Jump to: navigation, search

Difference between revisions of "SMILA/Documentation/Bundle org.eclipse.smila.processing.pipelets.xmlprocessing"

m (Configuration)
(5 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
== General ==
 
== General ==
 +
 +
All pipelets in this bundle support the configurable error handling as described in [[SMILA/Development_Guidelines/How_to_write_a_Pipelet#Implementation]]. When used in jobmanager workflows, records causing errors are dropped.
  
 
''' Read Type '''
 
''' Read Type '''
Line 24: Line 26:
 
|''outputName''||String||runtime||The name of the output attachment or the path to the output attribute (store result as literals of attribute).
 
|''outputName''||String||runtime||The name of the output attachment or the path to the output attribute (store result as literals of attribute).
 
|-
 
|-
|''xslFile''||String||init||The name (with relative or absolute path) of the XSL file to be used for transformation.
+
|''xslFile''||String||runtime||The name (with relative or absolute path) of the XSL file to be used for transformation.
 +
|-
 +
|''parameters''||Map or Boolean||runtime||Either a map of XSL parameters or a boolean that indicates to add all attributes as XSL parameters.
 
|}
 
|}
  
Line 224: Line 228:
  
 
On each created record the attribute <tt>__isXmlSnippet</tt>=<tt>true</tt> is set to true. Incoming records with this attribute set are not splitted again, but returned as the Pipelet result. This way it's possible to add further processing steps that should be done on the splitted records to the same pipeline that does the splitting. See [https://dev.eclipse.org/svnroot/rt/org.eclipse.smila/trunk/core/SMILA.application/configuration/org.eclipse.smila.processing.bpel/pipelines/XmlSplitAndAddPipeline.bpel XmlSplitAndAddPipeline.bpel] for an example.
 
On each created record the attribute <tt>__isXmlSnippet</tt>=<tt>true</tt> is set to true. Incoming records with this attribute set are not splitted again, but returned as the Pipelet result. This way it's possible to add further processing steps that should be done on the splitted records to the same pipeline that does the splitting. See [https://dev.eclipse.org/svnroot/rt/org.eclipse.smila/trunk/core/SMILA.application/configuration/org.eclipse.smila.processing.bpel/pipelines/XmlSplitAndAddPipeline.bpel XmlSplitAndAddPipeline.bpel] for an example.
 +
 +
Namespaces visible in its scope are added to each snippet, thus the result is a valid XML document.
  
 
=== Configuration ===
 
=== Configuration ===
Line 234: Line 240:
 
|''outputType''||String : ATTACHMENT, ATTRIBUTE||runtime||Defines whether the XML snippet should be stored in an attachment or in an attribute of the newly created record.  
 
|''outputType''||String : ATTACHMENT, ATTRIBUTE||runtime||Defines whether the XML snippet should be stored in an attachment or in an attribute of the newly created record.  
 
|-
 
|-
|''inputName''||String||runtime||The name of the input attachment or the path to the input attribute.
+
|''inputName''||String||runtime||The name of the input attachment or the path to the input attribute containing a path to an external data source, e.g. an  xml file name
 
|-
 
|-
 
|''outputName''||String||runtime||The name of the output attachment or the path to the output attribute (store result as literals of attribute).  
 
|''outputName''||String||runtime||The name of the output attachment or the path to the output attribute (store result as literals of attribute).  
Line 240: Line 246:
 
|''beginTagName''||String||runtime||The name of the tag to start the XML snippet with.
 
|''beginTagName''||String||runtime||The name of the tag to start the XML snippet with.
 
|-
 
|-
|''isBeginClosingTag''||Boolean||runtime||A boolean flag defining whether ''beginTagName'' is a closing tag (true) or not (false).
+
|''beginTagNamespace''||String||runtime||The namespace of the start tag. Namespaces are not checked, if not given (in that case any namespace matches).
 
|-
 
|-
|''endTagName''||String||runtime||The name of the tag to end the xml snippet with.
+
|''endTagName''||String||runtime||The name of the tag to end the xml snippet with, defaults to the value of ''beginTagName''
 
|-
 
|-
|''isEndClosingTag''||Boolean||runtime||A boolean flag defining whether ''endTagName'' is a closing tag (true) or not (false).
+
|''endTagNamespace''||String||runtime||The namespace of the end tag, defaults to the value of ''beginTagNamespace''.
 
|-
 
|-
 
|''keyTagName''||String||runtime||The name of the tag used to create a record ID.
 
|''keyTagName''||String||runtime||The name of the tag used to create a record ID.
Line 267: Line 273:
 
     <rec:Val key="outputName">xmlOut</rec:Val>
 
     <rec:Val key="outputName">xmlOut</rec:Val>
 
     <rec:Val key="beginTagName">document</rec:Val>
 
     <rec:Val key="beginTagName">document</rec:Val>
    <rec:Val key="isBeginClosingTag">false</rec:Val>
 
