# Shell command language code generator
#
# Copyright (C) 2016  Patrick "P. J." McDermott
#
# This file is part of the Eggshell Compiler.
#
# The Eggshell Compiler is free software: you can redistribute it
# and/or modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation, either version 3 of
# the License, or (at your option) any later version.
#
# The Eggshell Compiler is distributed in the hope that it will be
# useful, but WITHOUT ANY WARRANTY; without even the implied warranty
# of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with the Eggshell Compiler.  If not, see
# <http://www.gnu.org/licenses/>.

sc=
tu_id=
static_fn_n=

sgetc()
{
	sc="$(dd bs=1 count=1 2>/dev/null; printf '.')"
	sc="${sc%.}"
}

add_static_fn()
{
	local fn="${1}"

	printf '__%s_static_fns="${__%s_static_fns} %s:__%s_fn%d"\n' \
		"${tu_id}" "${tu_id}" "${fn}" ${static_fn_n}
	printf '__%s_fn%d' "${tu_id}" ${static_fn_n}
	static_fn_n=$((${static_fn_n} + 1))
}

codegen_sub()
{
	local array="${1}"
	shift 1
	local ionum=
	local fname=
	local static=
	local static_fname=

	ionum=false
	fname=''
	static=false
	static_fname=false

	IFS="${RS}"
	for t in ${array}; do
		# Handle I/O number false positives.
		if ${ionum}; then
			case "${t%${US}*}" in
				T_WORD|T_CMDNAME|T_FNAME)
					printf ' '
				;;
			esac
			ionum=false
		fi

		# Function names
		case "${t%${US}*}" in T_FNAME)
			fname="${t}"
			;;
		esac

		# State machine for static variables and functions
		case "${t%${US}*}" in
			T_STATIC)
				static=true
				continue
				;;
		esac
		if ${static}; then
			static=false
			case "${t%${US}*}" in
				T_FNAME)
					static_fname=true
					continue
					;;
				*)
					toktext T_STATIC
					printf ' '
					;;
			esac
		elif ${static_fname}; then
			static_fname=false
			case "${t%${US}*}" in
				T_LPAREN)
					add_static_fn "${fname#*${US}}"
					;;
				*)
					toktext T_STATIC
					printf ' '
					toktext "${fname}"
					;;
			esac
		fi

		# Function start/end tokens
		case "${t%${US}*}" in
			T_FN_START)
				printf '__fn_start %s\n' "${fname#*${US}}"
				;;
			T_FN_END|T_RETURN)
				printf '__fn_end\n'
				;;
		esac

		toktext "${t}"

		case "${t%${US}*}" in
			T_NEWLINE|T_LPAREN)
				# Indenting lines can mess up here-documents.
				# Adding a space after "(" can mess up function
				# definitions (on zsh at least, while other
				# shells accept the space).
				;;
			T_IO_NUMBER)
				ionum=true
				;;
			*)
				printf ' '
				;;
		esac
	done
	unset IFS
}

# The token stack is encoded in a string in the following grammar:
#     Terminal symbols:
#         TOKEN
#     Production rules:
#         stack  = tokens [ '<SOH>' type '<STX>' stack '<ETX>' [ tokens ] ] ;
#         tokens = TOKEN { '<RS>' TOKEN } ;
#         type   = 'C' ;
# We need to recurse through this stack to get to all the tokens.
# Each element in the stack (an array of tokens) gets run through the codegen to
# become text that is inserted into the array below.
sh_parse_stack()
{
	local array=

	array=''
	while :; do
		sgetc
		case "${sc}" in
			'')
				# EOF
				break
				;;
			"${SOH}")
				# New stack element
				sgetc
				case "${sc}" in
					'C')
						# Command substitution
						sgetc  # STX
						array="${array}$(\
							sh_parse_stack)."
						array="${array%.}"
						;;
				esac
				;;
			"${ETX}")
				# End of stack element
				break
				;;
			*)
				# Token character
				array="${array}${sc}"
				;;
		esac
	done
	codegen_sub "${array}"
}

sh_set_tu_id()
{
	local toks="${1}"

	tu_id="$(printf '%s' "${toks}" | sha256sum -)"
	tu_id="${tu_id%  -}"

	return 0
}

sh_start_tu()
{
	printf '__tu=%s\n' "${tu_id}"
	printf '__%s_static_vars=\n' "${tu_id}"
	printf '__%s_static_fns=\n' "${tu_id}"
	static_fn_n=0
}