Syntax

FACTORY_DEF <ReaderKeyword> SchemaMappingFactory
[FACTORY_NAME <factory name>]
[INPUT FEATURE_TYPE <factory type>
   [<attribute name> <attribute value>]*
   [<feature function>]*]*
[DB_TYPE <format name>]
[DB_DATASET <dataset>]
[DB_TABLE <table name>]
[DB_DIRECTIVES [<parameter> <value>]*]
[FILTER_COLUMNS [<attribute name column> <attribute value column>]*]
[INDEX_MAPPING_COLUMNS <source attribute value column>
   <destination attribute value column>
   <source attribute name> <destination attribute name>]
[ATTR_MAPPING_COLUMNS [<source attribute name column>
   <destination attribute name column>
   <default value column>]*]
[ROW_ID_ATTR <attribute name>]
[OUTPUT (MAPPED|UNMAPPED) FEATURE_TYPE <feature type>
   [<attribute name> <attribute value>]*
   [<feature function>]*]*

Overview

   This factory is used to modify the schema (data model) of incoming features,
   based on an external lookup table (schema mapping table).

   The schema mapping table defines:

      A series of conditions that are to be met (specified by the FILTER_COLUMNS
      clause)

      A series of actions that will be executed when conditions are met
      (specified by the INDEX_MAPPING_COLUMNS and ATTR_MAPPING_COLUMNS clauses)

   The factory processes each incoming feature in the following manner:

      1. The feature is matched against each row of the schema mapping table,
      where rows in which the filter does not match the feature are removed.

      2. If an index mapping is not defined, skip this step. For each row, those
      with index values matching that of the feature are processed. The index
      attribute is mapped to the specified target value, and all attribute maps
      will be executed.

      3. For each row, all attribute maps are executed.

   The factory loads the schema mapping table by the directives prefixed by DB_.
   In the standard manner, the DB_TYPE clause specifies the format type, the
   DB_DATASET clause specifies the dataset location, the DB_TABLE clause
   specifies the table(s) or feature(s) to read from the dataset, and the
   DB_DIRECTIVES clause specifies format-related parameters.

   The FILTER_COLUMNS clause specifies the columns used to create the mapping
   condition. The columns should be specified in pairs, where the first
   (<attribute name column>) is the column containing the attribute to test, and
   the second (<attribute value column>) is the column containing the value to
   test against.

   If the attribute value is blank for a particular row, then one of two
   behaviours may occur. By default, the filter is ignored. On the other hand,
   if one wishes to match a blank attribute value, add the control character
   ASCII 16 (DLE), or '\x10', to the end of <attribute value column>.

   The INDEX_MAPPING_COLUMNS clause specifies an index mapping. For each
   incoming feature, an index mapping is performed if both:

      1. The feature matches the filter condition of the given row

      2. The feature's index source attribute value matches that of the given
      row

   An index mapping produces a new index attribute with the name specified by
   <destination attribute name> and the value specified by the column specified
   by <destination attribute value column>.

   A common use case for index mapping is the mapping of feature types. In this
   case, the INDEX_MAPPING_COLUMNS clause should be specified in the following
   manner:

      [INDEX_MAPPING_COLUMNS <source feature type column>
         <destination feature type column> fme_feature_type
         fme_feature_type]

   The ATTR_MAPPING_COLUMNS clause specifies the columns used to create an
   attribute map. Depending on which values are present in a particular row, the
   attribute map may be categorized differently.

      Source Attribute:      yes
      Destination Attribute: yes
      Default Value:         no
      Type:                  Attribute Map

         Create a new attribute with the name from the row and the value from
         the source attribute.

      Source Attribute:      yes
      Destination Attribute: yes
      Default Value:         yes
      Type:                  Attribute Map

         Create a new attribute with the name from the row and the value from
         the source attribute. If the source attribute is missing, the
         default value from the row is used.

      Source Attribute:      no
      Destination Attribute: yes
      Default Value:         yes
      Type:                  New Attribute

         Create a new attribute with the name and value from the row.

   After an attribute map has been applied, the source schema, by default, is
   retained. If one wishes to remove the source schema, add the control
   character ASCII 16 (DLE), or '\x10', to the end of <default value column>.

   The ROW_ID_ATTR clause specifies an attribute to report the schema mapping
   rows that were applied to a feature. The rows are listed by row ID, separated
   by commas.

Output Tags

   The SchemaMappingFactory supports the following output tags.

   MAPPED

      The output features that have at least one schema mapping action applied.

   UNMAPPED

      The output features that have no schema mapping action applied.