This is better for performance because these then don't need to be reevaluated every time they are called.
When encountering an unqualified function or constant reference, PHP will first try to locate a symbol in the current namespace by that name, and then fall back to the global namespace.
This short-circuits the check, which has substantial performance effects in some cases - in particular, ord(), chr() and strlen() show ~1500x faster calls when they are fully qualified.
However, this doesn't mean that PM is getting a massive amount faster. In real world terms, this translates to about 10-15% performance improvement.
But before anyone gets excited, you should know that the CodeOptimizer in the PreProcessor repo has been applying fully-qualified symbol optimizations to Jenkins builds for years, which is one of the reasons why Jenkins builds have better performance than home-built or source installations.
We're choosing to do this for the sake of future SafePHP integration and also to be able to get rid of the buggy CodeOptimizer, so that phar and source are more consistent.
I don't know why this would be missing, but in some cases it is, as seen in the crash archive. Whatever the case, we shouldn't be shitting the bed because of this.
This takes advantage of two key behaviours of PHP:
1. Assigning a string does not copy the string
2. Changing an offset in a string causes the string to be copied.
These two factors combined, along with the fact that blocklight and skylight arrays are usually all-zeros, allow us to produce a significant memory usage reduction of loaded chunks.
A freshly generated PM world with 3,332 chunks loaded drops from 310MB to 200MB memory usage with these changes applied.
this process of fast-serialization, fast-deserialize, network-serialize is an order of magnitude slower than just doing the network encode directly on the main thread, and also copies more useless data.
For the main thread, the figures were something like 3x more expensive, and then an extra 7x for deserialization on the worker thread. This is a ridiculously large overhead.
This assumes that the region is properly garbage-collected and packed, but if the file contains uncollected garbage this may not be the case, resulting in a region larger than a gigabyte.
this goes on 3.1 because it changes the behaviour of chunk cloning, which might possibly break some plugins, and this isn't a bug fix.
This should see no change in behaviour other than a minor performance improvement and slight reduction in memory usage.
If the same chunk is set into multiple different places in the world, the chunk's position is no longer able to be relied on, because it will have the position of the last place it was set. This results in chunks not getting sent correctly when the same chunk is set in multiple places.
This avoids the bug by using known valid coordinates (using chunk hashes) to establish the real coordinates, and also adds an assert to notify developers should they unintentionally set a duplicate chunk by mistake.
This is a significant breaking change for anything utilizing Tiles as they now do not store their NBT at runtime anymore.
This is another step in the process of ridding PocketMine-MP of runtime NBT.
It can be noticed that all tiles are now using class fields to store properties instead of NBT, which is much faster, uses less memory and is also more concise when written in code.
Highlights:
- Tile->namedtag has been removed.
- Tile->saveNBT() now accepts a CompoundTag() parameter. Typically it's expected that this will be fed a newly-created CompoundTag (this part may be improved in the future).
- New internal methods Tile->readSaveData() and Tile->writeSaveData() have been added. Instead of overriding __construct() and saveNBT() to load and save properties from NBT, you should now implement these methods instead.
This is not final and will see further changes before it's done.
it appears that errors are occurring in the exception handler when handling corrupted regions, leaving regions in the provider index with incomplete location tables. This causes strange-looking errors later down the line.
This moves the region assignment to the end of the condition to avoid leaving incomplete/corrupted regions in the location table when errors occur.
This now removes logging from the level providers (for the most part) and replaces it with exception throws and catches. The implementation using the providers should catch these exceptions if they are thrown.
