### Table 1: Basic conditions on I 16Remark that the notion of constraint satisfaction refers to an arbitrary set of literals I, be it contradictory or otherwise.

1995

Cited by 88

### Table 1: Basic conditions on I 16Remark that the notion of constraint satisfaction refers to an arbitrary set of literals I, be it contradictory or otherwise.

1995

Cited by 88

### Table unique and table referential constraints are specialized syntactic forms for expressing key and foreign key properties: A quot;table check constraint quot; (TCC for short) is a generic constraint able to impose arbitrary restrictions on rows of the respective table. As an example consider:

### Table 1. Predefined aggregates. The goal, G, is an arbitrary conjunction of CHR constraints, guards and aggregates. The template X is an arbitrary term. Instantiations of the template, based on the different matchings of the goal, are used to increment and decrement the aggregates value (cf. infra). This template is similar to the first argument of the well-known findall/3 ISO Prolog predicate [1]. For all arithmetic aggregates, X must evaluate to a ground arithmetic expression at runtime.

"... In PAGE 4: ...f aggregate expressions is given in Section 3.3. Predefined aggregates. Table1 lists the proposed predefined aggregates. Sec- tion 2.... In PAGE 5: ... This paper uses Prolog as a host-language, but our approach is applicable to any host-language. The last three arguments correspond to the arguments of the predefined arguments of Table1 . The seman- tics of the arguments is further explained below, and more formally in Section 3.... ..."

### Table 1. Prede ned aggregates. The goal, G, is an arbitrary conjunction of CHR constraints, guards and aggregates. For all arithmetic aggregates, X has to be a ground arithmetic expression at runtime. For collect/3, X is just an arbitrary template, as e.g. in the well-known findall/3 ISO Prolog predicate [1].

2007

"... In PAGE 6: ...ate semantics is given in Section 3.3. Prede ned aggregates. Table1 lists the prede ned aggregates in our system.... In PAGE 7: ... This paper considers Prolog as the host- language, where the rst four are terms representing the Prolog predicates that have to be called (after extending them with some extra arguments). The last three arguments are analogous to those in Table1 . The concrete operational semantics of the arguments is explained below.... In PAGE 26: ... However, to the best of our knowledge, no PR system provides syntactic shorthands for commonly used aggregates (cf. Table1 ). Also, as far as we know, no PR system o ers incremental aggregate maintenance.... ..."

### Table 3.1 shows examples of five types of hard constraints. The first four types, from frequency to exclusivity, impose regularities that the source schema must conform to. The last type, column, imposes regularities that both the source schema and data must conform to. We can specify arbitrary hard constraints that involve only the schemas, because given any candidate mapping, they can always be checked. Constraints involving data elements cannot always be checked because we have access only to the current source data. (Even when all the data in the source at a given time conforms to a constraint, that still does not mean the constraint holds on the source.) In many cases, however, the few data instances we extract from the source will be enough to find a violation of such a constraint.

2002

### Table 4: Two sets of selectivity estimates for atomic probabilistic selection conditions. L gt; prob. The probability that an arbitrary time point t satis es the L gt; prob constraint can be bounded above by PROB gt; = min 1;

"... In PAGE 24: ... We propose two sets of cardinality estimations, stemming from two somewhat di erent approaches. Table4 summarizes the formulas for the two sets. In this table PROB gt; denotes the following expression: PROB gt; = min 1; 1 prob (MAX TP(r) ? MIN TP(r) + 1) : Some explanations about the intuition behind the formulas for both sets are in order.... ..."

### Table 7.7: arbitrary Distribution varying epsilon1FRAC. Goods = 50, Bids = 300. Frac-CG = total number of times we call constraint generation to check fractions. Dur-CG = total number of times we call constraint generation to check durations.

2005

### Table 7.8: arbitrary Distribution varying epsilon1DUR. Goods = 50, Bids = 300. Frac-CG = total number of times we call constraint generation to check fractions. Dur-CG = total number of times we call constraint generation to check durations.

2005

### Table 7.9: arbitrary Distribution varying per-epsilon1DUR. Goods = 50, Bids = 300. Frac-CG = total number of times we call constraint generation to check fractions. Dur-CG = total number of times we call constraint generation to check durations.

2005