assets + stylesheet assets #1475

Also deprecate a couple inlineImages tests

class

captureAssets directly

option.

TD, TR, TH, TBODY, THEAD, image, feImage, and cursor elements

config parameter

test

cacheable origins

And remove attribute while asset is loading

AssetManager

xlink:href is deprecated Most modern browsers rewrite `xlink:href` to `href` which is currently supported Tested in Firefox, Safari & Chrome

AssetManager.manageAttribute

different types of assets (not just img#src)

Co-authored-by: Eoghan Murray <[email protected]>

…the AssetManager needs to interrogate other attributes to determine how to manage, e.g. <link with href depending on rel="stylesheet"

…of this moving location & change to vitest

…tributes instead of storing config on the Meta event - a better approach as discussed with Justin and Yanzhen

…n determine that Assets won't be arriving later

…e existing nested iframe mechanism

…ic types that inline images or pdfs

…suite concurrently

…r handling stylesheet assets

…erives from rrweb-io/rrweb-snapshot#74

…en `srcset` and other url attributes in one place

…al recording config, i.e. the origins don't match. Should make the 'refused' status redundant

…s (that are otherwise blocked due to e.g. CORs issues) as assets

…ures stylesheets with `blob:` url'

…e of requestIdleCallback and not block the main thread during snapshotting

…is available https://developer.mozilla.org/en-US/docs/Web/API/HTMLLinkElement/sheet

…s on another branch)

…rocessing cost during initial snapshotting

…lves real problem in 'can record style text mutations' where we need to have the asset events emitted so that we can replay correctly

…ause` in 'can record style text mutations' test - actual fix is to skip in `applyEventsSynchronously` as the event has already been handled by preloadAllAssets on FullSnapshot event

…t is necessary for images. This fixes the 'can record style text mutations' test where we are scrubbing around the replayer synchronously and expect the DOM to keep up

…rectly using cssTextSplits by applying to the serializedNode rather than the DOM node. Another fix the 'can record style text mutations' test

…as a real element - otherwise rrweb-snapshot would need to import RRStyleElement which would create a cyclic dependency

…the `origins` mechanism to attempt to inline all images. The previous method will still be used in standalone snapshots

…rt - they are basically testing the same thing as the iframe versions still have <img> elements. These tests were also previously marked DEPRECATED due to association with older inlineImages setting

…et - the asset event will include the final content anyway

…xt node mutations

…a `splits` parameter when it is an asset so that there's a uniform way of processing it

…ting a zero or negative value for `config.processStylesheetsWithin`

…="stylesheet">

…rt of the captureAssets config where possible

…g new asset events

… config.stylesheets, as I want that to match inlineImages in that `true` will attempt to fetch all stylesheets

… (in terms of outer rules)

…r than just an asset

…synthetic property on the style element

…ther than browser context

Thanks for adding this document. I tested that even the translation result of Chatgpt4 is not good enough. So I changed some words to make the doc more readable for Chinese users.

… to vite

…ts in parallel - not needed when running test by itself

…se process

…, wasn't getting rebuilt with `adaptCssForReplay`

…istory via asset-event branch

… a `true` (ignore by default) - an empty (or maybe partial) captureAssets config where origins was unspecified was producing the error

… with { origins: true }

// presence of onAssetDetected means we should get // rr_captured_href (with contents promised later - i.e. using rrweb/record) Not sure when this test started failing

… be a fixup to 'Capture <style> element css via an asset event...'

…esume that as the machines are quicker there, the tests complete before stylesheets are processed

…ig.json

…apshots until we receive the stylesheet assets to avoid a flash of unstyled content (fouc)

…n the recording, and the ordering of asset arrival in the replayer was not regular. This change also means we only wait for stylesheets (which have a `timeout` associate with their status), which is good as other assets can be handled asyncrounously by the replayer asset manager without any negative effects (images get a placeholder image, whereas a missing stylesheet affect the entire page rendering)

… I'm about to remove the reset in favour of keeping assets around

…o that specified in the full snapshot

…multi-page session). - instead of resetting between FullSnapshots, we instead record assets against a timestamp - prefer assets with timestamps subseqent to a snapshot (as if a reset happened) - if none can be found, find the most recent prior asset for a url The asset manager now requires a rudimentary idea of where the replayer is at prior to applying an asset

…e types from rrweb-snapshot to @rrweb/types' but I don't know why it wasn't needed before

…that we don't delay fullsnapshot rendering awaiting for them

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

assets + stylesheet assets #1475

assets + stylesheet assets #1475

Commits on Oct 14, 2024

Commits on Oct 25, 2024

assets + stylesheet assets #1475

Are you sure you want to change the base?

assets + stylesheet assets #1475

Commits on Oct 14, 2024

Commits on Oct 25, 2024