|
Boost : |
From: Jeremy Siek (jsiek_at_[hidden])
Date: 2001-03-28 23:58:52
Right, that's an oversight in the documentation. I've added copy
constructible to the requirements of the visitors, and added the
suggestion that they be lightweight.
On Wed, 28 Mar 2001, Kevin S. Van Horn wrote:
kevin_> The documentation for the various Visitor concepts does not require that
kevin_> visitors be copyable. This strikes me as a bit odd, given that the
kevin_> algorithms that use them take parameters passed by value, and not by
kevin_> reference. In fact, taking a look at variant (3) of the
kevin_> dijkstra_shortest_paths algorithm, the last argument (UniformCostVisitor
kevin_> visit) is in fact copied to provide the last argument of a nested call to
kevin_> variant (4).
kevin_>
kevin_> So I think that either you need to eliminate any copying of visitors, and
kevin_> mention in the documentation that visitors are not required to be
kevin_> copyable, or the visitor concepts should require that visitors be
kevin_> copyable, and the documentation should mention that visitors are intended
kevin_> to be lightweight objects that are inexpensive to pass by value.
kevin_>
kevin_>
kevin_>
kevin_> To unsubscribe, send email to: <mailto:boost-unsubscribe_at_[hidden]>
kevin_>
kevin_>
kevin_> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
kevin_>
kevin_>
kevin_>
----------------------------------------------------------------------
Jeremy Siek www: http://www.lsc.nd.edu/~jsiek/
Ph.D. Candidate email: jsiek_at_[hidden]
Univ. of Notre Dame work phone: (219) 631-3906
----------------------------------------------------------------------
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk