summaryrefslogtreecommitdiffstats
path: root/parsing/codegen.sh
blob: 6c77eb6ed9f76b319088789fa420a67eeaaa2e93 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
sc=

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

toktext()
{
	local t="${1}"
	shift 1
	local n=

	case "${t%${US}*}" in
		# Operators
		T_EOF)       n='';;
		T_NEWLINE)   n="${LF}";;
		T_AND)       n='&';;
		T_SEMI)      n=';';;
		T_AND_IF)    n='&&';;
		T_OR_IF)     n='||';;
		T_DSEMI)     n=';;';;
		T_LESS)      n='<';;
		T_GREAT)     n='>';;
		T_DLESS)     n='<<';;
		T_DGREAT)    n='>>';;
		T_LESS)      n='<';;
		T_LESSAND)   n='<&';;
		T_GREAT)     n='>';;
		T_GREATAND)  n='>&';;
		T_LESSGREAT) n='<>';;
		T_DLESSDASH) n='<<-';;
		T_CLOBBER)   n='>|';;
		T_PIPE)      n='|';;
		T_LPAREN)    n='(';;
		T_RPAREN)    n=')';;
		# Reserved words
		T_IF)     n='if';;
		T_THEN)   n='then';;
		T_ELSE)   n='else';;
		T_ELIF)   n='elif';;
		T_FI)     n='fi';;
		T_DO)     n='do';;
		T_DONE)   n='done';;
		T_CASE)   n='case';;
		T_ESAC)   n='esac';;
		T_WHILE)  n='while';;
		T_UNTIL)  n='until';;
		T_FOR)    n='for';;
		T_LBRACE) n='{';;
		T_RBRACE) n='}';;
		T_BANG)   n='!';;
		T_IN)     n='in';;
		# Special symbols
		T_NAME)            n="${t#*${US}}";;
		T_FNAME)           n="${t#*${US}}";;
		T_CMDNAME)         n="${t#*${US}}";;
		T_IO_NUMBER)       n="${t#*${US}}";;
		T_WORD)            n="${t#*${US}}";;
		T_ASSIGNMENT_WORD) n="${t#*${US}}";;
		# Unknown
		*) n='';;
	esac

	printf '%s' "${n}"
}

codegen_sub()
{
	local array="${1}"
	shift 1

	IFS="${RS}"
	for t in ${array}; do
		toktext "${t}"
		case "${t%${US}*}" in
			T_NEWLINE)
				;;
			*)
				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.
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}$(parse_stack)."
						array="${array%.}"
						;;
				esac
				;;
			"${ETX}")
				# End of stack element
				break
				;;
			*)
				# Token character
				array="${array}${sc}"
				;;
		esac
	done
	codegen_sub "${array}"
}

codegen()
{
	local toks="${1}"
	shift 1

	if printf '%s' "${toks}" | parse_stack; then
		return 0
	else
		return 1
	fi
}