summaryrefslogtreecommitdiff
path: root/core/src/xmake/io/file_open.c
diff options
context:
space:
mode:
Diffstat (limited to 'core/src/xmake/io/file_open.c')
-rw-r--r--core/src/xmake/io/file_open.c215
1 files changed, 110 insertions, 105 deletions
diff --git a/core/src/xmake/io/file_open.c b/core/src/xmake/io/file_open.c
index 0cd9c6cb6..901a3f86d 100644
--- a/core/src/xmake/io/file_open.c
+++ b/core/src/xmake/io/file_open.c
@@ -22,8 +22,8 @@
/* //////////////////////////////////////////////////////////////////////////////////////
* trace
*/
-#define TB_TRACE_MODULE_NAME "file_open"
-#define TB_TRACE_MODULE_DEBUG (0)
+#define TB_TRACE_MODULE_NAME "file_open"
+#define TB_TRACE_MODULE_DEBUG (0)
/* //////////////////////////////////////////////////////////////////////////////////////
* includes
@@ -35,46 +35,40 @@
*/
// num of bytes read to guess encoding
-#define CHECK_SIZE (1024)
+#define CHECK_SIZE (1024)
// is utf-8 tail character
-#define IS_UTF8_TAIL(c) (c >= 0x80 && c < 0xc0)
+#define IS_UTF8_TAIL(c) (c >= 0x80 && c < 0xc0)
/* //////////////////////////////////////////////////////////////////////////////////////
* private implementation
*/
-static tb_size_t xm_io_file_detect_charset(tb_byte_t const** data_ptr, tb_long_t size)
-{
- // check
+static tb_size_t xm_io_file_detect_charset(tb_byte_t const **data_ptr, tb_long_t size) {
tb_assert(data_ptr && *data_ptr);
- tb_byte_t const* data = *data_ptr;
+ tb_byte_t const *data = *data_ptr;
tb_size_t charset = XM_IO_FILE_ENCODING_BINARY;
- do
- {
+ do {
// is luajit bitcode? open as binary
- if (size >= 3 && data[0] == 27 && data[1] == 'L' && data[2] == 'J')
+ if (size >= 3 && data[0] == 27 && data[1] == 'L' && data[2] == 'J') {
break;
+ }
// utf-8 with bom
- if (size >= 3 && data[0] == 239 && data[1] == 187 && data[2] == 191)
- {
+ if (size >= 3 && data[0] == 239 && data[1] == 187 && data[2] == 191) {
charset = TB_CHARSET_TYPE_UTF8;
data += 3; // skip bom
break;
}
- if (size >= 2)
- {
+ if (size >= 2) {
// utf16be
- if (data[0] == 254 && data[1] == 255)
- {
+ if (data[0] == 254 && data[1] == 255) {
charset = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_BE;
data += 2; // skip bom
break;
}
// utf16le
- else if (data[0] == 255 && data[1] == 254)
- {
+ else if (data[0] == 255 && data[1] == 254) {
charset = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_LE;
data += 2; // skip bom
break;
@@ -86,57 +80,59 @@ static tb_size_t xm_io_file_detect_charset(tb_byte_t const** data_ptr, tb_long_t
tb_sint16_t utf8_conf = 0;
tb_sint16_t ascii_conf = 0;
tb_sint16_t zero_count = 0;
- for (tb_long_t i = 0; i < (size - 4) && i < CHECK_SIZE; i++)
- {
- if (data[i] == 0) zero_count++;
+ for (tb_long_t i = 0; i < (size - 4) && i < CHECK_SIZE; i++) {
+ if (data[i] == 0) {
+ zero_count++;
+ }
- if (data[i] < 0x80)
+ if (data[i] < 0x80) {
ascii_conf++;
- else
+ } else {
ascii_conf = TB_MINS16;
+ }
- if (i % 2 == 0)
- {
