70961 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Callum Law	70cb8d23fb	Tests: Import/create tests related to absolutization of StyleValues ... Done in a separate commit to show progress	2025-08-06 17:44:12 +01:00
Callum Law	c8bd58c0ba	LibWeb: Add method to absolutize CalculatedOr	2025-08-06 17:44:12 +01:00
Callum Law	0df9c225aa	LibWeb: Add method to absolutize PercentageOr	2025-08-06 17:44:12 +01:00
Callum Law	ba4a57b34d	LibWeb: Absolutize StyleValueList in StyleComputer::absolutize_values	2025-08-06 17:44:12 +01:00
Callum Law	a2ceed27e2	LibWeb: Absolutize CalculatedSVs in StyleComputer::absolutize_values	2025-08-06 17:44:12 +01:00
Callum Law	9db3c66a85	LibWeb: Use correct font size when absolutizing animated properties	2025-08-06 17:44:12 +01:00
Callum Law	011ab5f714	LibWeb: Use correct SerializationMode when serializing CalculatedOr	2025-08-06 17:44:12 +01:00
Callum Law	a5cbcaf698	LibWeb: Use correct SerializationMode when serializing PercentageOr	2025-08-06 17:44:12 +01:00
Andreas Kling	74722cc499	LibWeb: Use the soon-to-be hovered element for tooltip notifications ... Instead of the currently hovered element.	2025-08-06 17:26:18 +01:00
Callum Law	a44e28fd56	LibWeb: Avoid premature creation of CSSPixels in calc simplification ... Previously we were converting lengths to CSSPixels values when we didn't need to, this had a couple of effects in that: - We rounded to CSSPixel resolution prematurely (sometimes giving incorrect results) - We converted NaN to 0 when we shouldn't have. We now avoid prematurely converting lengths to CSSPixels values in two places: - `CalculationResult::from_value` - `CalculatedStyleValue::resolve_length_deprecated` (the new method already avoided rounding). Gains us 16 WPT tests.	2025-08-06 14:57:34 +01:00
Callum Law	3fa7bc1919	LibWeb: Simplify calc products including unresolved percentages	2025-08-06 14:53:22 +01:00
Callum Law	c3b1013018	LibWeb: Skip inverted non-canonical values when simplifying product node ... This matches the logic with the non-inverted children. Gains us 2 WPT tests.	2025-08-06 14:53:22 +01:00
Jelle Raaijmakers	cd1cf75f7b	LibWeb: Unbreak SVGFilterElement compilation ... This was caused by an undetected merge conflict.	2025-08-06 15:47:44 +02:00
Tim Ledbetter	1dd3608960	LibWeb/SVG: Apply SVGFeBlendElement blend mode ... Previously, the blend mode was always assumed to be `normal`.	2025-08-06 15:21:03 +02:00
Tim Ledbetter	22225af994	LibGfx: Reorder CompositingAndBlendingOperator values ... This change aligns blending operator values with the constants in `SVGFEBlendElement`.	2025-08-06 15:21:03 +02:00
Jelle Raaijmakers	3d613a2ac4	LibWeb: Simplify SVGFilterElement::gfx_filter() ... No functional changes.	2025-08-06 14:00:31 +01:00
Jelle Raaijmakers	400a36aac1	LibWeb: Warn about unknown SVG filters ... This might give some debug noise, but it's better than remaining silent.	2025-08-06 14:00:31 +01:00
Sam Atkins	9b6c26e347	Tests: Include test URL in fuzzy mismatch reports ... The test logs tend to get a bit mixed together, so this makes it possible to identify which values correspond to which test when multiple are failing at once.	2025-08-06 13:51:36 +02:00
Sam Atkins	75046857ce	Tests: Report image differences on tests with no fuzzy config defined ... This can be helpful for identifying what the fuzzy config should be for new tests.	2025-08-06 13:51:36 +02:00
Sam Atkins	90b49b9015	Tests: Report what the fuzzy configuration parsing error was	2025-08-06 13:51:36 +02:00
Sam Atkins	9889710033	Tests: Allow whitespace around fuzzy-matching configurations ... Several WPT tests have a space after the semicolon, for example https://wpt.live/css/css-images/gradient/gradient-single-stop-001.html : ``` <meta name="fuzzy" content="maxDifference=0-1; totalPixels=0-6000"> ```	2025-08-06 13:51:36 +02:00
Andreas Kling	28c1dd551b	LibWeb: Don't let aspect-ratio influence size definiteness ... As it turns out, we still have to let the formatting contexts do a bit of layout work in order to correctly apply the aspect-ratio. Hence we can't just implicitly transfer definiteness from one size to the other. This is a revert of 1cfd8b3ac0.	2025-08-05 21:33:41 +02:00
Andreas Kling	41e8211405	LibWeb: Honor "should treat as auto" cases in aspect-ratio decisions	2025-08-05 21:33:41 +02:00
Aliaksandr Kalenik	1201447dfe	WebContent: Delete unused PageClient::layout_root()	2025-08-05 17:24:34 +02:00
Aliaksandr Kalenik	7a34bc2700	LibWeb: Re-evaluate media queries only when things they depend on change ... Before this change we were re-evaluating media queries on every frame which adds up in 1-4% in profiles on Discord.	2025-08-05 17:24:34 +02:00
Timothy Flynn	3920194bca	LibJS: Move regulating and balancing logic into ISODateSurpasses ... This is an editorial change in the Temporal proposal. See: eddb77f	2025-08-05 11:18:08 -04:00
Timothy Flynn	a95d3e2a5e	LibJS: Split ISO and non-ISO Temporal calendar operations ... This is an editorial change in the Temporal proposal. See: 47042f2	2025-08-05 11:18:08 -04:00
Timothy Flynn	6d1f90c739	AK: Remove now-unused UTF-16 length from UTF-8 string helper	2025-08-05 15:13:36 +02:00
Timothy Flynn	04d32deb52	LibWeb: Port SVGFormattingContext to UTF-16	2025-08-05 15:13:36 +02:00
