Re: new release status
Diego Biurrun <[email protected]> Wed, 25 Jan 2006 10:40:01 +0100
| Newsgroups | gmane.comp.video.ogle.devel |
|---|---|
| Message-ID | <[email protected]> |
On Sat, Jan 21, 2006 at 02:30:35AM +0100, Bj|rn Englund wrote: > > About your patches: > The asm patch will probably not go in. Nothing in libdvdread is so > time critical to need it so I'd rather not have any asm in there. Maybe this is worth benchmarking. Byteswapping can happen quite often. I don't see a bit of ASM as a problem, it's just a few lines... > The audio/subp_control is a bit problematic. > At the moment you have access to the data but only as a uint16_t > that you have to pick out the bits from yourself. > The patch would mean breaking source compatibility. > > An alternative would be to create macros to get at the bits instead, > but maybe not as nice. > > A third alt is to define just the audio/subp_control_t types in the .h file > and cast the uint16_t to the new types when you want to use the > bitfields. > > Or make a #define before including dvdread.h with something like > #define DVDREAD_NEW_CONTROL_INTERFACE if you want the new bitfields, > this alternative is probably just ugly. The comments in the source indicate that a new type was desirable, now you have it... > Any comments on this. Is it worth breaking source compatibility for > just these bitfields in this release? What about the next release? :) > Are you sure the control/mapping field should always be decoded the > same way so we don't have to change the bitfields later on? > Using macros would help in that case. No, it's not my patch. Shall I try contacting the patch sender? Diego