Skip to content

ABAP Keyword Documentation →  ABAP - Reference →  Data Interfaces and Communication Interfaces →  ABAP and XML →  Class Libraries for XML →  sXML Library →  sXML Library - Parse 

sXML Library, Token-Based Parsing

In token-based parsing, the parser iterates across all nodes (tokens) in the tree structure of the XML data, one after the other. By default, the iterator passes across all subnode branches until the final node. The parser pauses after every iteration step on a node whose properties are available in the attributes of the XML reader. If the parser pauses on the node for an opened element, the list of the XML attributes there can be accessed.

Other versions: 7.31 | 7.40 | 7.54

Procedure (Principles)

An XML reader is created using the factory method CREATE of the class in question and by passing the source xml to the XML data, for example:

DATA(reader) = cl_sxml_string_reader=>create( xml ).

The static type of the reference variables is then the interface IF_SXML_READER and its methods and attributes can be addressed directly.

In the simplest parsing case, the method NEXT_NODE is applied as many times as it takes to reach the end of the XML:

reader->next_node( ).

Once the method is called, the attributes of the reader with the required properties of the node can be accessed directly. If the parser pauses on the node of an opened element, the method NEXT_ATTRIBUTE can be used to iterate across the list of attributes of an XML element:

reader->next_attribute( ).

After this method, the attributes of the reader object contain the name and the value of the current XML attribute. The end of the data is displayed by the value of the constants IF_SXML_NODE=>CO_NT_FINAL in the attribute NODE_TYPE. Any exceptions should be caught and handled in a TRY control structure.


Example

See Token-Based Parsing.

Methods and Attributes for Token-Based Parsing

Methods for token-based parsing

Alongside the simple iteration across nodes and attributes, the interfaces IF_SXML_READER offers several other methods for token-based parsing, summarized as follows:

  • Methods for parsing across nodes
  • NEXT_NODE - Moves the parser to the next node
  • CURRENT_NODE - Moves the parser back to the node for the opened element of the current node, if available. This method is ignored by value nodes or closed elements.
  • PUSH_NODE - Moves the parser back to the node for the opened element of the direct parent node.
  • SKIP_NODE - Parses all nodes from a node for an opened element to the associated closed element. The parser then pauses on the closed element. If required, the current node and all subnodes are passed to an XML writer. This method is ignored by value nodes or closed elements.
  • Methods for reading the attribute list
  • NEXT_ATTRIBUTE - Reads the next attribute in the list
  • NEXT_ATTRIBUTE_VALUE - Sets the attribute VALUE or VALUE_RAW of the read explicitly to the value of the current attribute
  • GET_ATTRIBUTE_VALUE - Sets the attribute VALUE of the reader to the value of a specific attribute


Notes

  • Token-based parsing parsing is designed mainly for forwards interaction through the XML data. Free navigation, as possible in iXML Library in DOM, is not recommended here. The method PUSH_BACK makes it possible to move back a step, but does not restore the reader to the state it had when it reached the node using NEXT_NODE. After a PUSH_BACK, the parsed node does not always produced the same result as the last time. In particular, subnodes already parsed could be skipped.

  • If an XML element has several attributes with the same name, the system uses all of these attributes - in the order in which they appear (see Example).

  • The method CURRENT_NODE is most useful when reading an attribute list, so jump back to the start.

  • Apart from simply skipping nodes, the method SKIP_NODE can also be used to check the subtrees or the full tree for errors or to copy trees. In particular, the format of the copied tree or subtree can be transformed into a different format (see Example).

Examples

Attributes for token-based parsing

The attributes of the reader with the properties of the current node are:

  • NODE_TYPE - Node type in accordance with the constants of the interface IF_SXML_NODE
  • PREFIX - Namespace prefix
  • NAME - Element name
  • NSURI - Namespace
  • VALUE_TYPE - Type of the value in accordance with the constants of the interface IF_SXML_VALUE:
  • CO_VT_TEXT - Text data in the attribute VALUE
  • CO_VT_TEXT - Raw data in the attribute VALUE_RAW
  • VALUE - Character-like value (if text data)
  • VALUE_RAW - Byte-like value (if raw data)

The value of the constants IF_SXML_NODE=>CO_NT_FINAL in the attribute NODE_TYPE indicate that the end of the XML data was reached.


Notes

  • When parsing, the attributes of a reader are only overwritten by non-initial content. If the parser is set to a literal element without a name, for example, the preceding content of the attribute NAME is kept. This applies particularly to the attribute VALUE, which is not initialized if a node does not have a value.

  • The attributes are declared in the interface IF_SXML_READER. In a reader class, they can also be addressed using alias names.