Correction of Invalid XML Documents with Respect to Single Type Tree Grammars

Approximate XML structure validation based on document–grammar tree similarity

“…XML grammars has recently gained attention, as one of the aspects of XML data management [7,8,14,22,36,49]. Here, XML grammars are generally viewed as context-free (DTD-like) regular tree grammars [58]. Thus, verifying if an XML document D conforms to XML grammar G comes down to checking whether the document tree is included in the language defined by the grammar, i.e., if D 2 LðGÞ.…”

“…Methods for identifying the edit script transforming a given XML document, to another document conforming to a given DTD grammar, have been proposed in [57,58,79,80]. The approach in [57] builds a special graph structure G, based on the XML document tree D, underlining all possible transformation operations applicable to D (i.e., node insertion, deletion and update).…”

“…A problem comparable to that of document-to-grammar transformation is that of document-to-grammar correction [6,15,58]. The scenario considered here is that of dynamic XML documents which are modified and updated frequently, underlining the need to continuously test their conformance w.r.t.…”

“…The approach is shown exponential in the size of document node fanout (maximum node degree), and has been exploited to incrementally validate XML integrity constraints defined as XML functional dependencies [13]. In [54,58] the authors extend document-to-grammar (DTD) correction to deal with more expressive XML schemas, represented as simple type tree grammars (where the left-hand side of a production rule may be surrounded by context information, consisting of terminal symbols), representing repairs as sequences of edit operations to alter XML trees. The authors in [52,53] investigate user-defined XML document adaptations, i.e., sequences of document transformation operations intended to adapt documents valid for an original grammar G to a new grammar G 0 .…”

“…While the methods in [6,15,57,58,79] produce transformation and correction scripts, they do not however address the issue of XML document/grammar similarity.…”