unexpected query results against XML data source
by Mike Mascari
Hello. I've downloaded and installed the snapshot as of today (8/31/2009) and, using the designer:
1. Used Import->XML Schema as Relational Source Model
2. Selected Document over HTTP with no request parameters
3. Selected No Catalog
4. Chose "File System" as source of XML Schema (inlined below)
5. Chose "client" (I've tried w/"clients" as well) as the root element and "Finish"
6. Create a "purchasing" View Model and, with "purchasing" selected, created an "example" BaseTable. I then drag-and-drop "client" from clients.xmi Source Model into SOURCES pane. (This View abstraction turns out not to be the problem -- I query the Source via the Execute'd VDB with the same outcome.)
7. Created a "billing" vdb, cleaned, built, and synchronized.
8. Then, with this data served up by apache:
<?xml version="1.0" encoding="UTF-8"?>
<clients>
<client>
<name>acme</name>
<rate>15</rate>
</client>
<client>
<name>toolwerks</name>
<rate>20</rate>
</client>
</clients>
I issued the following query:
select * from clients.client
and received the expected result:
null;acme;15
null;toolwerks;20
9. Then I did this:
select * from clients.client where name = 'acme'
and received *two rows*:
null;acme;15
null;toolwerks;20
10. So I did this:
select * from clients.client where name != 'acme'
and received the expected one row:
null;toolwerks;20
???
Is there something goofy with the equality operator against XML data? Is there flag that needs flipped on these attributes? I went with the "string" mapping on the 'XML Schema as Relational Source Model' dialog wizard (pane 4 IIRC). Oughtn't I have? I also changed XML node names that might have been SQL keywords ('name' -> 'fullname') but no dice...
Thoughts?
- Mike
Schema:
<?xml version="1.0"?>
<!-- our client database -->
<xs:element name="clients">
<xs:complexType>
<xs:sequence>
<!-- the client -->
<xs:element name="client" minOccurs="1" maxOccurs="unbounded">
<xs:complexType>
<xs:sequence>
<!-- client name -->
<xs:element name="name" type="xs:string" minOccurs="1" maxOccurs="1"/>
<!-- bill rate -->
<xs:element name="rate" minOccurs="0">
<xs:simpleType>
<xs:restriction base="xs:decimal">
<xs:minInclusive value="50"/>
<xs:maxInclusive value="250"/>
</xs:restriction>
</xs:simpleType>
</xs:element>
</xs:sequence>
</xs:complexType>
</xs:element>
</xs:sequence>
</xs:complexType>
</xs:element>
</xs:schema>
15 years, 4 months
What should we call this artifact?
by Ramesh Reddy
Hi,
You all know that starting with 6.2 release we have a new feature called
"Connector Supplied Metadata", where connectors will be able to import
metadata dynamically at runtime for data integration purposes.
To use this feature user need to write ".def" file, which we been
calling internally "vdbLess" mode, that does not sound correct. We are
building a VDB and all the access is to this no different than a regular
VDB, so it is a VDB just not built using the Designer!
I have been calling it "DEF file based VDB", some names thrown out are
"dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
so that we all can refer it by the same name?
Thanks
Ramesh..
15 years, 4 months
Re: [teiid-users] teiid-users Digest, Vol 4, Issue 5
by Charles Simon
I think "Dynamic VDB" best describes the feature. The VDB is being
built at run time (aka, dynamically) from a meta data stream.
teiid-users-request(a)lists.jboss.org wrote:
> Send teiid-users mailing list submissions to
> teiid-users(a)lists.jboss.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://lists.jboss.org/mailman/listinfo/teiid-users
> or, via email, send a message with subject or body 'help' to
> teiid-users-request(a)lists.jboss.org
>
> You can reach the person managing the list at
> teiid-users-owner(a)lists.jboss.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of teiid-users digest..."
>
>
> Today's Topics:
>
> 1. What should we call this artifact? (Ramesh Reddy)
> 2. Re: What should we call this artifact? (Larry O'Leary)
> 3. Re: [teiid-dev] What should we call this artifact? (Ken Johnson)
> 4. Re: [teiid-dev] What should we call this artifact? (Ted Jones)
> 5. Re: [teiid-dev] What should we call this artifact?
> (Rama Kishore Yelamanchilli)
> 6. Re: What should we call this artifact? (John Doyle)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 28 Aug 2009 15:05:56 -0500
> From: Ramesh Reddy <rareddy(a)redhat.com>
> Subject: [teiid-users] What should we call this artifact?
