[RFC] Strive to prefer sbumpc() to snextc()?!?

Paolo Carlini pcarlini@suse.de
Thu Apr 22 11:45:00 GMT 2004


Paolo Carlini wrote:

> The latter results are obtained via the attached **broken** patch,
> which uses sbumpc() and, incorrectly, sungetc() (which may fail!).

Actually, *perhaps* the attached "proof of concept" patch is not so
broken, after all ;) ...

1- Passes the testsuite (we have got quite a few getline testcases)
2- sungetc() happens only after a sbumpc() not returning eof.

Paolo.



More information about the Libstdc++ mailing list