diff options
| author | Arseny Kapoulkine <arseny.kapoulkine@gmail.com> | 2014-11-07 19:08:49 +0100 |
|---|---|---|
| committer | Arseny Kapoulkine <arseny.kapoulkine@gmail.com> | 2014-11-07 19:08:49 +0100 |
| commit | d854b0219dcb1cecc42e87b1c397cb683967b74d (patch) | |
| tree | 7f0735de5e68668def06fcd1883799928fc35b13 /src | |
| parent | db78f34054275890000e0f114100730022e5ad16 (diff) | |
XPath: Only call apply_predicates if necessary
In some cases constant overhead on step evaluation is important - i.e. for
queries that evaluate a simple step in a predicate expression. Eliminating
a redundant function call thus can prove worthwhile.
This change makes some queries (e.g. //*[not(*)]) 4% faster.
Diffstat (limited to 'src')
| -rw-r--r-- | src/pugixml.cpp | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/src/pugixml.cpp b/src/pugixml.cpp index a3aaf76..1a93d47 100644 --- a/src/pugixml.cpp +++ b/src/pugixml.cpp @@ -9025,13 +9025,13 @@ PUGI__NS_BEGIN if (axis != axis_self && size != 0) ns.set_type(xpath_node_set::type_unsorted); step_fill(ns, *it, stack.result, once, v); - apply_predicates(ns, size, stack, eval); + if (_right) apply_predicates(ns, size, stack, eval); } } else { step_fill(ns, c.n, stack.result, once, v); - apply_predicates(ns, 0, stack, eval); + if (_right) apply_predicates(ns, 0, stack, eval); } // child, attribute and self axes always generate unique set of nodes |