source: trunk/gli/classes/xml/import.xml@ 4293

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE Info [
  <!ELEMENT Info      (Name, Desc, Arguments)>
  <!ELEMENT Arguments (Option*)>
  <!ELEMENT Option    (Name, Desc, Type, Required, Default?, List?)>
  <!ELEMENT Name      (#PCDATA)>
  <!ELEMENT Desc      (#PCDATA)>
  <!ELEMENT Type      (#PCDATA)>
  <!ELEMENT Required  (#PCDATA)>
  <!ELEMENT Default   (#PCDATA)>
  <!ELEMENT List      (Value*)>
  <!ELEMENT Value     (Name, Desc?)>
]>

<Info>
  <Name>import.pl</Name>
  <Desc>PERL script used to import files into a GML format ready for building.</Desc>
  <Arguments>
    <Option>
      <Name>archivedir</Name>
      <Desc>Where the converted material ends up</Desc>
      <Type>string</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>collectdir</Name>
      <Desc>Collection directory (defaults to /research/jmt12/gsdl/collect)</Desc>
      <Type>string</Type>
      <Required>no</Required>
      <Default>/research/jmt12/gsdl/collect</Default>
    </Option>
    <Option>
      <Name>debug</Name>
      <Desc>Print imported text to STDOUT</Desc>
      <Type>flag</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>faillog</Name>
      <Desc>Fail log filename. This log receives the filenames of any files which fail to be processed (defaults to &lt;collectdir&gt;/colname/etc/fail.log)</Desc>
      <Type>string</Type>
      <Required>no</Required>
      <Default>&lt;collectdir&gt;/colname/etc/fail.log</Default>
    </Option>
    <Option>
      <Name>groupsize</Name>
      <Desc>Number of import documents to group into one XML file</Desc>
      <Type>int</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>gzip</Name>
      <Desc>Use gzip to compress resulting xml documents (don't forget to include ZIPPlug in your plugin list when building from compressed documents)</Desc>
      <Type>flag</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>importdir</Name>
      <Desc>Where the original material lives</Desc>
      <Type>string</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>keepold</Name>
      <Desc>Will not destroy the current contents of the archives directory (the default)</Desc>
      <Type>flag</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>maxdocs</Name>
      <Desc>Maximum number of documents to import</Desc>
      <Type>int</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>OIDtype</Name>
      <Desc>The method to use when generating unique identifiers for each document. "hash" (the default) hashes the contents of the file and so will be the same every time the collection is imported. "incremental" is a simple document count and so will be significantly faster than "hash". It is not guaranteed to always assign the same identifier to a given document though and does not allow further documents to be added to existing xml archives</Desc>
      <Type>enum</Type>
      <Required>no</Required>
      <List>
        <Value>
          <Name>hash</Name>
        </Value>
        <Value>
          <Name>incremental</Name>
        </Value>
      </List>
    </Option>
    <Option>
      <Name>out</Name>
      <Desc>Filename or handle to print output status to. The default is STDERR</Desc>
      <Type>string</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>removeold</Name>
      <Desc>Will remove the old contents of the archives directory -- use with care</Desc>
      <Type>flag</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>sortmeta</Name>
      <Desc>Sort documents alphabetically by metadata for building. This will be disabled if groupsize > 1</Desc>
      <Type>string</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>statsfile</Name>
      <Desc>Filename or handle to print import statistics to. The default is STDERR</Desc>
      <Type>string</Type>
      <Required>no</Required>
    </Option>
    <Option>
      <Name>verbosity</Name>
      <Desc>A number between zero and three that controls how much information about the process is printed to the standard error stream; 0 gives a little, 3 gives lots</Desc>
      <Type>int</Type>
      <Required>no</Required>
    </Option>
  </Arguments>
</Info>