Repository navigation
Release - #1195
Open
pirate-bot wants to merge 9 commits into
Open
Release#1195pirate-bot wants to merge 9 commits into
pirate-bot wants to merge 9 commits into
Conversation
Individual cache tokens are stored as the API returned them, which can be a number. get_active_cache_booster() passed that value straight to Optimole::image(), whose cache buster is typed as a string, so image URL replacement threw a TypeError and broke the request. Cast the token to a string, and fall back to an empty token for non-scalar values. Closes #1187 Claude-Session: https://claude.ai/code/session_01WPwFxvA5VPq8sd4EzsNFFA Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Collaborator
Author
…URL (#1190) * fix: don't take a data: placeholder in a lazy-load attribute as the image URL The image regex matches the `src=` at the end of any `*-src=` attribute, so in `<img data-src="data:image/gif;..." src="photo.jpg">` it captured the placeholder. The tag replacer then ran against `data:image/gif`: no profiler srcset or dimensions, and a data-opt-id shared by every image with that placeholder. A lazy-load attribute whose value starts with data: is now skipped, so the real src is used. A data: value in a plain src is still matched, so images owned by another lazy-load script are handled as before. Fixes #1186 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WPwFxvA5VPq8sd4EzsNFFA * test: cover data: placeholders without a src and their lazy-load result Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WPwFxvA5VPq8sd4EzsNFFA --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* fix: stop retrying a failing transfer step forever When a background transfer step threw (an API error, a lookup that keeps failing), it was rescheduled 10 seconds later with the same step, with no limit. The transfer stayed "in progress" in the dashboard and never finished. The failed attempts of a step are now counted; after 5 in a row the chain moves on to the next step, so the transfer can reach its end and the images left behind can be retried. A successful step clears the count. Refs #1108 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Keep step failures per transfer in an option and don't report success after skips Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Fix the finish branch comment Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p --------- Co-authored-by: Claude <noreply@anthropic.com>
* feat: let an image keep its own sizes attribute The page profiler adds its measured breakpoints to an image's sizes attribute unless the value contains calc() or a similar formula. Sites with hand-written sizes had no supported way to keep them. An image with data-optml-keep-sizes now keeps its existing sizes attribute (srcset candidates are still added). The optml_keep_existing_sizes filter does the same site-wide without changing the markup. Refs #1115 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Clarify keep-sizes log, FAQ and tests Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Check the attribute keeps sizes before the filter overrides it Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p --------- Co-authored-by: Claude <noreply@anthropic.com>
* fix: bound the in-memory attachment ID cache The request-local map in Optml_Attachment_Cache kept every distinct URL looked up during a request, including misses, with no limit. It now keeps at most 2000 URLs and drops the older half when full; dropped URLs are still served from the persistent cache (transient or object cache). Refs #1139 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Cover the cache-miss path in the memory cap test and fix the value type Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p --------- Co-authored-by: Claude <noreply@anthropic.com>
…f sync" logs (#1196) * fix: pass the page through unprocessed after a fatal error A fatal error that hits while MySQL is still sending rows leaves the connection out of sync, so every profiler and cache read at shutdown logged "Commands out of sync". Skip processing in that case. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * test: cover both fatal-error pass-through paths Read the last PHP error through a private seam so tests can simulate a fatal error. New buffer tests check that the default in-handler mode and the shutdown-capture mode both pass the page through unprocessed; each fails when its guard is removed. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * docs: give the error reader seam a precise callable type Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p --------- Co-authored-by: Claude <noreply@anthropic.com>
…pted rollbacks (#1191) * fix: make the transfer lock atomic and clean up files left by interrupted rollbacks Two requests that started a transfer at the same moment could both take the transfer lock, because it was read and then written. The lock is now stored with an insert that fails when the row already exists (wp_cache_add with a persistent object cache). A rollback that died after moving the restored file into uploads left that file behind, so the retry got a suffixed name (.1-1.webp) while references kept the old one. The moved file is now recorded on the attachment until the rollback finishes, the retry removes it first, and the attachment points to the restored file right after the move instead of after the sub-sizes. Refs #1136 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * fix: reuse the file left by an interrupted rollback instead of deleting it Review follow-up on the rollback cleanup: - the retry writes the downloaded image over the file the interrupted attempt left behind, instead of deleting it before the sideload, so a failed move never leaves the attachment pointing at a missing file; - the pending-file meta is also cleared when the retry ends in the branch for attachments whose local metadata was already saved; - the uploads-dir check normalizes both paths, so it works on Windows; - the lock expiry is written again after the insert; - native parameter and return types on the new helpers. Refs #1136 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Write the restored file through a temporary copy and keep the lock expiry after a racing release Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p * Add native types to the failing stream test helper Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MEXWtYdE6e3ALYtqtP8j2p --------- Co-authored-by: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Linked issues
This release will close the following issues once merged:
Public changelog