Replace the naive close() loop in js_os_exec with closefrom().
On my system RLIMIT_NOFILE is set to 1 million and the delay from the
loop gets noticeable when I spawn many processes.
Use dlsym() to look up closefrom() because it's not available everywhere
(looking at you, musl.)
Fixes: https://github.com/quickjs-ng/quickjs/issues/711
Deserialized bytecode does not have IC state, i.e., `bc->ic == NULL`.
That may or may not be bug (IMO, it is and we should rebuild the
IC state during deserialization) but, either way, don't segfault.
DRY add_ic_slot() and its call sites in a hopefully NFC manner.
Don't raise a "invalid tag 12" exception when encountering bytecode
and JS_READ_OBJ_BYTECODE is not set, because no one knows what "tag 12"
means without looking it up, not even quickjs maintainers.
`JS_NewClassID(rt, &class_id)` where `class_id` is a global variable
is unsafe when called from multiple threads but that is exactly what
quickjs-libc.c did.
Add a new JS_AddRuntimeFinalizer function that lets quickjs-libc
store the class ids in JSRuntimeState and defer freeing the memory
until the runtime is destroyed. Necessary because object finalizers
such as js_std_file_finalizer need to know the class id and run after
js_std_free_handlers runs.
Fixes: https://github.com/quickjs-ng/quickjs/issues/577
I find it convenient to have access to JS_ReadObject and JS_WriteObject
from JS land. That's precisely the functionality that tests/bjson.c
provides, ergo, move it into quickjs-libc.c
- no longer pass the array length to `utf8_decode`
- add `utf8_decode_len` for border cases
- use switch based dispatch in `utf8_decode_len` to work around a gcc 12.2 optimizer bug
Ensure proper UTF-8 encoding (1 to 4 bytes).
Handle invalid encodings (return 0xFFFD and consume a single byte)
Individually encoded surrogate code points are accepted.
- add `utf8_scan()` to analyze a byte array for UTF-8 contents
detects invalid encoding, computes number of codepoints and content kind:
plain ASCII, 8-bit, 16-bit or larger codepoints.
- add `utf8_encode_len(c)` to compute the number of bytes to encode `c`
- rename `unicode_to_utf8` as `utf8_encode`
- rename `unicode_from_utf8` as `utf8_decode`
- add `utf8_decode_buf8(dest, size, src, len)` to decode a UTF-8 encoded
byte array known to contain only ASCII and 8-bit codepoints.
- add `utf8_decode_buf16(dest, size, src, len)` to decode a UTF-8 encoded
byte array into an array of 16-bit codepoints using UTF-16 surrogate pairs
for non-BMP1 codepoints.
- add `utf8_encode_buf8(dest, size, src, len)` to encode an array of 8-bit
codepoints as a UTF-8 encoded null terminated string
- add `utf16_encode_buf8(dest, size, src, len)` to decode an array of 16-bit
codepoints (including surrogate pairs) as a UTF-8 encoded null terminated string
- detect invalid UTF-8 encoding in RegExp parser
- simplify `JS_AtomGetStrRT`, `JS_NewStringLen` using the above functions
- simplify UTF-8 decoding and error testing
The shell script runs the tests and diffs stdout against v8.txt.
Lines added/removed means tests were broken/fixed.
Future work is to a) fix failing tests, and b) enable more tests.
A number are disabled for various reasons and mjsunit subdirectories are
currently skipped. Need to decide on a case-by-case basis what is and
isn't relevant to us.
At the moment about 430 tests are run of which approx. 80% pass.
JSValueConst was only used for the now removed CONFIG_CHECK_JSVALUE
build mode. It is kept around as an alias for JSValue in quickjs.h to
avoid breaking everyone's source builds but remove it everywhere else.
The default 256 kb stack is too small to run some of the test262 tests
when ASAN is enabled.
Double it to 512 kb and ensure threads created by quickjs have big
enough stacks.