> To: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Message-ID: <1251489956.3339.19.camel(a)localhost.localdomain>
> Content-Type: text/plain
>
> Hi,
>
> You all know that starting with 6.2 release we have a new feature called
> "Connector Supplied Metadata", where connectors will be able to import
> metadata dynamically at runtime for data integration purposes.
>
> To use this feature user need to write ".def" file, which we been
> calling internally "vdbLess" mode, that does not sound correct. We are
> building a VDB and all the access is to this no different than a regular
> VDB, so it is a VDB just not built using the Designer!
>
> I have been calling it "DEF file based VDB", some names thrown out are
> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
> so that we all can refer it by the same name?
>
> Thanks
>
> Ramesh..
>
>
>
> ------------------------------
>
> Message: 2
> Date: Fri, 28 Aug 2009 20:36:34 +0000
> From: "Larry O'Leary" <loleary(a)redhat.com>
> Subject: Re: [teiid-users] What should we call this artifact?
> To: Ramesh Reddy <rareddy(a)redhat.com>
> Cc: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Message-ID: <1251491795.2925.62.camel@loleary>
> Content-Type: text/plain
>
> On Fri, 2009-08-28 at 15:05 -0500, Ramesh Reddy wrote:
>
>> To use this feature user need to write ".def" file, which we been
>> calling internally "vdbLess" mode, that does not sound correct. We are
>> building a VDB and all the access is to this no different than a regular
>> VDB, so it is a VDB just not built using the Designer!
>>
>> I have been calling it "DEF file based VDB", some names thrown out are
>> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
>> so that we all can refer it by the same name?
>>
>
> How about:
> Model-less VDB.
> Headless VDB
> SVDB (Source VDB)
> PVDB (Physical VDB)
>
> --
> Larry O'Leary <loleary(a)redhat.com>
> Red Hat, Inc.
>
>
>
> ------------------------------
>
> Message: 3
> Date: Fri, 28 Aug 2009 20:26:43 -0400
> From: Ken Johnson <kejohnso(a)redhat.com>
> Subject: Re: [teiid-users] [teiid-dev] What should we call this
> artifact?
> To: Ramesh Reddy <rareddy(a)redhat.com>
> Cc: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Message-ID: <4A9875C3.9080804(a)redhat.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> How about:
>
> skinnyvdb
> minivdb
> tinyvdb
>
> -Ken
>
>
> Ramesh Reddy wrote:
>
>> Hi,
>>
>> You all know that starting with 6.2 release we have a new feature called
>> "Connector Supplied Metadata", where connectors will be able to import
>> metadata dynamically at runtime for data integration purposes.
>>
>> To use this feature user need to write ".def" file, which we been
>> calling internally "vdbLess" mode, that does not sound correct. We are
>> building a VDB and all the access is to this no different than a regular
>> VDB, so it is a VDB just not built using the Designer!
>>
>> I have been calling it "DEF file based VDB", some names thrown out are
>> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
>> so that we all can refer it by the same name?
>>
>> Thanks
>>
>> Ramesh..
>>
>> _______________________________________________
>> teiid-dev mailing list
>> teiid-dev(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/teiid-dev
>>
>>
>
>
> --
> Ken Johnson
> Sr. Product Manager
> JBoss Middleware Business Unit
> Red Hat, Inc
> 978.392.3917
> ken.johnson(a)redhat.com
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Fri, 28 Aug 2009 17:06:10 -0400 (EDT)
> From: Ted Jones <tejones(a)redhat.com>
> Subject: Re: [teiid-users] [teiid-dev] What should we call this
> artifact?
> To: Ramesh Reddy <rareddy(a)redhat.com>
> Cc: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Message-ID:
> <1492195497.633701251493570758.JavaMail.root(a)zmail01.collab.prod.int.phx2.redhat.com>
>
> Content-Type: text/plain; charset=utf-8
>
> How about:
>
> "Def-driven VDB" or "d-vdb"?
>
> Dynamic VDB seems good too.
>
>
> Ted
>
> ----- Original Message -----
> From: "Ramesh Reddy" <rareddy(a)redhat.com>
> To: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Sent: Friday, August 28, 2009 3:05:56 PM GMT -06:00 US/Canada Central
> Subject: [teiid-dev] What should we call this artifact?
>
> Hi,
>
> You all know that starting with 6.2 release we have a new feature called
> "Connector Supplied Metadata", where connectors will be able to import
> metadata dynamically at runtime for data integration purposes.
>
> To use this feature user need to write ".def" file, which we been
> calling internally "vdbLess" mode, that does not sound correct. We are
> building a VDB and all the access is to this no different than a regular
> VDB, so it is a VDB just not built using the Designer!
