apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Cliff Woolley <jwool...@virginia.edu>
Subject Re: svn commit: r107007 - /apr/apr/trunk/CHANGES /apr/apr/trunk/include/apr_lib.h /apr/apr/trunk/passwd/apr_getpass.c
Date Tue, 30 Nov 2004 15:26:31 GMT
On Tue, 30 Nov 2004, Jeff Trawick wrote:

> >   * @param pwbuf Buffer to store the password
> >   * @param bufsize The length of the password buffer.
> > + * @remark If the password entered must be truncated to fit in
> > + * the provided buffer, APR_ENAMETOOLONG will be returned.
> >   */
> >  APR_DECLARE(apr_status_t) apr_password_get(const char *prompt, char *pwbuf,
> >                                             apr_size_t *bufsize);
>
> another disturbance: we force caller to go to the trouble to pass by
> address, but we don't update the size on output to indicate either the
> number of bytes stored or the number of bytes needed; shrug or "fix"?

By number of bytes stored, I assume you mean bytes stored in the non-error
case, since bytes stored in the error case would just be bufsize.  Right?

If that's the case, then it's maybe worth fixing but I'm not sure I care.
How many bytes would be needed seems to make somewhat more sense as being
a useful value, but do we have any other functions that work that way?

Mime
View raw message