aboutsummaryrefslogtreecommitdiff
path: root/parse.c
diff options
context:
space:
mode:
Diffstat (limited to 'parse.c')
-rw-r--r--parse.c74
1 files changed, 38 insertions, 36 deletions
diff --git a/parse.c b/parse.c
index 73c5dc1d..0f32e021 100644
--- a/parse.c
+++ b/parse.c
@@ -7,6 +7,7 @@
#include <stdbool.h>
#include <string.h>
#include <unistr.h>
+#include <unigbrk.h>
#include <unictype.h>
#include <uniname.h>
#include <signal.h>
@@ -1283,13 +1284,12 @@ PARSER(parse_text) {
// Escape sequence, e.g. \r\n
if (*pos == '\\') {
- CORD cord = CORD_EMPTY;
+ Text_t text = Text("");
do {
const char *c = unescape(ctx, &pos);
- cord = CORD_cat(cord, c);
- // cord = CORD_cat_char(cord, c);
+ text = Texts(text, Text$from_str(c));
} while (*pos == '\\');
- return NewAST(ctx->file, start, pos, TextLiteral, .cord=cord);
+ return NewAST(ctx->file, start, pos, TextLiteral, .text=text);
}
char open_quote, close_quote, open_interp = '$';
@@ -1330,17 +1330,17 @@ PARSER(parse_text) {
int64_t string_indent = starting_indent + 1;
ast_list_t *chunks = NULL;
- CORD chunk = CORD_EMPTY;
+ Text_t chunk = Text("");
const char *chunk_start = pos;
int depth = 1;
bool leading_newline = false;
for (; pos < ctx->file->text + ctx->file->len && depth > 0; ) {
if (*pos == open_interp) { // Interpolation
const char *interp_start = pos;
- if (chunk) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk);
+ if (chunk.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=Text$format("%k", &chunk)); // Collapse text
chunks = new(ast_list_t, .ast=literal, .next=chunks);
- chunk = NULL;
+ chunk = Text("");
}
++pos;
ast_t *interp;
@@ -1353,7 +1353,7 @@ PARSER(parse_text) {
if (get_indent(ctx, pos) == starting_indent) {
++depth;
}
- chunk = CORD_cat_char(chunk, *pos);
+ chunk = Texts(chunk, Text$format("%c", *pos));
++pos;
} else if (!leading_newline && *pos == close_quote) { // Nested pair end
if (get_indent(ctx, pos) == starting_indent) {
@@ -1361,13 +1361,13 @@ PARSER(parse_text) {
if (depth == 0)
break;
}
- chunk = CORD_cat_char(chunk, *pos);
+ chunk = Texts(chunk, Text$format("%c", *pos));
++pos;
} else if (newline_with_indentation(&pos, string_indent)) { // Newline
- if (!leading_newline && !(chunk || chunks)) {
+ if (!leading_newline && !(chunk.length > 0 || chunks)) {
leading_newline = true;
} else {
- chunk = CORD_cat_char(chunk, '\n');
+ chunk = Texts(chunk, Text("\n"));
}
} else if (newline_with_indentation(&pos, starting_indent)) { // Line continuation (..)
if (*pos == close_quote) {
@@ -1378,16 +1378,19 @@ PARSER(parse_text) {
} else {
parser_err(ctx, pos, strchrnul(pos, '\n'), "This multi-line string should be either indented or have '..' at the front");
}
- } else { // Plain character
- chunk = CORD_cat_char(chunk, *pos);
- ++pos;
+ } else { // Regular grapheme cluster (no escapes etc.)
