mirror of
https://github.com/RubyMetric/chsrc
synced 2026-10-27 02:47:14 +08:00
296 lines
8.2 KiB
C
296 lines
8.2 KiB
C
/* ------------------------------------------------------------
|
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
|
* -------------------------------------------------------------
|
|
* Lib Authors : Mikachu2333 <linkchou@yandex.com>
|
|
* Contributors : 曾奥然 <ccmywish@qq.com>
|
|
* |
|
|
* Created On : <2026-08-02>
|
|
* Last Modified : <2026-08-13>
|
|
*
|
|
*
|
|
* mikachu-toml.h: 由 Mikachu2333 编写的简易 TOML 解析库
|
|
*
|
|
* 这里只实现 chsrc 当前需要的 TOML 子集，不是完整 TOML 解析器：
|
|
* 以行定位表头和键，并支持 basic/literal string 的有限读取与写入
|
|
* ------------------------------------------------------------*/
|
|
|
|
#ifndef MIKACHU_TOML_H
|
|
#define MIKACHU_TOML_H
|
|
|
|
static bool
|
|
mktoml_key_prefix (const char *p, const char *key)
|
|
{
|
|
size_t kl = strlen (key);
|
|
const char *end = NULL;
|
|
if (strncmp (p, key, kl) == 0)
|
|
end = p + kl;
|
|
else if ((*p == '"' || *p == '\'')
|
|
&& strncmp (p + 1, key, kl) == 0 && p[kl + 1] == *p)
|
|
end = p + kl + 2;
|
|
else
|
|
return false;
|
|
|
|
char c = *end;
|
|
return c == '=' || c == ' ' || c == '\t' || c == '\n' || c == '\0';
|
|
}
|
|
|
|
static const char *
|
|
mktoml_skip_indent (const char *p)
|
|
{
|
|
while (*p == ' ' || *p == '\t') p++;
|
|
return p;
|
|
}
|
|
|
|
static bool
|
|
mktoml_header_match (const char *p, const char *header)
|
|
{
|
|
p = mktoml_skip_indent (p);
|
|
size_t hl = strlen (header);
|
|
if (strncmp (p, header, hl) == 0)
|
|
{
|
|
char c = p[hl];
|
|
return c == '\n' || c == '\r' || c == '\0' || c == ' ' || c == '\t' || c == '#';
|
|
}
|
|
if (strcmp (header, "[tool.uv]") == 0
|
|
&& (strncmp (p, "[tool.\"uv\"]", 11) == 0
|
|
|| strncmp (p, "[tool.'uv']", 11) == 0))
|
|
{
|
|
char c = p[11];
|
|
return c == '\n' || c == '\r' || c == '\0' || c == ' ' || c == '\t' || c == '#';
|
|
}
|
|
return false;
|
|
}
|
|
|
|
static const char *
|
|
mktoml_next_line (const char *p)
|
|
{
|
|
while (*p && *p != '\n') p++;
|
|
if (*p == '\n') p++;
|
|
return p;
|
|
}
|
|
|
|
static const char *
|
|
mktoml_find_section_end (const char *start)
|
|
{
|
|
const char *p = start;
|
|
while (*p)
|
|
{
|
|
if (*mktoml_skip_indent (p) == '[') return p;
|
|
p = mktoml_next_line (p);
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
static const char *
|
|
mktoml_find_table (const char *content, const char *header)
|
|
{
|
|
const char *p = content;
|
|
while (*p)
|
|
{
|
|
const char *line = mktoml_skip_indent (p);
|
|
if (*line == '[' && mktoml_header_match (line, header)) return p;
|
|
p = mktoml_next_line (p);
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
static const char *
|
|
mktoml_find_key_in_section (const char *first, const char *end, const char *key)
|
|
{
|
|
for (const char *s = first; *s && (!end || s < end); s = mktoml_next_line (s))