Timothy Flynn	d893d3234d	LibGfx: Support UTF-16 SVG text placement	2025-08-05 15:13:36 +02:00
Timothy Flynn	2dc0a3b3ce	AK: Add trim methods to Utf16String that skip allocation when not needed ... If the string does not begin with any of the provided code units, we do not need to create a new string.	2025-08-05 15:13:36 +02:00
Timothy Flynn	0efa98a57a	LibJS+LibWeb+WebContent: Port JS::PropertyKey to UTF-16 ... This has quite a lot of fall out. But the majority of it is just type or UDL substitution, where the changes just fall through to other function calls. By changing property key storage to UTF-16, the main affected areas are: * NativeFunction names must now be UTF-16 * Bytecode identifiers must now be UTF-16 * Module/binding names must now be UTF-16	2025-08-05 07:07:15 -04:00
Timothy Flynn	cd276235d7	AK: Add a couple of validation-skipping UTF-16 string factories ... String and FlyString are known to be valid UTF-8, so we can skip validation when constructing a UTF-16 string from them.	2025-08-05 07:07:15 -04:00
Timothy Flynn	782f8c381c	AK: Implement the spaceship operator for UTF-16 strings	2025-08-05 07:07:15 -04:00
Timothy Flynn	5af99f4dd0	AK: Allow Utf16StringBase to hold null data ... This is required by JS::PropertyKey. This will also be needed when we implement an Optional<Utf16String> specialization.	2025-08-05 07:07:15 -04:00
Timothy Flynn	0bf565b97f	AK: Allow comparing UTF-16 strings to UTF-8 strings ... Before now, you could compare a Utf16View to a StringView, but it would only be valid if the StringView were ASCII. When porting code to UTF-16, it will be handy to have a code point-aware implementation for non-ASCII StringViews.	2025-08-05 07:07:15 -04:00
Timothy Flynn	319e7aa03b	AK: Do not replace lonely surragates with U+FFFD while iterating ... Utf8View doesn't do this either. The wobbly format is expected by JS.	2025-08-05 07:07:15 -04:00
Callum Law	a70a397501	LibWeb: Support percentages in word-spacing ... Fixes crash in the created test as well as https://wpt.live/css/css-text /word-spacing/reference/word-spacing-percent-001-ref.html. The WPT test hasn't been imported as it passing is currently a false-positive due to the fact that we don't yet respect `word-spacing` in most cases.	2025-08-05 11:43:55 +01:00
Jelle Raaijmakers	62cf33b98e	LibGfx: Remove WebP animation writer and utility ... This went unused.	2025-08-05 11:30:20 +02:00
Andreas Kling	b60d465029	LibWeb: Defer updating hovered node until end of mouse event handling ... Updating the hovered node may fire events, and so we can't assume the layout and paintable nodes we've found via hit testing will be valid after doing it.	2025-08-05 11:29:45 +02:00
Andreas Kling	7e06906157	LibWeb: Use GC::Ptr more instead of raw pointers in EventHandler ... Gives us nicer stack traces in debug builds.	2025-08-05 11:29:45 +02:00
Sam Atkins	e0c851b736	Tests/LibWeb: Generate screenshot test boilerplate in add_libweb_test.py ... Creating the image is left as an exercise to the user, but this saves some copypasta when adding a new test.	2025-08-05 09:59:47 +01:00
Idan Horowitz	110136b862	LibCrypto: Convert SignedBigInteger::import_data to accept Bytes ... This brings it up to par with UnsignedBigInteger.	2025-08-05 09:08:56 +02:00
Idan Horowitz	b0fdbe3756	LibCrypto: Convert UnsignedBigInteger::import_data to accept Bytes ... All the callers are already using Bytes and manually converting to the old style pointer + length when calling this API.	2025-08-05 09:08:56 +02:00
Idan Horowitz	2a3b072d0e	LibCrypto: Convert SignedBigInteger::export_data to return a span ... This brings it up to par with UnsignedBigInteger.	2025-08-05 09:08:56 +02:00
Idan Horowitz	660a499223	LibCrypto: Convert UnsignedBigInteger::export_data to return a span ... This helps make callers only use the slice of the output buffer that was written to. As part of updating the callers of the API several bugs were fixed and useless code paths were removed: - The exported data is not host-endianess dependent (always big endian) - The exported data does not contain leading zeros - The output buffer is only written up to the result's size	2025-08-05 09:08:56 +02:00
Callum Law	15cee5cc15	LibWeb: Correctly round non-integer z-indexes ... Previously we would just cast to an int. Gains us 3 WPT tests.	2025-08-05 06:37:40 +01:00
Tim Ledbetter	9a5b740da1	LibWeb: Remove unnecessary debug logging from SVGSVGElement	2025-08-05 01:46:36 +02:00
Idan Horowitz	e2fe46065a	LibWeb: Remove extraneous trailing '/' from service worker script scope ... The spec language specifies 'Set maxScopeString to "/", followed by the strings in XXXX’s path (including empty strings), separated from each other by "/"': That is, adjacent components are separated by a '/', but the last component does not get a trailing '/'. This resulted in the generated scope string ending with '//' in some cases, incorrectly tripping the 'Service-Worker-Allowed' security check	2025-08-05 00:37:53 +02:00
Sam Atkins	56852fb5c2	LibWeb/CSS: Set style sheet for children of ContentStyleValue ... Without this, any relative url()s in the `content` property don't know what style sheet they are in, which makes them load relative to the document instead.	2025-08-05 00:35:03 +02:00