Publications (2)0 Total impact
-
Article: Path Rewriting in Semistructured Data
[show abstract] [hide abstract]
ABSTRACT: We consider in this paper rooted labeled directed graphs as models for semistructured data. A path inclusion constraint, in such graphs, is an expression p <= q where p and q are regular expressions over the alphabet of labels. An path inclusion constraint p <= q is satisfied by a rooted labeled directed graph if the set of nodes reached, from the root, by the words of the language described by p is included in the set of nodes reached, from the root, by the words of the language described by q. We answer here to the following problem : given a set C of path inclusion constraints, given a regular expression p describing an infinite regular language, we want to compute, if it exists, a regular expression f describing a finite language, such that p <= f for every graph satisfying C.4th International Conference on Combinatorics on Words. -
Article: Extraction and Implication of Path Constraints
[show abstract] [hide abstract]
ABSTRACT: We consider semistructured data as rooted edge-labeled directed graphs, and path inclusion constraints on these graphs. In this paper, we give a new decision algorithm for the implication problem of a constraint $p \preceq q$ by a set of constraints $p_i \preceq u_i$ where $p$, $q$, and the $p_i$'s are regular path expressions and the $u_i$'s are non-empty paths, improving in this particular case, the more general algorithms of S. Abiteboul and V. Vianu, and Alechina et al. Moreover, in the case of a set of word equalities $u_i \equiv v_i$, we give a more efficient decision algorithm for the implication of a word equality $u \equiv v$, improving the more general algorithm of P. Buneman et al., and we prove that, in this case, the implication problem for non-deterministic models or for (complete) deterministic models are equivalent.29th Symposium on Mathematical Foundations of Computer Science.