>
> I have been calling it "DEF file based VDB", some names thrown out are
> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
> so that we all can refer it by the same name?
>
> Thanks
>
> Ramesh..
>
> _______________________________________________
> teiid-dev mailing list
> teiid-dev(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/teiid-dev
>
>
> ------------------------------
>
> Message: 5
> Date: Sat, 29 Aug 2009 07:02:37 +0000
> From: Rama Kishore Yelamanchilli <rkishore999(a)hotmail.com>
> Subject: Re: [teiid-users] [teiid-dev] What should we call this
> artifact?
> To: <tejones(a)redhat.com>, <rareddy(a)redhat.com>
> Cc: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
> Message-ID: <BAY121-W21007AC98F4EE2BB825A38E5F40(a)phx.gbl>
> Content-Type: text/plain; charset="windows-1252"
>
>
>
> I don't know whether I can participate in this? If I can, I liked dvdb (dynamic vdb) and would like to have extension as .dvdb instead of .def which will later be converted to .vdb file at runtime.
>
>
>
> Regards
>
>
>
> Ram
>
>
>
>
>> Date: Fri, 28 Aug 2009 17:06:10 -0400
>> From: tejones(a)redhat.com
>> To: rareddy(a)redhat.com
>> CC: teiid-users(a)lists.jboss.org; teiid-dev(a)lists.jboss.org
>> Subject: Re: [teiid-dev] What should we call this artifact?
>>
>> How about:
>>
>> "Def-driven VDB" or "d-vdb"?
>>
>> Dynamic VDB seems good too.
>>
>>
>> Ted
>>
>> ----- Original Message -----
>> From: "Ramesh Reddy" <rareddy(a)redhat.com>
>> To: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
>> Sent: Friday, August 28, 2009 3:05:56 PM GMT -06:00 US/Canada Central
>> Subject: [teiid-dev] What should we call this artifact?
>>
>> Hi,
>>
>> You all know that starting with 6.2 release we have a new feature called
>> "Connector Supplied Metadata", where connectors will be able to import
>> metadata dynamically at runtime for data integration purposes.
>>
>> To use this feature user need to write ".def" file, which we been
>> calling internally "vdbLess" mode, that does not sound correct. We are
>> building a VDB and all the access is to this no different than a regular
>> VDB, so it is a VDB just not built using the Designer!
>>
>> I have been calling it "DEF file based VDB", some names thrown out are
>> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
>> so that we all can refer it by the same name?
>>
>> Thanks
>>
>> Ramesh..
>>
>> _______________________________________________
>> teiid-dev mailing list
>> teiid-dev(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/teiid-dev
>> _______________________________________________
>> teiid-dev mailing list
>> teiid-dev(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/teiid-dev
>>
>
> _________________________________________________________________
> Log on to MSN India for a lowdown on what?s hot in the world today
> http://in.msn.com
>
15 years, 4 months
Re: [teiid-users] [teiid-dev] What should we call this artifact?
by Ted Jones
How about:
"Def-driven VDB" or "d-vdb"?
Dynamic VDB seems good too.
Ted
----- Original Message -----
From: "Ramesh Reddy" <rareddy(a)redhat.com>
To: teiid-users(a)lists.jboss.org, teiid-dev(a)lists.jboss.org
Sent: Friday, August 28, 2009 3:05:56 PM GMT -06:00 US/Canada Central
Subject: [teiid-dev] What should we call this artifact?
Hi,
You all know that starting with 6.2 release we have a new feature called
"Connector Supplied Metadata", where connectors will be able to import
metadata dynamically at runtime for data integration purposes.
To use this feature user need to write ".def" file, which we been
calling internally "vdbLess" mode, that does not sound correct. We are
building a VDB and all the access is to this no different than a regular
VDB, so it is a VDB just not built using the Designer!
I have been calling it "DEF file based VDB", some names thrown out are
"dynamic vdb" or "direct vdb" etc. Can you suggest a good name for this,
so that we all can refer it by the same name?
Thanks
Ramesh..
_______________________________________________
teiid-dev mailing list
teiid-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/teiid-dev
15 years, 4 months
Re: [teiid-users] What should we call this artifact?
by John Doyle
I like dynamic vdb more than any other name tossed out so far. But I'll toss some out as well.
thin VDB
instant VDB
fast VDB
rapid VDB
~jd
----- "Ramesh Reddy" <rareddy(a)redhat.com> wrote:
> Hi,
>
> You all know that starting with 6.2 release we have a new feature
> called
> "Connector Supplied Metadata", where connectors will be able to
> import
> metadata dynamically at runtime for data integration purposes.