    <rec:Val key="endTagName">document</rec:Val>
 
    <rec:Val key="isEndClosingTag">true</rec:Val>
 
 
     <rec:Val key="keyTagName">docId</rec:Val>
 
     <rec:Val key="keyTagName">docId</rec:Val>
 
     <rec:Val key="idSeparator">#</rec:Val>
 
     <rec:Val key="idSeparator">#</rec:Val>

Revision as of 04:14, 27 March 2012

General

All pipelets in this bundle support the configurable error handling as described in SMILA/Development_Guidelines/How_to_write_a_Pipelet#Implementation. When used in jobmanager workflows, records causing errors are dropped.

Read Type

  • runtime: Parameters are read when processing records. Parameter value can be set per Record.
  • init: Parameters are read once from Pipelet configuration when initializing the Pipelet. Parameter value can not be overwritten in Record.

org.eclipse.smila.processing.pipelets.xmlprocessing.XslTransformationPipelet

Description

This pipelet performs an XSL transformation on an attribute or attachment value and stores the transformed document in an attribute or attachment.

Configuration

Property Data Type Read Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the transformed output should be stored in an attachment or in an attribute of the record.
inputName String runtime The name of the input attachment or the path to the input attribute (process literals of attribute).
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
xslFile String runtime The name (with relative or absolute path) of the XSL file to be used for transformation.
parameters Map or Boolean runtime Either a map of XSL parameters or a boolean that indicates to add all attributes as XSL parameters.

Example

Pipelet configuration for XslTransformationPipelet

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Val key="xslFile">./configuration/data/author.xsl<rec:Val>
</proc:configuration>

org.eclipse.smila.processing.pipelets.xmlprocessing.XPathExtractorPipelet

Description

This pipelet extracts elements selected by XPath, converts them to appropriate data types (Boolean, Double, String), and stores the transformed value in an attribute or attachment.

Configuration

Property Data Type Read Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the transformed output should be stored in an attachment or in an attribute of the record.
inputName String runtime The name of the input attachment or the path to the input attribute (process literals of attribute).
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
xpath String runtime The XPath expression to be evaluated.
separator String runtime The optional separator.
namespace String runtime The optional XML namespace.

Example

Pipelet configuration for XPathExtractorPipelet

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Val key="xpath">author/email<rec:Val>
    <rec:Val key="separator"><rec:Val>
    <rec:Val key="namespace"><rec:Val>
</proc:configuration>

org.eclipse.smila.processing.pipelets.xmlprocessing.XPathFilterPipelet

Description

This pipelet filters elements by XPath expressions (either using include or exclude mode) and stores the filtered elements as a new document in an attribute or attachment.

Configuration

Property Data Type Read Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the transformed output should be stored in an attachment or in an attribute of the record.
inputName String runtime The name of the input attachment or the path to the input attribute (process literals of attribute).
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
xpath String runtime The XPath expressions to be evaluated (multi-valued property).
filterMode String : INCLUDE, EXCLUDE runtime The filter mode, defining whether to include or exclude the elements matched by the XPath expressions.
namespace String runtime The optional XML namespace.

Examples

Pipelet configuration for XPathFilterPipelet with multi-valued xpath

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Seq key="xpath">
        <rec:Val>author/email<rec:Val>
        <rec:Val>author/name<rec:Val>
    </rec:Seq>
    <rec:Val key="filterMode">EXCLUDE<rec:Val>
    <rec:Val key="seperator"><rec:Val>
    <rec:Val key="namespace"><rec:Val>
</proc:configuration>

Pipelet configuration for XPathFilterPipelet with single-valued xpath

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Val key="xpath">author/email<rec:Val>
    <rec:Val key="filterMode">EXCLUDE<rec:Val>
    <rec:Val key="seperator"><rec:Val>
    <rec:Val key="namespace"><rec:Val>
</proc:configuration>

org.eclipse.smila.processing.pipelets.xmlprocessing.RemoveElementFromXMLPipelet

Description

This pipelet removes a selected element from an XML document and stores the manipulated document in an attribute or attachment.

Configuration

Property Data Type Read Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the transformed output should be stored in an attachment or in an attribute of the record.
inputName String runtime The name of the input attachment or the path to the input attribute (process literals of attribute).
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
elementId String runtime The ID of the XML element to be removed.

Example

Pipelet configuration for RemoveElementFromXMLPipelet

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Val key="elementId">title<rec:Val>
</proc:configuration>

org.eclipse.smila.processing.pipelets.xmlprocessing.TidyPipelet

Description

This pipelet performs a Tidy transformation on an attribute or attachment value and stores the result in an attribute or attachment.

