On Sat, Aug 03, 2019 at 07:29:34PM -0400, Sam Varshavchik wrote:
> I believe that the following construct trips this assertion:
>
> # std::vector<int> foo;
> #
> # std::vector<int> bar;
> #
> # // Populate foo with something.
> #
> # std::copy(&foo[0], &foo[foo.size()], std::back_insert_iterator{bar});
It's true that something like
copy(&foo[0], &foo[foo.size()], back_inserter(bar));
triggers an assertion when compiled with -D_GLIBCXX_ASSERTIONS. Because
the checking code just can't look at the context of its invocation.
Which is fine, because you really don't need to use such constructs.
> There's nothing wrong with this. There is no out of bounds access. I do not
> believe that this is undefined behavior. The defined semantics of
> std::vector, and its operator[], are well defined here.
It's defined behaviour but there is something wrong with it: it isn't
idiomatic C++ and it's tedious and error-prone.
> I ran into these new assertions with my own code as well, after updating to
> F28 (where they were enabled by default the first time, IIRC, not sure why
> this shows up only now, for that package).
>
> I ended up tweaking my code to avoid the assertions, rather than disabling
> them. For this particular situation, my original change was to try
>
> std::copy(&foo[0], &foo[0]+foo.size(), std::back_insert_iterator{bar});
>
> But that still tripped the assertion when the foo vector was empty, so I had
> wrap this inside an if().
But why don't you use the idiomatic
copy(foo.begin(), foo.end(), back_inserter(bar));
?
No need to wrap it into an extra if statement.
With GCC, the generated code is basically the same as when using:
copy(foo.data(), foo.data()+foo.size(), back_inserter(bar));
vector::data() is available since C++11. This doesn't trigger
any assertion and works with an empty source vector, as well.
What also works in general and doesn't trigger any assertion is the
'ultra ugly':
copy(&*foo.begin(), &*foo.end(), back_inserter(bar));
But there is also really no need to use an back_insert_iterator here - a
direct solution:
bar.insert(bar.end(), foo.begin(), foo.end());
Best regards
Georg
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
