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
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
|
// Logic for parsing text literals
#include <stdarg.h>
#include <stdbool.h>
#include <string.h>
#include "../unistr-fixed.h"
#include <unictype.h>
#include <uniname.h>
#include "../ast.h"
#include "../stdlib/text.h"
#include "../util.h"
#include "context.h"
#include "errors.h"
#include "expressions.h"
#include "types.h"
#include "utils.h"
static ast_list_t *_parse_text_helper(parse_ctx_t *ctx, const char **out_pos, bool allow_interps) {
const char *pos = *out_pos;
int64_t starting_indent = get_indent(ctx, pos);
int64_t string_indent = starting_indent + SPACES_PER_INDENT;
const char *quote, *interp;
bool allow_escapes = true;
if (match(&pos, "\"\"\"")) { // Triple double quote
quote = "\"\"\"", interp = "$", allow_escapes = false;
} else if (match(&pos, "'''")) { // Triple single quote
quote = "'''", interp = "$", allow_escapes = false;
} else if (match(&pos, "```")) { // Triple backtick
quote = "```", interp = "@", allow_escapes = false;
} else if (match(&pos, "\"")) { // Double quote
quote = "\"", interp = "$", allow_escapes = true;
} else if (match(&pos, "'")) { // Single quote
quote = "'", interp = "$", allow_escapes = true;
} else if (match(&pos, "`")) { // Backtick
quote = "`", interp = "@", allow_escapes = true;
} else {
parser_err(ctx, pos, pos, "I expected a valid text here");
}
if (!allow_interps) interp = NULL;
ast_list_t *chunks = NULL;
Text_t chunk = EMPTY_TEXT;
const char *chunk_start = pos;
bool leading_newline = false;
int64_t plain_span_len = 0;
#define FLUSH_PLAIN_SPAN() \
do { \
if (plain_span_len > 0) { \
chunk = Texts(chunk, Text$from_strn(pos - plain_span_len, (size_t)plain_span_len)); \
plain_span_len = 0; \
} \
} while (0)
for (const char *end = ctx->file->text + ctx->file->len; pos < end;) {
const char *after_indentation = pos;
const char *interp_start = pos;
if (interp != NULL && strncmp(pos, interp, strlen(interp)) == 0) { // Interpolation
FLUSH_PLAIN_SPAN();
if (chunk.length > 0) {
ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text = chunk);
chunks = new (ast_list_t, .ast = literal, .next = chunks);
chunk = EMPTY_TEXT;
}
pos += strlen(interp);
if (*pos == ' ' || *pos == '\t')
parser_err(ctx, pos, pos + 1, "Whitespace is not allowed before an interpolation here");
ast_t *value =
expect(ctx, interp_start, &pos, parse_term_no_suffix, "I expected an interpolation term here");
chunks = new (ast_list_t, .ast = value, .next = chunks);
chunk_start = pos;
} else if (allow_escapes && *pos == '\\') {
FLUSH_PLAIN_SPAN();
const char *c = unescape(ctx, &pos);
chunk = Texts(chunk, Text$from_str(c));
} else if (!leading_newline && strncmp(pos, quote, strlen(quote)) == 0) { // Nested pair end
if (get_indent(ctx, pos) == starting_indent) break;
plain_span_len += 1;
++pos;
} else if (newline_with_indentation(&after_indentation, string_indent)) { // Newline
FLUSH_PLAIN_SPAN();
pos = after_indentation;
if (!leading_newline && !(chunk.length > 0 || chunks)) {
leading_newline = true;
} else {
chunk = Texts(chunk, Text("\n"));
}
} else if (newline_with_indentation(&after_indentation, starting_indent)) { // Line continuation (..)
FLUSH_PLAIN_SPAN();
pos = after_indentation;
if (strncmp(pos, quote, strlen(quote)) == 0) {
break;
} else if (some_of(&pos, ".") >= 2) {
// Multi-line split
continue;
} else {
parser_err(ctx, pos, eol(pos),
"This multi-line string should be either indented or have '..' at the front");
}
} else { // Plain character
ucs4_t codepoint;
const char *next = (const char *)u8_next(&codepoint, (const uint8_t *)pos);
plain_span_len += (int64_t)(next - pos);
if (next == NULL) break;
pos = next;
}
}
FLUSH_PLAIN_SPAN();
#undef FLUSH_PLAIN_SPAN
expect_closing(ctx, &pos, quote, "I was expecting a ", quote, " to finish this string");
if (chunk.length > 0) {
ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text = chunk);
chunks = new (ast_list_t, .ast = literal, .next = chunks);
}
REVERSE_LIST(chunks);
*out_pos = pos;
return chunks;
}
ast_t *parse_text(parse_ctx_t *ctx, const char *pos, bool allow_interps) {
// ('"' ... '"' / "'" ... "'" / "`" ... "`")
// "$" [name] quote-char ... close-quote
const char *start = pos;
const char *lang = NULL;
if (match(&pos, "$")) {
lang = get_id(&pos);
if (lang == NULL) parser_err(ctx, start, pos, "I expected a language name after the `$`");
}
if (!(*pos == '"' || *pos == '\'' || *pos == '`')) return NULL;
ast_list_t *chunks = _parse_text_helper(ctx, &pos, allow_interps);
bool colorize = match(&pos, "~") && match_word(&pos, "colorized");
return NewAST(ctx->file, start, pos, TextJoin, .lang = lang, .children = chunks, .colorize = colorize);
}
ast_t *parse_inline_c(parse_ctx_t *ctx, const char *pos) {
const char *start = pos;
if (!match_word(&pos, "C_code")) return NULL;
spaces(&pos);
type_ast_t *type = NULL;
if (match(&pos, ":")) {
type = expect(ctx, start, &pos, parse_type, "I couldn't parse the type for this C_code code");
spaces(&pos);
}
static const char *quote_chars = "\"'`|/;([{<";
if (!strchr(quote_chars, *pos))
parser_err(ctx, pos, pos + 1,
"This is not a valid string quotation character. Valid characters are: \"'`|/;([{<");
ast_list_t *chunks = _parse_text_helper(ctx, &pos, true);
return NewAST(ctx->file, start, pos, InlineCCode, .chunks = chunks, .type_ast = type);
}
ast_t *parse_path(parse_ctx_t *ctx, const char *pos) {
// "(" ("~/" / "./" / "../" / "/") ... ")"
const char *start = pos;
if (!match(&pos, "(")) return NULL;
if (!(*pos == '~' || *pos == '.' || *pos == '/')) return NULL;
const char *path_start = pos;
size_t len = 1;
int paren_depth = 1;
while (pos + len < ctx->file->text + ctx->file->len - 1) {
if (pos[len] == '\\') {
len += 2;
continue;
} else if (pos[len] == '(') {
paren_depth += 1;
} else if (pos[len] == ')') {
paren_depth -= 1;
if (paren_depth <= 0) break;
} else if (pos[len] == '\r' || pos[len] == '\n') {
parser_err(ctx, path_start, &pos[len - 1], "This path was not closed");
}
len += 1;
}
pos += len + 1;
char *path = String(string_slice(path_start, .length = len));
for (char *src = path, *dest = path;;) {
if (src[0] == '\\') {
*(dest++) = src[1];
src += 2;
} else if (*src) {
*(dest++) = *(src++);
} else {
*(dest++) = '\0';
break;
}
}
return NewAST(ctx->file, start, pos, Path, .path = path);
}
|