- if (data[i] == 0) utf16be_conf++;
- if (data[i + 1] == 0) utf16le_conf++;
+ if (i % 2 == 0) {
+ if (data[i] == 0) {
+ utf16be_conf++;
+ }
+ if (data[i + 1] == 0) {
+ utf16le_conf++;
+ }
}
- if (IS_UTF8_TAIL(data[i]))
- ;
- else if (data[i] < 0x80)
+ if (IS_UTF8_TAIL(data[i])) {
+ // continue
+ } else if (data[i] < 0x80) {
utf8_conf++;
- else if (data[i] >= 0xc0 && data[i] < 0xe0 && IS_UTF8_TAIL(data[i + 1]))
+ } else if (data[i] >= 0xc0 && data[i] < 0xe0 && IS_UTF8_TAIL(data[i + 1])) {
utf8_conf++;
- else if (data[i] >= 0xe0 && data[i] < 0xf0 && IS_UTF8_TAIL(data[i + 1]) && IS_UTF8_TAIL(data[i + 2]))
+ } else if (data[i] >= 0xe0 && data[i] < 0xf0 && IS_UTF8_TAIL(data[i + 1]) && IS_UTF8_TAIL(data[i + 2])) {
utf8_conf++;
- else if (data[i] >= 0xf0 && data[i] < 0xf8 && IS_UTF8_TAIL(data[i + 1]) && IS_UTF8_TAIL(data[i + 2]) && IS_UTF8_TAIL(data[i + 3]))
+ } else if (data[i] >= 0xf0 && data[i] < 0xf8 && IS_UTF8_TAIL(data[i + 1]) && IS_UTF8_TAIL(data[i + 2]) &&
+ IS_UTF8_TAIL(data[i + 3])) {
utf8_conf++;
- else
+ } else {
utf8_conf = TB_MINS16;
+ }
}
- if (ascii_conf > 0 && zero_count <= 1)
- {
+ if (ascii_conf > 0 && zero_count <= 1) {
charset = TB_CHARSET_TYPE_UTF8;
break;
}
- if (utf8_conf > 0 && zero_count <= 1)
- {
+ if (utf8_conf > 0 && zero_count <= 1) {
charset = TB_CHARSET_TYPE_UTF8;
break;
}
- if (utf16be_conf > 0 && utf16be_conf > utf16le_conf)
- {
+ if (utf16be_conf > 0 && utf16be_conf > utf16le_conf) {
charset = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_BE;
break;
}
- if (utf16le_conf > 0 && utf16le_conf >= utf16be_conf)
- {
+ if (utf16le_conf > 0 && utf16le_conf >= utf16be_conf) {
charset = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_LE;
break;
}
- if (utf8_conf > 0)
- {
+ if (utf8_conf > 0) {
charset = TB_CHARSET_TYPE_UTF8;
break;
}
@@ -150,20 +146,17 @@ static tb_size_t xm_io_file_detect_charset(tb_byte_t const** data_ptr, tb_long_t
*data_ptr = data;
return charset;
}
-static tb_size_t xm_io_file_detect_encoding(tb_stream_ref_t stream, tb_long_t* pbomoff)
-{
- // check
+static tb_size_t xm_io_file_detect_encoding(tb_stream_ref_t stream, tb_long_t *pbomoff) {
tb_assert_and_check_return_val(stream && pbomoff, XM_IO_FILE_ENCODING_BINARY);
// detect encoding
- tb_byte_t* data = tb_null;
- tb_size_t encoding = XM_IO_FILE_ENCODING_BINARY;
- tb_long_t size = tb_stream_peek(stream, &data, CHECK_SIZE);
- if (size > 0)
- {
- tb_byte_t const* p = data;
- encoding = xm_io_file_detect_charset(&p, size);
- *pbomoff = p - data;
+ tb_byte_t *data = tb_null;
+ tb_size_t encoding = XM_IO_FILE_ENCODING_BINARY;
+ tb_long_t size = tb_stream_peek(stream, &data, CHECK_SIZE);
+ if (size > 0) {
+ tb_byte_t const *p = data;
+ encoding = xm_io_file_detect_charset(&p, size);
+ *pbomoff = p - data;
}
return encoding;
}
@@ -173,114 +166,126 @@ static tb_size_t xm_io_file_detect_encoding(tb_stream_ref_t stream, tb_long_t* p
*/
// io.file_open(path, modestr)
