The previous algorithm would seek every time. This meant that although reading
from a filesystem would be quick (especially with disk caching), but reading
from a zip would be extremely slow.
The new version is rouhgly as fast on a filesystem, but appears not to slow
down on zips at all. Not in my tests, at least.
--HG--
branch : minor
Because of course that went wrong. I'm not sure whether it's the merge's fault,
or the merged in boot.lua.h was outdated. Either way, here we are.
--HG--
branch : minor
This way lua code can stop every iteration (usually frame), and return to the
c++ code, so love can work in environments with co-operative multithreading
like emscripten.
It may be interesting to see if we can make this optional, so this only happens
on platforms that require it, but I doubt it's significant in the grand scheme
of things.
--HG--
branch : minor
The new StringMap code for luax_enumerror causes code to be generated for
std::vector. By using an extern template (an old feature standardised in c++11)
we can instantiate it once instead, in StringMap.cpp.
--HG--
branch : minor
Previously every (internal) update the decoder would advance 0 or 1 frame. Now
it will advance until it catches up. Like before, if it takes too many steps
(5) before catching up, it seeks instead. This should mostly mean seeking
forwards has less latency.
--HG--
branch : minor
This makes the internal API more modular, but the split also forces a cleaner
implementation. It also means the OggDemuxer is now responsible for the
ogg-specific stuff like seeking and (ogg-specific) type detection.
Backported from the love.video-vpx branch of love-experiments
--HG--
branch : minor
This means you can use VERTEX or PIXEL in your shader code without them
disappearing (and getting a vague error). It's still not necessarily a good
idea to use them as names, but at least it doesn't mysteriously break now.
--HG--
branch : minor
All enum errors have (hopefully) been changed to a luax_enumerror, which has a
fixed error message. If additionally a list of valid options is passed, it
lists that in the error message. For every enum error with few options I've
implemented this using a getConstants call.
This solution has been designed specifically to reduce the number of template
instantiations (as I've been told that was a concern). All new template code
happens in places where there already was an instantiation of the relevant
StringMap. Of course there is still std::vector<std::string>...
--HG--
branch : minor
- Moved love.math.compress / decompress / decode / encode / hash to love.data.
- Changed love.data.compress/decompress to take the format argument first instead of second.
- Added love.data.newDataView. Returns a read-only subsection of an existing Data object.
- Added love.data.newByteData. Mostly useful in combination with LuaJIT's FFI as it has no extra methods currently.
- Added implementations of Lua 5.3's data packing APIs: love.data.packString / packData / unpack / getPackedSize.
Resolves issue #1336.
Resolves issue #1331.
--HG--
branch : minor
Also changed the optional 'columnMajor' argument of the matrix variant of Shader:send and Transform:setMatrix to accept enum constants instead. Valid values are "row" and "column".
--HG--
branch : minor
The pop("all") api was a bit awkward, especially considering there's a
push("all") that does something else. I think any use case that benefited from
pop("all") benefits from getStackDepth just as much if not more. There is also
additional functionality, like being able to find the source of a mismatched
push/pop.
--HG--
branch : minor
Alex Szpakowski
Bart van Strien