|
|
{
|
|
if (mktoml_key_prefix (mktoml_skip_indent (s), key)) return s;
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
static char *
|
|
mktoml_escape_basic_string (const char *value)
|
|
{
|
|
size_t n = 0;
|
|
for (const char *p = value; *p; p++)
|
|
{
|
|
switch (*p)
|
|
{
|
|
case '\\': case '"': case '\b': case '\t': case '\n':
|
|
case '\f': case '\r': n += 2; break;
|
|
default: n++;
|
|
}
|
|
}
|
|
|
|
char *ret = malloc (n + 1);
|
|
size_t pos = 0;
|
|
for (const char *p = value; *p; p++)
|
|
{
|
|
switch (*p)
|
|
{
|
|
case '\\': ret[pos++] = '\\'; ret[pos++] = '\\'; break;
|
|
case '"': ret[pos++] = '\\'; ret[pos++] = '"'; break;
|
|
case '\b': ret[pos++] = '\\'; ret[pos++] = 'b'; break;
|
|
case '\t': ret[pos++] = '\\'; ret[pos++] = 't'; break;
|
|
case '\n': ret[pos++] = '\\'; ret[pos++] = 'n'; break;
|
|
case '\f': ret[pos++] = '\\'; ret[pos++] = 'f'; break;
|
|
case '\r': ret[pos++] = '\\'; ret[pos++] = 'r'; break;
|
|
default: ret[pos++] = *p;
|
|
}
|
|
}
|
|
ret[pos] = '\0';
|
|
return ret;
|
|
}
|
|
|
|
static int
|
|
mktoml_hex_digit (char c)
|
|
{
|
|
if (c >= '0' && c <= '9') return c - '0';
|
|
if (c >= 'a' && c <= 'f') return c - 'a' + 10;
|
|
if (c >= 'A' && c <= 'F') return c - 'A' + 10;
|
|
return -1;
|
|
}
|
|
|
|
static bool
|
|
mktoml_append_utf8 (char **out, size_t *pos, unsigned long codepoint)
|
|
{
|
|
if (codepoint <= 0x7f)
|
|
(*out)[(*pos)++] = (char)codepoint;
|
|
else if (codepoint <= 0x7ff)
|
|
{
|
|
(*out)[(*pos)++] = (char)(0xc0 | (codepoint >> 6));
|
|
(*out)[(*pos)++] = (char)(0x80 | (codepoint & 0x3f));
|
|
}
|
|
else if (codepoint <= 0xffff)
|
|
{
|
|
if (codepoint >= 0xd800 && codepoint <= 0xdfff) return false;
|
|
(*out)[(*pos)++] = (char)(0xe0 | (codepoint >> 12));
|
|
(*out)[(*pos)++] = (char)(0x80 | ((codepoint >> 6) & 0x3f));
|
|
(*out)[(*pos)++] = (char)(0x80 | (codepoint & 0x3f));
|
|
}
|
|
else if (codepoint <= 0x10ffff)
|
|
{
|
|
(*out)[(*pos)++] = (char)(0xf0 | (codepoint >> 18));
|
|
(*out)[(*pos)++] = (char)(0x80 | ((codepoint >> 12) & 0x3f));
|
|
(*out)[(*pos)++] = (char)(0x80 | ((codepoint >> 6) & 0x3f));
|
|
(*out)[(*pos)++] = (char)(0x80 | (codepoint & 0x3f));
|
|
}
|
|
else return false;
|
|
return true;
|
|
}
|
|
|
|
static char *
|
|
mktoml_extract_string_value (const char *line)
|
|
{
|
|
const char *eq = strchr (mktoml_skip_indent (line), '=');
|
|
if (!eq) return NULL;
|
|
const char *v = eq + 1;
|
|
while (*v == ' ' || *v == '\t') v++;
|
|
if (*v != '"' && *v != '\'') return NULL;
|
|
char quote = *v++;
|
|
|
|
size_t max_len = strcspn (v, "\n") + 1;
|
|
char *ret = malloc (max_len * 4);
|
|
size_t pos = 0;
|
|
bool closed = false;
|
|
while (*v && *v != '\n')
|
|
{
|
|
if (*v == quote)
|
|
{
|
|
closed = true;
|
|
break;
|
|
}
|
|
if (quote == '"' && *v == '\\')
|
|
{
|
|
v++;
|
|