-tb_int_t xm_io_file_open(lua_State* lua)
-{
- // check
+tb_int_t xm_io_file_open(lua_State *lua) {
tb_assert_and_check_return_val(lua, 0);
// get file path and mode
- tb_char_t const* path = luaL_checkstring(lua, 1);
- tb_char_t const* modestr = luaL_optstring(lua, 2, "r");
+ tb_char_t const *path = luaL_checkstring(lua, 1);
+ tb_char_t const *modestr = luaL_optstring(lua, 2, "r");
tb_assert_and_check_return_val(path && modestr, 0);
// get file mode value
tb_size_t mode;
- switch (modestr[0])
- {
- case 'w': mode = TB_FILE_MODE_RW | TB_FILE_MODE_CREAT | TB_FILE_MODE_TRUNC; break;
- case 'a': mode = TB_FILE_MODE_RW | TB_FILE_MODE_APPEND | TB_FILE_MODE_CREAT; break;
- case 'r': default: mode = TB_FILE_MODE_RO; break;
+ switch (modestr[0]) {
+ case 'w':
+ mode = TB_FILE_MODE_RW | TB_FILE_MODE_CREAT | TB_FILE_MODE_TRUNC;
+ break;
+ case 'a':
+ mode = TB_FILE_MODE_RW | TB_FILE_MODE_APPEND | TB_FILE_MODE_CREAT;
+ break;
+ case 'r':
+ default:
+ mode = TB_FILE_MODE_RO;
+ break;
}
// get file encoding
- tb_long_t bomoff = 0;
- tb_stream_ref_t stream = tb_null;
- tb_bool_t update = !!tb_strchr(modestr, '+');
+ tb_long_t bomoff = 0;
+ tb_stream_ref_t stream = tb_null;
+ tb_bool_t update = !!tb_strchr(modestr, '+');
tb_size_t encoding = XM_IO_FILE_ENCODING_UNKNOWN;
- if (modestr[1] == 'b' || (update && modestr[2] == 'b'))
+ if (modestr[1] == 'b' || (update && modestr[2] == 'b')) {
encoding = XM_IO_FILE_ENCODING_BINARY;
- else if (tb_strstr(modestr, "utf8") || tb_strstr(modestr, "utf-8"))
+ } else if (tb_strstr(modestr, "utf8") || tb_strstr(modestr, "utf-8")) {
encoding = TB_CHARSET_TYPE_UTF8;
- else if (tb_strstr(modestr, "utf16le") || tb_strstr(modestr, "utf-16le"))
+ } else if (tb_strstr(modestr, "utf16le") || tb_strstr(modestr, "utf-16le")) {
encoding = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_LE;
- else if (tb_strstr(modestr, "utf16be") || tb_strstr(modestr, "utf-16be"))
+ } else if (tb_strstr(modestr, "utf16be") || tb_strstr(modestr, "utf-16be")) {
encoding = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_BE;
- else if (tb_strstr(modestr, "utf16") || tb_strstr(modestr, "utf-16"))
+ } else if (tb_strstr(modestr, "utf16") || tb_strstr(modestr, "utf-16")) {
encoding = TB_CHARSET_TYPE_UTF16 | TB_CHARSET_TYPE_NE;
- else if (tb_strstr(modestr, "ansi"))
+ } else if (tb_strstr(modestr, "ansi")) {
encoding = TB_CHARSET_TYPE_ANSI;
- else if (tb_strstr(modestr, "gbk"))
+ } else if (tb_strstr(modestr, "gbk")) {
encoding = TB_CHARSET_TYPE_GBK;
- else if (tb_strstr(modestr, "gb2312"))
+ } else if (tb_strstr(modestr, "gb2312")) {
encoding = TB_CHARSET_TYPE_GB2312;
- else if (tb_strstr(modestr, "iso8859"))
+ } else if (tb_strstr(modestr, "iso8859")) {
encoding = TB_CHARSET_TYPE_ISO8859;
- else if (modestr[0] == 'w' || modestr[0] == 'a') // set to utf-8 if not specified for the writing mode
+ } else if (modestr[0] == 'w' || modestr[0] == 'a') { // set to utf-8 if not specified for the writing mode
