Re: apr_off_t is of an ambiguous size.

Ben Reser wrote:

>On Sun, Jan 18, 2004 at 08:53:21AM +0100, Branko ??ibej wrote:
>
>
>>Not on Windows, where apr_off_t is always 64-bit.
>>
>>
>
>I didn't say anything about Windows. All I was saying is we know there
>are platforms with 32-bit apr_off_t's. Nor did I say every platform has
>a 32-bit apr_off_t.
>
>
I wasn't saying you had.

>>If anyone takes the time to make this a complete patch, please don't
>>introduce svn_offset_t. Use the existing svn_filesize_t. Yes, I know
>>that's unsigned, but we can fix that (i.e., either make it signed, or
>>tell svn_io_file_seek of we're seeking backwards or forwards from the
>>current position). But I don't want to see two different types.
>>
>>
>
>What's wrong with two types? We already have two types if we make no
>change. It's not like this is really introducing a new type. It's just
>replacing apr_off_t with svn_offset_t.
>
>
I know we have two types now, and I don't like it, for the reasons you
describe below.

-- 
Brane Čibej   <brane_at_xbc.nu>   http://www.xbc.nu/brane/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org