Mapping by similar name

Learn about the rules that apply when you select the Create transforms when the names of inputs and outputs are more similar than option in Auto map (automap).

About this task

When you select Create transforms when the names of inputs and outputs are more similar than, the following rules apply:
  1. Fixed value outputs and mapped outputs are excluded in name matching.
  2. The similarity test is done using the name of an element, a database column, or an attribute regardless of its type. If an input and an output have the same name, it is a match, regardless of the kind of, and XSD type of, the input and output. An element, a database column, or an attribute can all form a match if their names are the same.
  3. The similarity test applies in the same way to case sensitivity and alphanumeric characters as for Mapping by same name.
  4. Namespace or namespace prefixes do not participate in the similarity test. XML namespaces are excluded from name matching. Therefore, abc:something and xyz:something are considered the same, as are {http://www.abc.com}:something and {http://www.xyz.com}:something.
  5. When multiple inputs have the same name as one output, one mapping is created. However, if you have multiple inputs with the same name as one output and you choose to map by the same name (or similar name) and to match descendants, an attempt is made first to match by path and name. If a match is found, one transform is made, and no further matches are considered.
  6. When you select Map all simple descendants of the selected elements, the following steps are taken to match names.
    Inputs and outputs matched in a previous step are ignored in later steps:
    1. Compare the relative path and item name of the selected input and output
    2. Compare the item names without relative path
    3. Compare similar item names without relative path

    When you select the Map the immediate children of the selected elements option, the only step taken to match names is to compare similar item names without the relative path.

  7. You can select the similarity threshold for two words to be considered similar.
  8. You cannot use any other similarity algorithm.