NSIS Unicode - How to get all entries in section of INI file - nsis

Question

At https://nsis.sourceforge.io/Get_all_entries_in_section_of_INI_file there is an ANSI version, but I need an unicode solution.
My first try was remove the A in the api calls (GetPrivateProfileSection and lstrlen), and increasing the memory to 8192.
Function GetSection
Exch $2
Exch
Exch $1
Exch
Exch 2
Exch $0
Exch 2
Push $3
Push $4
Push $5
Push $6
Push $8
Push $9
System::Alloc 8192
Pop $3
StrCpy $4 $3
System::Call "kernel32::GetPrivateProfileSection(t, i, i, t) i(r1, r4, 8192, r0) .r5"
enumok:
System::Call 'kernel32::lstrlen(t) i(i r4) .r6'
StrCmp $6 '0' enumex
System::Call '*$4(&t8192 .r9)'
Push $0
Push $1
Push $2
Push $3
Push $4
Push $5
Push $6
Push $8
Call $2
Pop $9
Pop $8
Pop $6
Pop $5
Pop $4
Pop $3
Pop $2
Pop $1
Pop $0
StrCmp $9 'StopGetSection' enumex
IntOp $4 $4 + $6
IntOp $4 $4 + 1
goto enumok
enumex:
System::Free $3
Pop $9
Pop $8
Pop $6
Pop $5
Pop $4
Pop $3
Pop $2
Pop $1
Pop $0
FunctionEnd
Despite that my application crashes.

Answer 1

Unicode characters are 2 bytes, use System::StrAlloc to allocate the buffer or double the number passed to Alloc.
The ini function deals with number of characters, not number of bytes. It overflowed the buffer because you had not allocated as much as you told it.
I edited the wiki page, you can try the new version...