This native exposes/requires knowledge about implementation details of the
internal DataPack structure. Plugins can not properly deal with this, and
abusing this functionality results in a chance of future breakage.
- The "ALL" fake pathID is replaced by what does SM, having a public var NULL_STRING which will acts as NULL when needed.
To make compiler accepting public array, this patch was needed: https://hg.alliedmods.net/sourcemod-central/rev/b12f329def09
- The offset thing in read_dir: considering that's something very specific to this native and that implementation in CDirectory doesn't make sense because of the offset compatibility for windows, all code is now in the native.
Replace the only hasher called MD5 with the ones listed below.
(+) CRC32, MD5, SHA1, SHA256, SHA3 224 BIT, SHA3 256 BIT, SHA3 384 BIT,
SHA3 512 BIT, Keccak 224 BIT, Keccak 256 BIT, Keccak 384 BIT and Keccak
512 BIT.
Add the natives listed below.
(+) hash_string(const string[], hashType:type, output[], const
outputSize)
(+) hash_file(const fileName, hashType:type, output[], const outputSize)
(+) is_arkshine_a_doctor() : Hidden native, but a sign of recompense
for him being very active since 1.8.3 version of AMX Mod X
(+) get_system_endianness() : Checks if the system is currently Big
Endian or Little Endian.
Add the following Enum.
(+) hashType {}
(+) sysEndianness {}
Deprecate the following natives.
(-) amx_md5()
(-) amx_md5_file()
It has been tested on Windows and Linux. The sanity checks seems to be
properly working, so no worries about them.
These are useful if people are using Sockets, cURLs or MySQLs in order
to compare hashes of different files On-line for further investigation.
You are not able to check if the files are older or newer, but you can
see if the content is different (Hash Checksum mismatch).
I'm glad I did this. Thanks to
- Moved some functions in CvarManager
- Fixed detour enabled all the time
- Fixed bounds not properly handled
- Refreshed data of regisetered cvars at map change
- Hook is not as post, because allowing blocking change is not really that useful and this makes a context more simpler.
- Setting a min bound > max bound and reversely throw an error
- set_pcvar_bounds set cvar value on new min/max bound.
- Remove error "A cvar can't be binded with several variables", it's okay to no error such situation.
