incubator-stdcxx-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Martin Sebor <se...@roguewave.com>
Subject Re: test for lib.string.io
Date Sun, 16 Jul 2006 21:19:23 GMT
Farid Zaripov wrote:
>  > -----Original Message-----
>  > From: Martin Sebor [mailto:sebor@roguewave.com]
>  > Sent: Wednesday, July 12, 2006 3:39 AM
>  > To: stdcxx-dev@incubator.apache.org
>  > Subject: Re: test for lib.string.io
>  >
>  > With the patch applied I get 120 failed assertions out of the
>  > io test, all of them on lines 872 through 886. They all look
>  > like they are due to the test expecting failbit for
>  > insertions into the stream buffer that fail and the library
>  > setting badbit. I think the library does the right thing and
>  > the standard needs to be fixed to say that
>  > operator<<(ostream&, string) a) is a Formatted Output
>  > Function and b) should behave according to the requirements
>  > on such functions and thus should set badbit and not failbit
>  > when it fails to produce the expected output.
> 
>   I updated the test (failbit checking is replaced to the badbit 
> checking). The updated diff in attach.

I assume this needs the recent changes to rw_streambuf.h, right?
(They will need to be committed first.)

I made a few trivial formatting changes (please take a look at
them to make sure they make sense to you and follow the same
convention in the future -- thanks :) With the changes I say
go ahead and commit it. The next step will be replacing the
ctype facet with the one in rw_ctype.h.

Martin

Mime
View raw message