This mod for Canvas is intended to make it easier for Canvas mod developers to test and prototype their ideas using the Lua scripting language.
The mod gives access to most of the Canvas APIs.
The cipher table is the table that contains all of the static APIs of Canvas.
number cipher.getGameVersion
returns the game's version number
boolean cipher.isGameBeta
returns if the game that is currently running is a beta version of the game
light_userdata cipher.get_libBase
returns the pointer to the start of the library in memory as light userdata
string cipher.get_libName
returns the name of the game's library (libBootloader.so)
light_userdata cipher.CipherScan(string pattern, string mask)
pattern - the pattern, consisting of raw characters
mask - the mask for the pattern
The pattern's characters will be masked with characters from the mask (x in mask meaning an exact match, ? meaning an "any char" match), and if a result matching this would be found, the function will return the start of that result as light userdata, or nil if it was not found
returns result of the pattern scan as light userdata or nil if not found
the lu table contains functions to operate with light userdata
light_userdata lu.fromhex(string hex)
hex - the string of hexadecimal characters
returns string hex converted to light userdata
light_userdata lu.add(light_userdata data, integer x)
data - source light userdata
x - the integer to add to light userdata
returns light userdata, equal to data+x
light_userdata lu.multiply(light_userdata data, integer x)
data - source light userdata
x - the integer to multiply light userdata by
returns light userdata, equal to data*x
light_userdata lu.divide(light_userdata data, integer x)
data - source light userdata
x - the integer to divide light userdata by
returns light userdata, equal to data/x
integer lu.read<u>int<8/16/32/64>(light_userdata where_from)
NOTE: there are no functions to read the int8 data type since its pointless
where_from - the light userdata from which the (u)int<8/16/32/64> will be read
returns an integer read from memory that where_from points to
number lu.read<float/double>(light_userdata where_from)
where_from - the light userdata from which the float or double will be read
returns a number with a value of the float/double read from memory that where_from points to
light_userdata lu.readptr(light_userdata where_from)
where_from - the light userdata from which the pointer will be read
returns a light userdata with the value read from memory that where_from points to
string lu.readstring(light_userdata where_from)
where_from - the light userdata from which the null terminated C string will be read
returns the string corresponding to the null terminated C string read from memory that where_from points to
string lu.readstring(light_userdata where_from, integer length)
where_from - the light userdata from which the string will be read
length - how much characters needs to be read into the string
returns the string corresponding to bytes starting from where_from and ending at where_from+length (unless a null character preceeds where_from+length)
void lu.write<u>int<8/16/32/64>(light_userdata where_to, integer value)
NOTE: there are no functions to write the int8 data type since its pointless
where_to - the light userdata to where the (u)int<8/16/32/64> will be written
value - the value that will be written
returns nothing, writes a int<8/16/32/64> to the memory pointed by where_to
void lu.write<float/double>(light_userdata where_to, number value)
where_to - the light userdata to where the float or double will be written
value - the value that will be written
returns nothing, writes a float or double to the memory pointed by where_to
void lu.writestring(light_userdata where_to, string value, boolean null_terminated=true)
where_to - the light userdata to where the string will be written
value - the string that will be written
null_terminated - true if the written string needs to have a null terminator on the end, false otherwise
returns nothing, writes a string to the memory pointed by where_to with or without the null terminator depending on the null_terminated argument
cipherpatch is a generic and relatively safe way to patch the game library code
CipherPatch cipherpatch.new
returns a new CipherPatch
NOTE - you need to store this globally, if the patch gets garbage collected it will revert itself
void CipherPatch:set_Opcode(string opcode)
opcode - a string with an even amount of hexadecimal characters
returns nothing, sets the replacement byte(s) for the patch
void CipherPatch:set_Address(light_userdata address, boolean is_local=true)
address - the start address where byte(s) would be replaced
is_local - set to true if the address is relative to the library, false otherwise
returns nothing, sets the address where the byte(s) would be replaced
void CipherPatch:set_Address(string symbol)
symbol - the symbol at the address of which the byte(s) would be replaced
returns nothing, sets the address where the byte(s) would be replaced
void CipherPatch:set_Address(string pattern, string mask) NOTE: USE AT YOUR OWN RISK! This function uses an irrelibale KittyMemory::find_from_lib API which is not guaranteed to fully scan the library
pattern - the pattern, consisting of raw characters
mask - the mask for the pattern
Uses the same pattern scan algorithm as CipherScan to find the address
returns nothing, sets the address where the byte(s) would be replaced to the result of the pattern scan
void CipherPatch:set_libName(string name)
name - the library name
NOTE: must be called before set_Address
returns nothing, sets the library from which the local adresses would be relative to and which would be scanned when using set_Address
void CipherPatch:set_Lock(boolean lock)
lock - if this patch is locked or not
If the patch is set to locked, no other CipherPatch can make a patch to this memory address
returns nothing, sets the locked status of the patch
void CipherPatch:Fire()
returns nothing, applies the patch
void CipherPatch:Restore()
returns nothing, reverts the patch
dyncall allows you to call any C(++) function by its pointer, if you know the arguments and the return value
mixed dyncall(<cfunction/number/light_userdata> function_ptr, string signature, ...)
function_ptr - the pointer to the function that needs to be called
signature - a dyncall function signature, as described in the dyncall manual
... - arguments for the function that needs to be called, as specified in the dyncall signature
returns the result of the function call, with the type dictated by the dyncall signature