The supported metadata for the document is the following:

  <root>

    <document id="doc001" name="my document">

      <data mime="application/x-indesign">pallino.indd</data>

      <meta>

        <value name="title">My title</value>

        <value name="date">2011-01-03</value>

        <value name="keyword">keyword1</value>

        <value name="keyword">keyword2</value>

        <value name="category">local</value>

        <value name="category">politics</value>

        <value name="summary">summary</value>

        <value name="authors">author1 author 2</value>

        <value name="instructions">instructions</value>

        <value name="credit">credit</value>

        <value name="orgSource">orgSource</value>

        <value name="location">location</value>

        <value name="city">city</value>

        <value name="state">state</value>

        <value name="country">country</value>

        <value name="countryCode">NL</value>

        <value name="position-lat">49.37</value>

        <value name="position-lon">12.94</value>

      </meta>

    <document>

  </root> 

The 'date' value goes into the 'eventDate' attribute.

All the 'keyword' values go in the 'keywords' attribute.

All the 'category' values go in the 'cats' attribute - and they must be the codes of existing category in the category set associated with the 'cats' attribute - if they are not the import will fail.

The 'position-lat' and 'position-lon' values go in the 'position' attribute, and they must be valid decimal latitude and longitude.

Instead that a single 'authors' values it is possible to specify one or more 'author' values that will be concatenated and placed in the 'authors' attribute.

All the other values go in the attribute with the same name.

String values that are too long for the corresponding attribute will be truncated, adding '...' at the end to indicate that a truncation took place.

All the metadata is optional. If no 'title' metadata value is specified the title of the imported document will be set to the specified document name ('my document' in the exaple above).

The metadata for stories is identical to that of documents, e.g.:

  <root>

    <story id="story0001" name="my story">

      <data>storyText.xml</data>

      <meta>

        <value name="title">story title</value>

        <value name="date">2011-01-04</value>

        <value name="keyword">story keyword1</value>

        <value name="keyword">story keyword2</value>

        <value name="category">local</value>

        <value name="category">politics</value>

        <value name="summary">new story summary</value>

        <value name="author">author1</value>

        <value name="author">author 2</value>

        <value name="instructions">story instructions</value>

        <value name="credit">story credit</value>

        <value name="orgSource">story orgSource</value>

        <value name="location">story location</value>

        <value name="city">city</value>

        <value name="state">state</value>

        <value name="country">country</value>

        <value name="countryCode">NZ</value>

        <value name="position-lat">-42.37</value>

        <value name="position-lon">173.94</value>

      </meta>

    </story>

  </root>

Everything said above concerning documents' metadata applies to stories' metadata (note in this example the use of two separate 'author' values).

For stories if there is no 'summary' metadata a summary is generated extracting the initial part of the text itself.

The metadata of images and videos is identical to that of stories and documents, e.g. (for an image)

  <root>

    <image id="image0001">

      <data>IPTC.jpg</data>

      <meta>

        <value name="title">image title</value>

        <value name="date">2011-01-03</value>

        <value name="keyword">image keyword1</value>

        <value name="keyword">image keyword2</value>

        <value name="category">local</value>

        <value name="category">politics</value>

        <value name="summary">image summary</value>

        <value name="author">author1</value>

        <value name="author">author 2</value>

        <value name="instructions">image instructions</value>

        <value name="credit">image credit</value>

        <value name="orgSource">image orgSource</value>

        <value name="location">image location</value>

        <value name="city">image city</value>

        <value name="state">image state</value>

        <value name="country">image country</value>

        <value name="countryCode">NZ</value>

        <value name="position-lat">-42.37</value>

        <value name="position-lon">173.94</value>

      </meta>

    </image>

  </root>

Again, the same remarks concerning the documents' metadata apply to the images' and videos' ones. The only difference is that if metdata values are not specified in the XML the workflow try to fill the corresponding attributes from the IPTC or XMP data embedded in the image or video file. See the Config\Global\GN4_Tark4_Common\xsl_ImportDocumentImage.xml and Config\Global\GN4_Tark4_Common\xsl_ImportDocumentImage.xml XLST for details of these mappings.

Finally, the metadata of audio is similar to that of other objects, e.g.

  <root>

    <audio id="audio001">

      <data>07 Get It Hot.mp3</data>

      <meta>

        <value name="kind">music</value>

        <value name="title">audio title</value>

        <value name="album">audio album</value>

        <value name="date">2011-01-02</value>

        <value name="keyword">audio keyword1</value>

        <value name="keyword">audio keyword2</value>

        <value name="category">local</value>

        <value name="category">politics</value>

        <value name="summary">new audio summary</value>

        <value name="author">author1</value>

        <value name="author">author 2</value>

        <value name="instructions">audio instructions</value>

        <value name="credit">audio credit</value>

        <value name="orgSource">audio orgSource</value>

        <value name="location">audio location</value>

        <value name="city">audio city</value>

        <value name="state">audio state</value>

        <value name="country">audio country</value>

        <value name="countryCode">NZ</value>

        <value name="position-lat">-42.37</value>

        <value name="position-lon">173.94</value>

      </meta>

    </audio>

  </root>

The difference is in the addition of the two values 'kind' and 'album' - mapped to the attribute with the same name of the audio object type. Note that the 'kind' attribute is an enumeration (currently with two values 'music', 'spken word'), so using values not in the enumeration will cause an error in import.

As for images and videos if metdata values are not specified in the XML the workflow try to fill the corresponding attributes from the IPTC or XMP data embedded in the audio file. See the Config\Global\GN4_Tark4_Common\xsl_ImportDocumentAudio.xml XLST for details of this mappings.

As noted the mapping of metadata values to object attributes and the extraction of metadata from XMP / IPTC data is done by external XSLT (Config\Global\GN4_Tark4_Common\xsl_ImportDocuments*.xml), so it is possible to customize them without touching the main workflow.

See also

Sources