Lines Matching full:we
109 * As a heuristic, we stop prediction when we see any conflicting subset
110 * unless we see a state that only has one alternative associated with it.
118 * `12|1|[]` and `12|2|[]` conflict, but we cannot stop processing this
123 * After matching input A, we reach the stop state for rule A, state 1.
126 * However, alternative 3 will be able to continue and so we do not stop
127 * working on this state. In the previous example, we're concerned with
129 * associated with the conflicting configs, but since we can continue
134 * To handle pure SLL parsing, all we have to do is make sure that we
136 * predicate. From there, we can do the usual SLL termination heuristic.
148 * semantic predicate contexts so we might see two configurations like the
153 * Before testing these configurations against others, we have to merge
155 * For example, we test `(x+x') === x''` when looking for conflicts in
170 * to match additional input so we terminate prediction.
179 // contexts if we can fail over to full LL; costs more time
180 // since we'll often fail over anyway.
192 // now we have combined contexts for configs with dissimilar preds
246 * Can we stop looking ahead during ATN simulation or is there some
247 * uncertainty as to which alternative we will ultimately pick, after
248 * consuming more input? Even if there are partial conflicts, we might know
250 * means we can stop since no more lookahead will change that fact. On the
252 * minimums. That means we need more look ahead to decide which of those
253 * alternatives we should predict.
279 * more lookahead will help us. We will always pick that alternative. If,
280 * however, there is more than one alternative, then we are uncertain which
281 * alternative to predict and must continue looking for resolution. We may
285 * The biggest sin is to terminate early because it means we've made a
286 * decision but were uncertain as to the eventual outcome. We haven't used
299 * but `x=x'`. Because we merge all `(s, i, _)` configurations together,
320 * conflicting alternative subsets has more than one alternative. We are
324 * alternatives are still in the running for the amount of input we've
326 * configurations that won't lead to more states because we resolve
349 * If all states report the same conflicting set of alternatives, then we
350 * know we have the exact ambiguity set.
354 * In other words, we continue examining lookahead until all `A_i`
358 * ambiguity is, we have to know whether the ambiguity is between one and
359 * two or one and three so we keep going. We can only stop prediction when
360 * we need exact ambiguity detection when the sets look like