>
> To use this feature user need to write ".def" file, which we been
> calling internally "vdbLess" mode, that does not sound correct. We
> are
> building a VDB and all the access is to this no different than a
> regular
> VDB, so it is a VDB just not built using the Designer!
>
> I have been calling it "DEF file based VDB", some names thrown out
> are
> "dynamic vdb" or "direct vdb" etc. Can you suggest a good name for
> this,
> so that we all can refer it by the same name?
>
> Thanks
>
> Ramesh..
>
> _______________________________________________
> teiid-users mailing list
> teiid-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/teiid-users
15 years, 4 months
Understanding Connectors
by Torsten Schlabach
Hi all!
In the Designer perspective, first of all, I had some problems defining
a connector at all, until I found out that I will only see the
connectors which I have defined after I disable the "Hide connector
types" button.
I then defined my connector and dragged my model into it. That part
seems to be fine.
Just when I go to my VDB definition, on the "Execute" page, I need to
assign a connector in order for my VDB to be able to execute.
But I just cannot select my pre-defined connector there as the "Loaded
Connectors" tab is empty. I can define a new connector there, using the
same properties as my existing connector, but just using a different
name, and it will work fine.
Is there anything I don't quite understand or did I find some kind of bug?
Regards,
Torsten
15 years, 4 months
Simple example (reading a CSV file) needed
by Torsten Schlabach
Dear list!
I am currently trying to make sense of Teiid. I am trying to follow the Quick Start Example at:
http://docs.jboss.org/teiid/6.1/quick-start-guide/en-US/html/
While it's clear to me how this is supposed to work, I fail to put it into practice. I am looking at:
http://docs.jboss.org/teiid/6.1/quick-start-guide/en-US/html/example-expl...
Where in the Designer would I have to create those two files? Looking at the examples in the designer, I cannot spot any files which look like that.
Then actually, on the next page:
http://docs.jboss.org/teiid/6.1/quick-start-guide/en-US/html/designer.html
> 1. Switch to the Designer Perspective, by opening the "Window->Open
> Perspective.."
Ok, no problem!
> 2. Create new "Model Project", using "File->New->ModelProject"
Also no problem!
> 3. Import the Derby database's "account" schema into a source model.
How would I do that???
> 4. Manually create the source model for the "Text File". Create a Table
called "Price" with "stock" and "price" as column names
Again, how would I do that? I don't see anything called "Source Model" in my designer.
Also for the remaining steps, I just cannot find the elements in my designer which the steps talk about.
Any hints are welcome!
Regards,
Torsten
15 years, 4 months
Logging warnings with Teiid 6.2-M2
by Ramesh Reddy
Here is the new log4J file, that fixes the issues. Will fix the
configuration for RC1.
Ramesh..
On Tue, 2009-08-18 at 10:45 -0400, Paul Nittel wrote:
> Hiya!
>
> I downloaded M2 and am running a test using a simple JDBC class. When i starts I get this:
>
> log4j:WARN Continuable parsing error 68 and column 13
> log4j:WARN The content of element type "logger" must match "(level?,appender-ref*)".
> log4j:WARN Continuable parsing error 72 and column 13
> log4j:WARN The content of element type "logger" must match "(level?,appender-ref*)".
> log4j:WARN Continuable parsing error 76 and column 13
> log4j:WARN The content of element type "logger" must match "(level?,appender-ref*)".
> log4j:WARN Continuable parsing error 80 and column 13
> log4j:WARN The content of element type "logger" must match "(level?,appender-ref*)".
>
>
15 years, 4 months
Teiid 6.2 M2 and more
by Steven Hawkins
Hello all,
6.2 M2 status: There were a lot of changes over the weekend, there hasn't been much feedback on M1, and the pass-through feature has not been added (again not much feedback), so M2 will be delayed until 8/13. In the meantime let's keep assessing current 6.2 defects.
6.3 planning: Start capturing your thoughts (vote, send messages to the dev-list, etc.) for what is appropriate in the next release. This will again be targeted at a 12 week cycle.
7.0 planning: I am in favor of ending the 6.x line at 6.3 and start on major changes in a longer 7.0 cycle. This would also probably necessitate a 6.3.1 to address defects while 7.0 is still solidifying. As a start the current 6.x items will be rolled into the 7.x bucket and a 7.0 bucket will be created. Community direction will be essential for this release.
Steve
15 years, 4 months
