OXYGENHUB
◆ main openmsx / src / events / AdhocCliCommParser.cc
4.1 KiB · C++ · last changed chore(): initial commit 1 month ago source blame history raw
1 
#include "AdhocCliCommParser.hh"
2
3 
#include "utf8_unchecked.hh"
4
5 
AdhocCliCommParser::AdhocCliCommParser(std::function<void(const std::string&)> callback_)
6 
	: callback(std::move(callback_))
7 
{
8 
}
9
10 
void AdhocCliCommParser::parse(std::span<const char> buf)
11 
{
12 
	for (auto c : buf) {
13 
		parse(c);
14 
	}
15 
}
16
17 
void AdhocCliCommParser::parse(char c)
18 
{
19 
	// Whenever there is a parse error we return to the initial state
20 
	switch (state) {
21 
	case O0: // looking for opening tag
22 
		state = (c == '<') ? O1 : O0; break;
23 
	case O1: // matched <
24 
		state = (c == 'c') ? O2 : O0; break;
25 
	case O2: // matched <c
26 
		state = (c == 'o') ? O3 : O0; break;
27 
	case O3: // matched <co
28 
		state = (c == 'm') ? O4 : O0; break;
29 
	case O4: // matched <com
30 
		state = (c == 'm') ? O5 : O0; break;
31 
	case O5: // matched <comm
32 
		state = (c == 'a') ? O6 : O0; break;
33 
	case O6: // matched <comma
34 
		state = (c == 'n') ? O7 : O0; break;
35 
	case O7: // matched <comman
36 
		state = (c == 'd') ? O8 : O0; break;
37 
	case O8: // matched <command
38 
		if (c == '>') {
39 
			state = C0;
40 
			command.clear();
41 
		} else {
42 
			state = O0;
43 
		}
44 
		break;
45 
	case C0: // matched <command>, now parsing xml entities and </command>
46 
		if      (c == '<') state = C1;
47 
		else if (c == '&') state = A1;
48 
		else command += c;
49 
		break;
50 
	case C1: // matched <
51 
		state = (c == '/') ? C2 : O0; break;
52 
	case C2: // matched </
53 
		state = (c == 'c') ? C3 : O0; break;
54 
	case C3: // matched </c
55 
		state = (c == 'o') ? C4 : O0; break;
56 
	case C4: // matched </co
57 
		state = (c == 'm') ? C5 : O0; break;
58 
	case C5: // matched </com
59 
		state = (c == 'm') ? C6 : O0; break;
60 
	case C6: // matched </comm
61 
		state = (c == 'a') ? C7 : O0; break;
62 
	case C7: // matched </comma
63 
		state = (c == 'n') ? C8 : O0; break;
64 
	case C8: // matched </comman
65 
		state = (c == 'd') ? C9 : O0; break;
66 
	case C9: // matched </command
67 
		if (c == '>') callback(command);
68 
		state = O0;
69 
		break;
70 
	case A1: // matched &
71 
		if      (c == 'l') state = L2;
72 
		else if (c == 'a') state = A2;
73 
		else if (c == 'g') state = G2;
74 
		else if (c == 'q') state = Q2;
75 
		else if (c == '#') { state = H2; unicode = 0; }
76 
		else               state = O0; // error
77 
		break;
78 
	case A2: // matched &a
79 
		if      (c == 'm') state = A3;
80 
		else if (c == 'p') state = P3;
81 
		else               state = O0; // error
82 
		break;
83 
	case A3: // matched &am
84 
		state = (c == 'p') ? A4 : O0; break;
85 
	case A4: // matched &amp
86 
		if (c == ';') {
87 
			command += '&';
88 
			state = C0;
89 
		} else {
90 
			state = O0; // error
91 
		}
92 
		break;
93 
	case P3: // matched &ap
94 
		state = (c == 'o') ? P4 : O0; break;
95 
	case P4: // matched &apo
96 
		state = (c == 's') ? P5 : O0; break;
97 
	case P5: // matched &apos
98 
		if (c == ';') {
99 
			command += '\'';
100 
			state = C0;
101 
		} else {
102 
			state = O0; // error
103 
		}
104 
		break;
105 
	case Q2: // matched &q
106 
		state = (c == 'u') ? Q3 : O0; break;
107 
	case Q3: // matched &qu
108 
		state = (c == 'o') ? Q4 : O0; break;
109 
	case Q4: // matched &quo
110 
		state = (c == 't') ? Q5 : O0; break;
111 
	case Q5: // matched &quot
112 
		if (c == ';') {
113 
			command += '"';
114 
			state = C0;
115 
		} else {
116 
			state = O0; // error
117 
		}
118 
		break;
119 
	case G2: // matched &g
120 
		state = (c == 't') ? G3 : O0; break;
121 
	case G3: // matched &gt
122 
		if (c == ';') {
123 
			command += '>';
124 
			state = C0;
125 
		} else {
126 
			state = O0; // error
127 
		}
128 
		break;
129 
	case L2: // matched &l
130 
		state = (c == 't') ? L3 : O0; break;
131 
	case L3: // matched &lt
132 
		if (c == ';') {
133 
			command += '<';
134 
			state = C0;
135 
		} else {
136 
			state = O0; // error
137 
		}
138 
		break;
139 
	case H2: // matched &#
140 
		// This also parses invalid input like '&#12xab;' but let's
141 
		// ignore that. It also doesn't check for overflow etc.
142 
		if (c == ';') {
143 
			utf8::unchecked::append(unicode, back_inserter(command));
144 
			state = C0;
145 
		} else if (c == 'x') {
146 
			state = H3;
147 
		} else {
148 
			unicode *= 10;
149 
			if (('0' <= c) && (c <= '9')) unicode += c - '0';
150 
			else state = O0;
151 
		}
152 
		break;
153 
	case H3: // matched &#x
154 
		if (c == ';') {
155 
			utf8::unchecked::append(unicode, back_inserter(command));
156 
			state = C0;
157 
		} else {
158 
			unicode *= 16;
159 
			if      (('0' <= c) && (c <= '9')) unicode += c - '0';
160 
			else if (('a' <= c) && (c <= 'f')) unicode += c - 'a' + 10;
161 
			else if (('A' <= c) && (c <= 'F')) unicode += c - 'A' + 10;
162 
			else state = O0;
163 
		}
164 
		break;
165 
	}
166 
}