Martin Sebor wrote: > This seems to include the changes for reverse_iterator. Shouldn't those be implemented separately (and in the driver instead)?
Yes. I committed changes for range overloads in cons test here: http://svn.apache.org/viewvc?rev=414235&view=rev The updated test and test driver which supports the reverse iterators are here: http://people.apache.org/~antonp/stdcxx06142006/ Thanks, Anton Pevtsov -----Original Message----- From: Martin Sebor [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 13, 2006 21:05 To: [email protected] Subject: Re: svn commit: r412249 - in /incubator/stdcxx/trunk/tests: include/21.strings.h src/21.strings.cpp Anton Pevtsov wrote: > The update test version is here: > http://people.apache.org/~antonp/stdcxx06132006/ This seems to include the changes for reverse_iterator. Shouldn't those be implemented separately (and in the driver instead)? Martin > > I'll commit it with changes to other tests correcting the typos. > > Thanks, > Anton Pevtsov > > > -----Original Message----- > From: Martin Sebor [mailto:[EMAIL PROTECTED] > Sent: Friday, June 09, 2006 21:39 > To: [email protected] > Subject: Re: svn commit: r412249 - in /incubator/stdcxx/trunk/tests: > include/21.strings.h src/21.strings.cpp > > > Anton Pevtsov wrote: > >>Here are the diffs to the ctor test required by the testing of the >>range >>overloads: >>http://people.apache.org/~antonp/stdcxx06092006/ > > > Why don't we reuse the virtual RangeBase member function used for > other template members and have test_ctor pass it str_arg? CtorRange > would ignore the argument and CtorRangeOverload would use it instead > of re-creating its own copy. > > Btw., I think you have a copy-n-paste typo in test_cons: the name of > the RangeBase and CtorRangeOverload objects should probably not be > "app." Your "rng" (from test_ctor() argument) sounds better and, if > used consistently throughout the all the tests (which I should have > thought > of) would avoid these kinds of typos. > > Finally, I'd like to use either "cons" or "ctor" in the names of all > functions. I don't really care which as long as we are consistent > (although given the name of the test we need to go with cons :) > > With that: okay to commit. > > Thanks > Martin
