| Age | Commit message (Collapse) | Author |
|---|
|
|
|
This is somewhat noticable difference (~10s), as these two headers are
basically included in every single JS file anyway.
|
|
Until we get the goodness that C++ modules are supposed to be, let's try
to shave off some parse time using precompiled headers.
This commit only adds some very common AK headers, only to binaries,
libraries and the kernel (tests are not covered due to incompatibility
with AK/TestSuite.h).
This option is on by default, but can be disabled by passing
`-DPRECOMPILE_COMMON_HEADERS=OFF` to cmake, which will disable all
header precompilations.
This makes the build about 30 seconds faster on my machine (about 7%).
|
|
Also adds a missing AK::Formatter specialisation for DNSName, required
for the debug calls to actually work (!)
|
|
Fixes #4393 :^)
|
|
In LibDebug this required implementing the Kill debug action, and
in HackStudio this required making the toolbar's stop action stop
the debugger if active.
|
|
Standard: https://pubs.opengroup.org/onlinepubs/009695399/functions/pthread_attr_getscope.html
Needed for https://fio.readthedocs.io
|
|
Add a basic non-thread safe implementation of pwrite, following the
existing pread(..) design.
This is needed for https://fio.readthedocs.io
|
|
Needed for porting https://fio.readthedocs.io
|
|
Although serenity doesn't implement IPv6 yet, applications
will often declare buffers to be of size `INET6_ADDRSTRLEN`
so it's guaranteed to work in both IPv4 / IPv6 environments.
This is needed specifically for a port of Flexible IO Tester
https://fio.readthedocs.io
This is standardized here:
https://pubs.opengroup.org/onlinepubs/009695399/basedefs/netinet/in.h.html
|
|
LibC is no different than any other code, it should be unit tested where
appropriate / possible.
|
|
@linusg noticed this bug in the original implementation during code review.
This makes all of these API's more spec conforming.
|
|
Need this for a port of FIO (Flexible IO Tester)
https://fio.readthedocs.io/
|
|
Move LibCompress unit tests to LibCompress/Tests directory and register
them with CMake's add_test. This allows us to run these tests with
ninja test instead of running a separate executable.
Also split the existing tests in 3 test files that better follow the
source code structure (inspired by AK tests).
|
|
This will eventually allow us to implement HTMLFormControlsCollection.
|
|
|
|
We should never enter the syscall handler from a kernel address.
|
|
|
|
|
|
The stress-ng port depends on these now that it detects we have
thread support.
|
|
GCC will insert various calls to pthread functions when compiling
C++ code with static initializers, even when the user doesn't link
their program against libpthread explicitly.
This is used to make static initializers thread-safe, e.g. when
building a library that does not itself use thread functionality
and thus does not link against libpthread - but is intended to
be used with other code that does use libpthread explicitly.
This makes these symbols available in libc.
|
|
GCC inserts calls to pthread_mutex_lock when compiling C++ code with
threads enabled.
|
|
This enables POSIX threads for GCC and makes the -pthread
argument available.
|
|
A website with a 4xx status code is still a valid website, we should not
artificially hide it. In fact, many websites implement custom 404 error
pages for example, often containing search functionality or links back
to the homepage.
This might have implications regarding the loading of stylesheets where
the request 404s, but since we were not handling 5xx status codes as
errors either, I think that's fine for now (but might need additional
work later). Worst case, the parser rejects to load some error page HTML
as CSS :^)
|
|
E.g. for "0" we have to contruct a PropertyName with Type::Number so it
looks in the indexed properties as expected.
|
|
We were getting the attributes of the existing value and then
immediately assigned the default attributes instead.
|
|
|
|
The Inspector hangs when user tries to do that.
|
|
In the ListItemMarkerBox render the correct thing when painting.
This covers decimal counting for ordered lists as well as square,
disc, and circle for unordered lists. Thus all currently supported
list-style-types are displayed correctly.
This closes #2059
|
|
In the ListItemBox we get the index of the current <li> element in the
parent and pass it to the ListItemMarkerBox.
This patch is work towards #2059
|
|
For Elements depending on the index they are inside their parent. Most
notably the <ol> element.
Also added a typed version to only count children of a certain type.
This patch is work towards #2059
|
|
|
|
|
|
|
|
This adds a simple REPL command line utility for (eventually) executing
SQL statements / files. Currently, it just validates statements from
stdin and prints any errors.
|
|
This parser builds on the LibSQL lexer and currently only allows users
to parse 'CREATE TABLE' statements.
|
|
LibSQL aims to be a SQLite clone for SerenityOS. Step 1 is creating a
tokenizer to lex SQL tokens. This lexer is heavily influenced by the
LibJS lexer.
|
|
|
|
- Replaced /Root with
- Improved documentation.
- Removed a few typos.
- Replaced with
- Added brackets in some cases.
Most of the changes were reviewed and applied manually.
|
|
|
|
SerenityOS repositories
|
|
Otherwise we'd have to update them in this repository
every time something changes.
|
|
program name
|
|
|
|
|
|
As is required by the specification:
"Let isRegExp be ? IsRegExp(searchString). If isRegExp is true,
throw a TypeError exception."
|
|
A missing searchString is allowed by the specification and is treated
as js_undefined. ("undefined test".startsWith() => true)
|
|
The current event name was accidentally set for the onmousedown event.
|
|
|
|
HTMLInputElement now inherits from FormAssociatedElement, which will
be a common base for the handful of elements that need to track their
owner form (and register with it for the form.elements collection.)
At the moment, the owner form is assigned during DOM insertion/removal
of an HTMLInputElement. I didn't implement any of the legacy behaviors
defined by the HTML parsing spec yet.
|