Configuration

Property Data Type Read Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the transformed output should be stored in an attachment or in an attribute of the record.
inputName String runtime The name of the input attachment or the path to the input attribute (process literals of attribute).
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
tidyFile String init The name (with relative or absolute path) of the Tidy configuration file to be used by the transformation.

Example

Pipelet configuration for TidyPipelet

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE<rec:Val>
    <rec:Val key="outputType">ATTRIBUTE<rec:Val>
    <rec:Val key="inputName">xmlIn<rec:Val>
    <rec:Val key="outputName">xmlOut<rec:Val>
    <rec:Val key="tidyFile">./configuration/data/tidy_config.txt<rec:Val>
</proc:configuration>

org.eclipse.smila.processing.pipelets.xmlprocessing.XmlSplitterPipelet

Description

This pipelet splits an XML stream into multiple XML snippets. For each snippet a new record is created where the XML snippet is stored in either an attribute or attachment. The created records are not returned as a PipeletResult (this is just the same as the incoming RecordIds) but are directly sent to the ConnectivityManager and are routed once more to the queue.

On each created record the attribute __isXmlSnippet=true is set to true. Incoming records with this attribute set are not splitted again, but returned as the Pipelet result. This way it's possible to add further processing steps that should be done on the splitted records to the same pipeline that does the splitting. See XmlSplitAndAddPipeline.bpel for an example.

Namespaces visible in its scope are added to each snippet, thus the result is a valid XML document.

Configuration

Property Type Description
inputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML input is found in an attachment or in an attribute of the record. An input attribute is not interpreted as to contain XML content itself but rather as a file path or an URL to the XML document.
outputType String : ATTACHMENT, ATTRIBUTE runtime Defines whether the XML snippet should be stored in an attachment or in an attribute of the newly created record.
inputName String runtime The name of the input attachment or the path to the input attribute containing a path to an external data source, e.g. an xml file name
outputName String runtime The name of the output attachment or the path to the output attribute (store result as literals of attribute).
beginTagName String runtime The name of the tag to start the XML snippet with.
beginTagNamespace String runtime The namespace of the start tag. Namespaces are not checked, if not given (in that case any namespace matches).
endTagName String runtime The name of the tag to end the xml snippet with, defaults to the value of beginTagName
endTagNamespace String runtime The namespace of the end tag, defaults to the value of beginTagNamespace.
keyTagName String runtime The name of the tag used to create a record ID.
maxBufferSize Integer runtime The maximum size of the internal record buffer (optional, default is 20).
idSeparator String runtime The separator used to create the record IDs of the split records (optional, default is "#").
xmlSnippetJobName String runtime The JobManager job name to submit the split records to. It must be running when the pipelet is executed.

The first four attributes can be set only in the pipelet configuration. All other attributes can seperately customized for each single record by setting them as values of the the _parameters map in the record.

Example

Pipelet configuration for XmlSplitterPipelet

<proc:configuration>
    <rec:Val key="inputType">ATTRIBUTE</rec:Val>
    <rec:Val key="outputType">ATTRIBUTE</rec:Val>
    <rec:Val key="inputName">xmlIn</rec:Val>
    <rec:Val key="outputName">xmlOut</rec:Val>
    <rec:Val key="beginTagName">document</rec:Val>
    <rec:Val key="keyTagName">docId</rec:Val>
    <rec:Val key="idSeparator">#</rec:Val>
    <rec:Val key="xmlSnippetJobName">indexUpdateXml</rec:Val>
</proc:configuration>


The above configuration would split this XML format:

<sampleCollection>
    ...
    <document>
        <docId>4711</docId>
        <title>Some title</title>
        ...
        <text>Some text</text>
    </document>
    <document>
        <docId>0815</docId>  
        ...
    </document>
    ...
</sampleCollection>

into XML snippets like this one:

<document>
    <docId>4711</docId>
    <title>Some title</title>
    ...
    <text>Some text</text>
</document>

And for each snippet a record would be created and submitted to a job run of job "indexUpdateXml".

<Record xmlns="http://www.eclipse.org/smila/record" version="2.0">
  <Val key="_recordid">xmlsplitter:someBigXmlfile.xml#4711</Val>
  <Val key="_source">xmlsplitter</Val>
  <Val key="__isXmlSnippet">true</Val>
  <Val key="xmlOut">
         <document>
            <docId>4711</docId>
            <title>Some title</title>
            ...
            <text>Some text</text>
        </document>   
  </Val>
</Record>

To use the XmlSplitAndAddPipeline.bpel via the predefined asynchronous workflow indexUpdateXml you should create a job definition that sets the "xmlSnippetJobName" to the job's own name:

{
  "name": "indexUpdateXmlJob",
  "workflow": "indexUpdateXml",
  "parameters": {
    "tempStore": "xmlbulks",
    "xmlSnippetJobName": "indexUpdateXmlJob"
  }
}