encoding = TB_CHARSET_TYPE_UTF8;
- else if (modestr[0] == 'r') // detect encoding if not specified for the reading mode
- {
+ } else if (modestr[0] == 'r') { // detect encoding if not specified for the reading mode
stream = tb_stream_init_from_file(path, mode);
- if (stream && tb_stream_open(stream))
+ if (stream && tb_stream_open(stream)) {
encoding = xm_io_file_detect_encoding(stream, &bomoff);
- else
- {
- if (stream) tb_stream_exit(stream);
+ } else {
+ if (stream) {
+ tb_stream_exit(stream);
+ }
xm_io_return_error(lua, "file not found!");
}
+ } else {
+ xm_io_return_error(lua, "invalid open mode!");
}
- else xm_io_return_error(lua, "invalid open mode!");
tb_assert_and_check_return_val(encoding != XM_IO_FILE_ENCODING_UNKNOWN, 0);
// write data with utf bom? e.g. utf8bom, utf16lebom, utf16bom
tb_bool_t utfbom = tb_false;
- if (tb_strstr(modestr, "bom"))
+ if (tb_strstr(modestr, "bom")) {
utfbom = tb_true;
+ }
// open file
- tb_bool_t open_ok = tb_false;
+ tb_bool_t open_ok = tb_false;
tb_stream_ref_t file_ref = tb_null;
- tb_stream_ref_t fstream = tb_null;
- do
- {
+ tb_stream_ref_t fstream = tb_null;
+ do {
// init stream from file
- stream = stream? stream : tb_stream_init_from_file(path, mode);
+ stream = stream ? stream : tb_stream_init_from_file(path, mode);
tb_assert_and_check_break(stream);
// is transcode?
tb_bool_t is_transcode = encoding != TB_CHARSET_TYPE_UTF8 && encoding != XM_IO_FILE_ENCODING_BINARY;
- if (is_transcode)
- {
- if (modestr[0] == 'r')
+ if (is_transcode) {
+ if (modestr[0] == 'r') {
fstream = tb_stream_init_filter_from_charset(stream, encoding, TB_CHARSET_TYPE_UTF8);
- else
+ } else {
fstream = tb_stream_init_filter_from_charset(stream, TB_CHARSET_TYPE_UTF8, encoding);
+ }
tb_assert_and_check_break(fstream);
// use fstream as file
file_ref = fstream;
+ } else {
+ file_ref = stream;
}
- else file_ref = stream;
// open file stream
- if (!tb_stream_open(file_ref)) break;
+ if (!tb_stream_open(file_ref)) {
+ break;
+ }
// skip bom characters if exists
- if (bomoff > 0 && !tb_stream_seek(stream, bomoff)) break;
+ if (bomoff > 0 && !tb_stream_seek(stream, bomoff)) {
+ break;
+ }
- // ok
open_ok = tb_true;
} while (0);
// open failed?
- if (!open_ok)
- {
+ if (!open_ok) {
// exit stream
- if (stream) tb_stream_exit(stream);
+ if (stream) {
+ tb_stream_exit(stream);
+ }
stream = tb_null;
// exit charset stream filter
- if (fstream) tb_stream_exit(fstream);
+ if (fstream) {
+ tb_stream_exit(fstream);
+ }
fstream = tb_null;
// return errors
@@ -288,7 +293,7 @@ tb_int_t xm_io_file_open(lua_State* lua)
}
// make file
- xm_io_file_t* file = (xm_io_file_t*)lua_newuserdata(lua, sizeof(xm_io_file_t));
+ xm_io_file_t *file = (xm_io_file_t *)lua_newuserdata(lua, sizeof(xm_io_file_t));
tb_assert_and_check_return_val(file, 0);
// init file
generated by cgit v1.3.1 (git 2.54.0) at 2026-09-03 07:00:03 +0000