if (*v == 'u' || *v == 'U')
|
|
{
|
|
int digits = (*v == 'u') ? 4 : 8;
|
|
unsigned long codepoint = 0;
|
|
v++;
|
|
for (int i = 0; i < digits; i++, v++)
|
|
{
|
|
int digit = mktoml_hex_digit (*v);
|
|
if (digit < 0) { free (ret); return NULL; }
|
|
codepoint = (codepoint << 4) | (unsigned)digit;
|
|
}
|
|
if (!mktoml_append_utf8 (&ret, &pos, codepoint))
|
|
{ free (ret); return NULL; }
|
|
continue;
|
|
}
|
|
switch (*v)
|
|
{
|
|
case '"': ret[pos++] = '"'; break;
|
|
case '\\': ret[pos++] = '\\'; break;
|
|
case 'b': ret[pos++] = '\b'; break;
|
|
case 't': ret[pos++] = '\t'; break;
|
|
case 'n': ret[pos++] = '\n'; break;
|
|
case 'f': ret[pos++] = '\f'; break;
|
|
case 'r': ret[pos++] = '\r'; break;
|
|
default: free (ret); return NULL;
|
|
}
|
|
if (*v) v++;
|
|
continue;
|
|
}
|
|
ret[pos++] = *v++;
|
|
}
|
|
if (!closed)
|
|
{
|
|
free (ret);
|
|
return NULL;
|
|
}
|
|
ret[pos] = '\0';
|
|
return ret;
|
|
}
|
|
|
|
static bool
|
|
mktoml_value_is_true (const char *line)
|
|
{
|
|
const char *eq = strchr (mktoml_skip_indent (line), '=');
|
|
if (!eq) return false;
|
|
const char *v = mktoml_skip_indent (eq + 1);
|
|
return strncmp (v, "true", 4) == 0
|
|
&& (v[4] == '\0' || v[4] == '\r' || v[4] == '\n' || v[4] == ' ' || v[4] == '\t' || v[4] == '#');
|
|
}
|
|
|
|
static char *
|
|
mktoml_replace_line (const char *content, const char *line, const char *new_line_text)
|
|
{
|
|
const char *line_end = strchr (line, '\n');
|
|
if (!line_end) line_end = content + strlen (content);
|
|
|
|
const char *indent_end = mktoml_skip_indent (line);
|
|
size_t ind = indent_end - line;
|
|
|
|
size_t len = strlen (content) + ind + strlen (new_line_text) + 8;
|
|
char *ret = calloc (len, 1);
|
|
size_t pos = 0;
|
|
pos += snprintf (ret + pos, len - pos, "%.*s", (int)(line - content), content);
|
|
pos += snprintf (ret + pos, len - pos, "%.*s", (int)ind, line);
|
|
pos += snprintf (ret + pos, len - pos, "%s", new_line_text);
|
|
strcpy (ret + pos, line_end);
|
|
return ret;
|
|
}
|
|
|
|
static char *
|
|
mktoml_insert_before (const char *content, const char *insert_at, const char *insert_text)
|
|
{
|
|
size_t head = insert_at - content;
|
|
bool need_sep = (head > 0 && content[head - 1] != '\n');
|
|
|
|
size_t text_len = strlen (insert_text);
|
|
size_t len = strlen (content) + text_len + (need_sep ? 1 : 0) + 16;
|
|
char *ret = calloc (len, 1);
|
|
size_t pos = 0;
|
|
pos += snprintf (ret + pos, len - pos, "%.*s", (int)head, content);
|
|
if (need_sep) ret[pos++] = '\n';
|
|
memcpy (ret + pos, insert_text, text_len);
|
|
pos += text_len;
|
|
strcpy (ret + pos, content + head);
|
|
return ret;
|
|
}
|
|
|
|
static char *
|
|
mktoml_append_segment (const char *content, const char *segment)
|
|
{
|
|
return mktoml_insert_before (content, content + strlen (content), segment);
|
|
}
|
|
|
|
#endif
|