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
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
|
#pragma once
//BML v1.0 parser
//revision 0.04
namespace nall::BML {
//metadata is used to store nesting level
struct ManagedNode;
using SharedNode = shared_pointer<ManagedNode>;
struct ManagedNode : Markup::ManagedNode {
protected:
//test to verify if a valid character for a node name
auto valid(char p) const -> bool { //A-Z, a-z, 0-9, -.
return p - 'A' < 26u || p - 'a' < 26u || p - '0' < 10u || p - '-' < 2u;
}
//determine indentation level, without incrementing pointer
auto readDepth(const char* p) -> u32 {
u32 depth = 0;
while(p[depth] == '\t' || p[depth] == ' ') depth++;
return depth;
}
//determine indentation level
auto parseDepth(const char*& p) -> u32 {
u32 depth = readDepth(p);
p += depth;
return depth;
}
//read name
auto parseName(const char*& p) -> void {
u32 length = 0;
while(valid(p[length])) length++;
if(length == 0) throw "Invalid node name";
_name = slice(p, 0, length);
p += length;
}
auto parseData(const char*& p, string_view spacing) -> void {
if(*p == '=' && *(p + 1) == '\"') {
u32 length = 2;
while(p[length] && p[length] != '\n' && p[length] != '\"') length++;
if(p[length] != '\"') throw "Unescaped value";
_value = {slice(p, 2, length - 2), "\n"};
p += length + 1;
} else if(*p == '=') {
u32 length = 1;
while(p[length] && p[length] != '\n' && p[length] != '\"' && p[length] != ' ') length++;
if(p[length] == '\"') throw "Illegal character in value";
_value = {slice(p, 1, length - 1), "\n"};
p += length;
} else if(*p == ':') {
u32 length = 1;
while(p[length] && p[length] != '\n') length++;
_value = {slice(p, 1, length - 1).trimLeft(spacing, 1L), "\n"};
p += length;
}
}
//read all attributes for a node
auto parseAttributes(const char*& p, string_view spacing) -> void {
while(*p && *p != '\n') {
if(*p != ' ') throw "Invalid node name";
while(*p == ' ') p++; //skip excess spaces
if(*(p + 0) == '/' && *(p + 1) == '/') break; //skip comments
SharedNode node(new ManagedNode);
u32 length = 0;
while(valid(p[length])) length++;
if(length == 0) throw "Invalid attribute name";
node->_name = slice(p, 0, length);
node->parseData(p += length, spacing);
node->_value.trimRight("\n", 1L);
_children.append(node);
}
}
//read a node and all of its child nodes
auto parseNode(const vector<string>& text, u32& y, string_view spacing) -> void {
const char* p = text[y++];
_metadata = parseDepth(p);
parseName(p);
parseData(p, spacing);
parseAttributes(p, spacing);
while(y < text.size()) {
u32 depth = readDepth(text[y]);
if(depth <= _metadata) break;
if(text[y][depth] == ':') {
_value.append(slice(text[y++], depth + 1).trimLeft(spacing, 1L), "\n");
continue;
}
SharedNode node(new ManagedNode);
node->parseNode(text, y, spacing);
_children.append(node);
}
_value.trimRight("\n", 1L);
}
//read top-level nodes
auto parse(string document, string_view spacing) -> void {
//in order to simplify the parsing logic; we do an initial pass to normalize the data
//the below code will turn '\r\n' into '\n'; skip empty lines; and skip comment lines
char* p = document.get(), *output = p;
while(*p) {
char* origin = p;
bool empty = true;
while(*p) {
//scan for first non-whitespace character. if it's a line feed or comment; skip the line
if(p[0] == ' ' || p[0] == '\t') { p++; continue; }
empty = p[0] == '\r' || p[0] == '\n' || (p[0] == '/' && p[1] == '/');
break;
}
while(*p) {
if(p[0] == '\r') p[0] = '\n'; //turns '\r\n' into '\n\n' (second '\n' will be skipped)
if(*p++ == '\n') break; //include '\n' in the output to be copied
}
if(empty) continue;
memory::move(output, origin, p - origin);
output += p - origin;
}
document.resize(document.size() - (p - output)).trimRight("\n");
if(document.size() == 0) return; //empty document
auto text = document.split("\n");
u32 y = 0;
while(y < text.size()) {
SharedNode node(new ManagedNode);
node->parseNode(text, y, spacing);
if(node->_metadata > 0) throw "Root nodes cannot be indented";
_children.append(node);
}
}
friend auto unserialize(const string&, string_view) -> Markup::Node;
};
inline auto unserialize(const string& markup, string_view spacing = {}) -> Markup::Node {
SharedNode node(new ManagedNode);
try {
node->parse(markup, spacing);
} catch(const char* error) {
node.reset();
}
return (Markup::SharedNode&)node;
}
inline auto serialize(const Markup::Node& node, string_view spacing = {}, u32 depth = 0) -> string {
if(!node.name()) {
string result;
for(auto leaf : node) {
result.append(serialize(leaf, spacing, depth));
}
return result;
}
string padding;
padding.resize(depth * 2);
padding.fill(' ');
vector<string> lines;
if(auto value = node.value()) lines = value.split("\n");
string result;
result.append(padding);
result.append(node.name());
if(lines.size() == 1) result.append(":", spacing, lines[0]);
result.append("\n");
if(lines.size() > 1) {
padding.append(" ");
for(auto& line : lines) {
result.append(padding, ":", spacing, line, "\n");
}
}
for(auto leaf : node) {
result.append(serialize(leaf, spacing, depth + 1));
}
return result;
}
}
|
