Hi Duncan,
On Oct 29, 2012, at 9:24 AM, Duncan Temple Lang wrote:
>
> Hi Ben
>
> Can you tell us the slightly bigger picture, please?
> Do you want to create a single similar node entirely in isolation
> or do you want to create it as part of an XML tree/document?
> Who will be reading the resulting XML.
>
> You can use a parent node
>
> top = newXMLNode("storms", namespaceDefinitions = c(weather =
> "http://my.weather.com/events";))
>
> Then
>
> newXMLNode("storm", "ripsnorter", namespace = "weather",
> attrs = c(type = "hurrican", name = "Sandy"),
> parent = top )
>
>
> That gives you
>
> <weather:storm type="hurrican" name="Sandy">ripsnorter</weather:storm>
>
> So now what are you going to do with that node?
>
> The namespace prefix is local to a document, chosen by the author of that XML
> document.
> The namespace URI is the global key that authors and consumers must agree
> upon.
> While your database may use "udf", you may chose a different prefix or even
> the default
> prefix to correspond to that same URI. So each document must explicitly
> declare the
> prefix = URI mapping for it to be understood.
>
Ah! The parent - of course! This makes perfect sense (and says so in the
docs). I think I was focussing on building up the hierarchy from the
inside-to-the-outside. That's obviously backwards (of course it's obvious now
that I have gone public!)
I must have I missed the boat when I read the documentation for the namespace
argument to newXMLNode() ...
"If this is a character vector with a) one element and b) with an empty names
attribute and c) whose value does not start with http:/ or ftp:/, then it is
assumed that the value is a namespace prefix for a namespace defined in an
ancestor node. To be able to resolve this prefix to a namespace definition,
parent must be specified so that we can traverse the chain of ancestor nodes. "
I wonder if that last sentence might be amended to read, "To be able to
resolve this prefix to a namespace definition, parent must be specified so that
we can traverse the chain of ancestor nodes _to where the namespace is
defined._" I realize that this addition repeats the meaning of the previous
sentence, but repetition never hurts boneheads like me.
The bigger picture is that I am using this with a RESTful system which has been
very kindly to this neophyte. The parent tree, of which these user defined
fields are children, will be POSTed to the system (using RCurl).
I have the parent just as you have shown 'top' above, but I had not included
the namespace definitions that any of its children might need. I have added
the namespace definitions needed to the parent definition and all is well.
Thanks, once again, for making the XML package available. It is incredibly
useful.
Cheers,
Ben
> D.
>
>
> On 10/29/12 5:54 AM, Ben Tupper wrote:
>> Hello,
>>
>> I am working with a database system from which I can retrieve these kinds of
>> user defined fields formed as XML ...
>>
>> <udf:field unit="uM" type="Numeric" name="facs.Stain final
>> concentration">5</udf:field>
>>
>> You can see in the above example that "field" is defined in the namespace
>> "udf", but that the "udf" namespace is not defined along with the attributes
>> of the node. That is, 'xmlns:udf = "http://blah.blah.com/blah";' doesn't
>> appear.
>>
>> I would like to create a similar node from scratch, but I can't seem to
>> define the node with a namespace without providing the namespace definition.
>>
>>
>> library(XML)
>>
>> node1 <- newXMLNode("storm", "ripsnorter",
>> namespace = "weather",
>> namespaceDefinitions = c(weather = "http://my.weather.com/events";),
>> attrs = c(type = "hurricane", name = "Sandy"))
>> node1
>>
>> # this returns the new node with the namespace prefix (which I want)
>> # and the definition (which I don't want)
>>
>> # <weather:storm xmlns:weather="http://my.weather.com/events";
>> type="hurricane" name="Sandy">ripsnorter</weather:storm>
>>
>>
>> node2 <- newXMLNode("storm", "ripsnorter",
>> namespace = "weather",
>> attrs = c(type = "hurricane", name = "Sandy"),
>> suppressNamespaceWarning = TRUE)
>> node2
>>
>> # produces the node without the namespace prefix and without the definition
>>
>> # <storm type="hurricane" name="Sandy">ripsnorter</storm>
>>
>> Is there some way to create a node with a namespace prefix but without
>> embedding the namespace definition along with the attributes?
>>
>> Thanks!
>> Ben
>>
>> Ben Tupper
>> Bigelow Laboratory for Ocean Sciences
>> 180 McKown Point Rd. P.O. Box 475
>> West Boothbay Harbor, Maine 04575-0475
>> http://www.bigelow.org
>>
>>
>>> sessionInfo()
>> R version 2.15.0 (2012-03-30)
>> Platform: i386-apple-darwin9.8.0/i386 (32-bit)
>>
>> locale:
>> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>>
>> attached base packages:
>> [1] stats graphics grDevices utils datasets methods base
>>
>> other attached packages:
>> [1] tripack_1.3-4 RColorBrewer_1.0-5 Biostrings_2.24.1 IRanges_1.14.2
>> BiocGenerics_0.2.0 RCurl_1.91-1
>> [7] bitops_1.0-4.1 XML_3.9-4
>>
>> loaded via a namespace (and not attached):
>> [1] stats4_2.15.0 tools_2.15.0
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>>
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Ben Tupper
Bigelow Laboratory for Ocean Sciences
180 McKown Point Rd. P.O. Box 475
West Boothbay Harbor, Maine 04575-0475
http://www.bigelow.org
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
