IT in general.

regards,

/m

(Continue reading)

Ken McCloskey | 19 Jun 20:57 2003
Picon

Documentation suggestion and question

Hi,

I'm new to o:XML, trying to discover how to use it and the caveats of doing 
so.

One thing I really like about o:XML is the database extensions. However the 
documentation in my opinion has a serious omission. To get the database 
extensions to work you have to do:

java org.oXML.engine.ObjectBox -e org.oXML.extras.db.DatabaseExtensions

This isn't mentioned in the docs. A reasonably intuitive person might 
surmise that, but I think many would not.

Also, a question, does the output of an o:XML program always need to include 
the <program> tag? I'd prefer if my output did not include it.

Thanks for your work on this interesting project.

Ken

_________________________________________________________________
MSN 8 with e-mail virus protection service: 2 months FREE*  
http://join.msn.com/?page=features/virus
Ken McCloskey | 19 Jun 22:29 2003
Picon

processing-instruction/comment/text

<o:processing-instruction target="'xml-stylesheet'">href="style.xsl" 
type="text/xsl"</o:processing-instruction>
<o:comment select="'i am commenting this'"/>
<o:text>  leading and trailing whitespace might otherwise have been removed  
</o:text>

None of the above seem to work correctly.

My output is as follows:

<processing-instruction target="'xml-stylesheet
'">href="style.xsl" type="text/xsl"</processing-instruction>
<comment select="'i am commenting this'"/>
<text>  leading and trailing whitespace might otherwise have been removed  
</text>

Do I need to have an extension loaded in order to make this work correctly?

Ken McCloskey
818.281.7811

_________________________________________________________________
Tired of spam? Get advanced junk mail protection with MSN 8. 
http://join.msn.com/?page=features/junkmail
Bill Keese | 20 Jun 04:07 2003
Picon

threads?

I just read the manual for o:XML and it looks cool.  I like the way that input and output to functions, etc. is all XML, and I like how you can use X-Path anywhere.  Basically, o:XML is the scripting language equivalent of XSLT.
 
However, although you can expand the language by defining your own procedures (ie, defining your own tags), I can't figure out anyway to fire off multiple threads and then wait for them to finish.  Any ideas?
 
Bill
martin | 20 Jun 10:33 2003
Picon

Re: threads?


Bill, threads will make it into o:XML later this year. In the meantime you
can use the java extensions to create a type that derives from
java.lang.Thread. I'm planning to start experimenting with threads when i
get back from hols in a couple of weeks - I'll keep you posted.

cheers,

/m

On Fri, 20 Jun 2003, Bill Keese wrote:

> I just read the manual for o:XML and it looks cool.  I like the way that
> input and output to functions, etc. is all XML, and I like how you can use
> X-Path anywhere.  Basically, o:XML is the scripting language equivalent of
> XSLT.
>
> However, although you can expand the language by defining your own
> procedures (ie, defining your own tags), I can't figure out anyway to fire
> off multiple threads and then wait for them to finish.  Any ideas?
>
> Bill
>
martin | 20 Jun 10:45 2003
Picon

Re: processing-instruction/comment/text


Ken, the <text> tags still aren't fully implemented in the ObjectBox, and
Comment exists as a type but maybe not as a <comment> mapping. PI's i used
just the other day though! what version are you running, 0.9.5?

maybe try
<o:eval select="Comment('blabla')"/>
<o:eval select="String('text')"/>
etc

I'm sorry if you were working through the examples and they weren't
functioning with the current version of ObjectBox. I promise to go through
it all and tie up the loose ends when i return to the UK in a couple of
weeks.

cheers!

/m

On Thu, 19 Jun 2003, Ken McCloskey wrote:

> <o:processing-instruction target="'xml-stylesheet'">href="style.xsl"
> type="text/xsl"</o:processing-instruction>
> <o:comment select="'i am commenting this'"/>
> <o:text>  leading and trailing whitespace might otherwise have been removed
> </o:text>
>
> None of the above seem to work correctly.
>
> My output is as follows:
>
> <processing-instruction target="'xml-stylesheet
> '">href="style.xsl" type="text/xsl"</processing-instruction>
> <comment select="'i am commenting this'"/>
> <text>  leading and trailing whitespace might otherwise have been removed
> </text>
>
> Do I need to have an extension loaded in order to make this work correctly?
>
> Ken McCloskey
> 818.281.7811
>
> _________________________________________________________________
> Tired of spam? Get advanced junk mail protection with MSN 8.
> http://join.msn.com/?page=features/junkmail
>
> _______________________________________________
> o-xml mailing list
> o-xml@...
> http://lists.pingdynasty.com/mailman/listinfo/o-xml
>
martin | 20 Jun 11:18 2003
Picon

Re: Documentation suggestion and question


Hey Ken, I'm glad you like it!
You're right about the docs, there should be a section on how to enable
the extensions. I'm just not sure the command-line way is really going to
be standard. I'll include it for the 1.0 release later this summer though.
It would be nice to do some more work on the db extensions, at least to
handle transactions and connection pooling cleanly.

The <program> tag is just a placeholder. you can embedd o:XML anywhere in
an XML document - it's designed that way to make inline dynamic processing
easier. For example I use it on the o:XML web site to generate the
non-static content. The o:XML tags are embedded in DocBook files and
processed by an ant buildscript, which then does an XSL transformation to
generate the HTML and PDF output.
Would like to write an article on using o:XML for publishing purposes,
just need to find the time...

later,

/m

On Thu, 19 Jun 2003, Ken McCloskey wrote:

> Hi,
>
> I'm new to o:XML, trying to discover how to use it and the caveats of doing
> so.
>
> One thing I really like about o:XML is the database extensions. However the
> documentation in my opinion has a serious omission. To get the database
> extensions to work you have to do:
>
> java org.oXML.engine.ObjectBox -e org.oXML.extras.db.DatabaseExtensions
>
> This isn't mentioned in the docs. A reasonably intuitive person might
> surmise that, but I think many would not.
>
> Also, a question, does the output of an o:XML program always need to include
> the <program> tag? I'd prefer if my output did not include it.
>
> Thanks for your work on this interesting project.
>
> Ken
>
> _________________________________________________________________
> MSN 8 with e-mail virus protection service: 2 months FREE*
> http://join.msn.com/?page=features/virus
>
> _______________________________________________
> o-xml mailing list
> o-xml@...
> http://lists.pingdynasty.com/mailman/listinfo/o-xml
>

Gmane