Notable changes:
- Updated and the port API interface, new functions have been added
and some have been changed. The port library is now cleaned up to
not have any dependency on jerry-core, as it should be. The port library
is now strictly a collection of functions that implement
embedding/platform specific behavior.
- The default port implementation has been split for windows and unix.
Implemented port functions have been categorized and reorganized,
and marked with attribute((weak)) for better reusability.
- External context allocation has been moved to the port API instead
of a core API callback. The iterface has also been extended with a
function to free the allocated context. When external context is
enabled, jerry_init now automatically calls the port implementation
to allocate the context and jerry_cleanup automatically calls the port
to free the context.
- jerry_port_log has been changed to no longer require formatting to
be implemented by the port. The reason beind this is that it was vague what
format specifiers were used by the engine, and in what manner. The port
function now takes a zero-terminated string, and should only implement
how the string should be logged.
- Logging and log message formatting is now handled by the core jerry library
where it can be implemented as necessary. Logging can be done through a new
core API function, which uses the port to output the final log message.
- Log level has been moved into jerry-core, and an API function has
been added to set the log level. It should be the library that
filters log messages based on the requested log level, instead of
logging everything and requiring the user to do so.
- Module resolving logic has been moved into jerry-core. There's no
reason to have it in the port library and requiring embedders to
duplicate the code. It also added an unnecessary dependency on
jerry-core to the port. Platform specific behavior is still used through
the port API, like resolving module specifiers, and reading source file
contents. If necessary, the resolving logic can still be overridden as
previously.
- The jerry-ext library has also been cleaned up, and many utility
functions have been added that previously were implemented in
jerry-main. This allows easier reusability for some common operations,
like printing unhandled exceptions or providing a repl console.
- Debugger interaction with logged/printed messages has been fixed, so
that it's no longer the port implementations responsibility to send
the output to the debugger, as the port should have no notion of what a
debugger is. The printing and logging functions will now pass the
result message to the debugger, if connected.
- Cleaned up TZA handling in the date port implementation, and simplified
the API function prototype.
- Moved property access helper functions that use ASCII strings as
keys from jerry-ext to the core API.
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai dbatyai@inf.u-szeged.hu
* Added more details into documentation.
* Moved the C unit-test into its own file.
* Added extra test cases.
* Extended the API reference documentation with doctests.
JerryScript-DCO-1.0-Signed-off-by: Peter Gal pgal.usz@partner.samsung.com
Overflows in conversions from floating-point to integer are
undefined behavior in the C99 standard. (Clause 6.3.1.4: "If the
value of the integral part cannot be represented by the integer
type, the behavior is undefined.")
When UBSAN is enabled, this gets reported at `srand()` calls. (The
random seed is usually initialized using the date port API, which
represents dates as `double`s. But `srand` takes an `unsigned int`.
A simple cast from `double` to `unsigned` becomes undefined
behavior if the value is too large. And "now" is too large
nowadays. So, effectively, all executions start with an undefined
behavior.)
This patch fixes this by casting the floating-point value of the
date to an integer through a union.
JerryScript-DCO-1.0-Signed-off-by: Akos Kiss akiss@inf.u-szeged.hu
MSVC doen't support __attribute__((unused)),
we should use JERRY_UNUSED macro instead.
Additionally removed the internal jrt.h include from
tests/unit-core/test-common.h which was layering violation.
It made JERRY_ASSERT unavailable, we should use TEST_ASSERT.
JerryScript-DCO-1.0-Signed-off-by: Csaba Osztrogonác oszi@inf.u-szeged.hu
Their length (size) is known at compile time. Therefore `sizeof`
is more efficient for them.
JerryScript-DCO-1.0-Signed-off-by: Akos Kiss akiss@inf.u-szeged.hu
In general, public headers should not have compiler-specific
constructs but both the core and the port headers have attributes,
which are non-standard. It's better to factor out such constructs
to a common place (a new header) and hide them behind macros, which
can then be defined on a per-compiler basis.
This patch moves the existing definitions of function attributes and
likely/unlikely builtins to the new header. At the same time, it
unifies the names of these attribute defines and where they are
used. Moreover, it touches on jerry-main and removes the uses of
`__attribute__((unused))` entirely and replaces them with the
elsewhere used `(void) ...` pattern.
JerryScript-DCO-1.0-Signed-off-by: Akos Kiss akiss@inf.u-szeged.hu
Issue #2053 has highlighted the fact that random numbers are always generated with the same seed.
An example of generating different random numbers, other than the original seed, has been added to the documentation.
Furthermore srand initialization has been added to jerry-main, and targets.
Update test-common.h with srand call.
JerryScript-DCO-1.0-Signed-off-by: Daniel Balla dballa@inf.u-szeged.hu
The unit tests should follow the component structure, so this patch
moves all `jerry-core` unit tests under `tests/unit-core` and the
`jerry-libm` unit tests under `tests/unit-libm`.
JerryScript-DCO-1.0-Signed-off-by: Akos Kiss akiss@inf.u-szeged.hu
