extern	NR_write
extern	strlen
extern	strcat

section .rodata
	NL		db	0xA
section .bss
	printfBuff	resb	4096
section .text
	global	print
	global	puts
	global	printf

;----- print (char* string) -----;
; return value: N/A
print:
	push	rbp
	mov	rbp, rsp

	call	strlen
	mov	rdx, rax
	mov	rax, NR_write
	mov	rsi, rdi
	mov	rdi, 1
	syscall

	leave
	ret
;----- puts (char* string) -----;
; return  value: N/A
puts:
	push	rbp
	mov	rbp, rsp

	mov	r10, rdi
	call	print
	mov	rdi, r10
	mov	rax, NR_write
	mov	rdi, 1
	mov	rsi, NL
	mov	rdx, 1
	syscall

	leave
	ret
;----- printf(const char* string, ...) -----;
printf:
	push	rbp
	mov	rbp, rsp

	push	r12
	push	r13
	;sub	rsp, 8

	xor	r10, r10
	lea	r11, [rel printfBuff]
	.makeStr:
		cmp	byte [rdi], 0x0
		je	.finish
		cmp	byte [rdi], '%'
		je	.replaceArg
		mov	r12b, byte [rdi]
		mov	byte [r11], r12b
		jmp	.continue
		.replaceArg:
			cmp	byte [rdi+1], 0x0
			je	.continue
			cmp	byte [rdi+1], 'd'
			je	.rep_d
			cmp	byte [rdi+1], 's'
			je	.rep_s

			.rep_d:
			;TODO
			jmp	.continue

			.rep_s:
			push	rdi
			;mov	byte [r11], 's' ;%s => Ss
			cmp	r10, 0
			je	.sinsertLoop ;nothing to do, rsi already correct
			cmp	r10, 1
			je	.rep_s_rdx
			cmp	r10, 2
			je	.rep_s_rcx
			cmp	r10, 3
			je	.rep_s_r8
			cmp	r10, 4
			je	.rep_s_r9

			;get from stack
			; rsp + 6*8 : RIP to printf call (+1), PUSH rbp,r12,r13,rdi (+4), +1 should be tS6
			mov	rsi, qword [rsp + 5*8 + (r10-5)*8]
			jmp	.sinsertLoop

			.rep_s_rdx:
			mov	rsi, rdx
			jmp	.sinsertLoop
			.rep_s_rcx:
			mov	rsi, rcx
			jmp	.sinsertLoop
			.rep_s_r8:
			mov	rsi, r8
			jmp	.sinsertLoop
			.rep_s_r9:
			mov	rsi, r9

			.sinsertLoop:
				cmp	byte [rsi], 0x0
				je	.s0f
				mov	r13b, byte [rsi]
				mov	byte [r11], r13b
				inc	rsi
				inc	r11
				jmp	.sinsertLoop
			.s0f:
			inc	r10
			pop	rdi
			add	rdi, 2
			jmp	.makeStr
		.continue:
		inc	rdi
		inc	r11
		jmp	.makeStr
	.finish:
	mov	byte [r11], 0x0
	lea	rdi, [rel printfBuff]
	call	print

	;add	rsp, 8
	pop	r13
	pop	r12

	leave
	ret