+ char *next = (char*)u8_grapheme_next((const uint8_t*)pos, (const uint8_t*)ctx->file->text + ctx->file->len);
+ while (next < ctx->file->text + ctx->file->len && !isascii(*next))
+ next = (char*)u8_grapheme_next((const uint8_t*)next, (const uint8_t*)ctx->file->text + ctx->file->len);
+ Text_t cluster = Text$from_strn(pos, (size_t)(next-pos));
+ chunk = Texts(chunk, cluster);
+ pos = next;
}
}
- if (chunk) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk);
+ if (chunk.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=Text$format("%k", &chunk)); // Collapse text
chunks = new(ast_list_t, .ast=literal, .next=chunks);
- chunk = NULL;
}
REVERSE_LIST(chunks);
@@ -1407,13 +1410,13 @@ PARSER(parse_path) {
const char *chunk_start = start + 1;
ast_list_t *chunks = NULL;
- CORD chunk_text = CORD_EMPTY;
+ Text_t chunk_text = Text("");
int paren_depth = 1;
while (pos < ctx->file->text + ctx->file->len) {
switch (*pos) {
case '\\': {
++pos;
- chunk_text = CORD_asprintf("%r%.*s%c", chunk_text, (size_t)(pos - chunk_start), chunk_start, *pos);
+ chunk_text = Text$format("%k%.*s%c", &chunk_text, (size_t)(pos - chunk_start), chunk_start, *pos);
++pos;
chunk_start = pos;
continue;
@@ -1422,12 +1425,12 @@ PARSER(parse_path) {
const char *interp_start = pos;
if (pos > chunk_start)
- chunk_text = CORD_asprintf("%r%.*s", chunk_text, (size_t)(pos - chunk_start), chunk_start);
+ chunk_text = Text$format("%k%.*s", &chunk_text, (size_t)(pos - chunk_start), chunk_start);
- if (chunk_text) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk_text);
+ if (chunk_text.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=chunk_text);
chunks = new(ast_list_t, .ast=literal, .next=chunks);
- chunk_text = CORD_EMPTY;
+ chunk_text = Text("");
}
++pos;
if (*pos == ' ' || *pos == '\t')
@@ -1455,10 +1458,10 @@ PARSER(parse_path) {
end_of_path:;
if (pos > chunk_start)
- chunk_text = CORD_asprintf("%r%.*s", chunk_text, (size_t)(pos - chunk_start), chunk_start);
+ chunk_text = Text$format("%k%.*s", &chunk_text, (size_t)(pos - chunk_start), chunk_start);
- if (chunk_text != CORD_EMPTY) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk_text);
+ if (chunk_text.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=chunk_text);
chunks = new(ast_list_t, .ast=literal, .next=chunks);
}
@@ -2283,7 +2286,7 @@ PARSER(parse_inline_c) {
if (depth != 0)
parser_err(ctx, start, start+1, "I couldn't find the closing '}' for this inline C code");
- CORD c_code = GC_strndup(c_code_start, (size_t)((pos-1) - c_code_start));
+ Text_t c_code = Text$format("%.*s", (size_t)((pos-1) - c_code_start), c_code_start);
return NewAST(ctx->file, start, pos, InlineCCode, .code=c_code, .type_ast=type);
}
@@ -2322,16 +2325,16 @@ PARSER(parse_say) {
spaces(&pos);
ast_list_t *chunks = NULL;
- CORD chunk = CORD_EMPTY;
+ Text_t chunk = Text("");
const char *chunk_start = pos;
const char open_interp = '$';
while (pos < ctx->file->text + ctx->file->len) {
if (*pos == open_interp) { // Interpolation
const char *interp_start = pos;
- if (chunk) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk);
+ if (chunk.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=Text$format("%k", &chunk));
chunks = new(ast_list_t, .ast=literal, .next=chunks);
- chunk = NULL;
+ chunk = Text("");
}
++pos;
ast_t *interp;
@@ -2343,15 +2346,14 @@ PARSER(parse_say) {
} else if (*pos == '\r' || *pos == '\n') { // Newline
break;
} else { // Plain character
- chunk = CORD_cat_char(chunk, *pos);
+ chunk = Texts(chunk, Text$format("%c", *pos));
++pos;
}
}
- if (chunk) {
- ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .cord=chunk);
+ if (chunk.length > 0) {
+ ast_t *literal = NewAST(ctx->file, chunk_start, pos, TextLiteral, .text=Text$format("%k", &chunk));
chunks = new(ast_list_t, .ast=literal, .next=chunks);
- chunk = NULL;
}
REVERSE_LIST(chunks);
generated by cgit v1.2.3 (git 2.25.1) at 2026-03-01 04:47:43 +0000