v1.12.0

This is a minor release with some new features, as well as some bug fixes and perf improvements.

Many thanks to @sorin-davidoi and @sgenoud for their help with this release!

New features

Blurhash

Sensitive images now show a blurred preview (blurhash). This feature can be disabled in the general settings, under "show a plain gray color for sensitive media."

show commits

fix(components/media): Handling of grouped videos #1409
fix: improve style of "large inline" media #1428
fix: fix blurhash for reblogged images #1412
fix(components/media): Handle autoplay with blurhash #1425
perf: use more efficient CSS selectors for inline media #1423
fix: ensure "disable blurhash" setting is honored #1422
fix: remove unused nCols computation #1420
fix: Fix the blurhash alignment and general style #1418
perf: call revokeObjectURL() on stale blurhash blob URLs #1403
fix(components/media): Autoplay and blurhash #1400
fix(workers/blurhash): Don't use OffscreenCanvas #1398
fix: reduce number of blurhash performance measures #1395
fix: move blurhash worker operations to before status rendering #1391
fix: tweak position of sensitive media button in blurhash mode #1390
fix: tweak colors for blurhash sensitive media button/text #1389
feat(media): Blurhash #1381

Extract text from images

Inspired by an upcoming Mastodon feature, Pinafore can now use tesseract.js to generate a description (alt text) for an image. Click the image edit (pencil) button to either edit the description, edit the focal point, or generate text from the image.

This can also be considered an accessibility feature, because now it's easier than ever to add alt text to an image!

show commits

perf: terminate tesseract worker after a delay #1449
fix: make the OCR progress bar accessible #1445
fix: add progress bar for OCR #1444
fix: improve style of focal/alt editor on mobile #1439
fix: use url over id as media upload cache key #1437
fix: prefer local file URLs for OCR #1436
fix: tweak style of focal point coordinate inputs #1435
feat: add OCR for image descriptions using tesseract.js #1433
fix: tweak backdrop filter in focal point editor #1434
fix: fix cursor when grabbing in draggable UI #1432
feat: add length gauge for media alt text editor #1431
feat: combine alt/focal point into single "media edit" dialog #1430

Bug fixes and perf improvements

More reliable notifications

In the past, you may have had to refresh Pinafore to get the most recent notifications or new toots in your timeline. With this fix, new toots and notifications should hopefully arrive without needing a refresh.

fix: reset websocket on online/offline/active events #1429

Accessibility fixes

The toot length gauge now respects the "reduce motion setting," the tappable area for settings buttons is larger.

show commits

fix: length gauge should respect reduceMotion #1440
feat(settings): Increase clickable area #1401

Don't over-fetch

Pinafore no longer eagerly fetches the custom emoji and lists from an instance whenever the page is loaded.

show commits

perf: don't fetch custom emoji on every page load #1451
perf: fetch lists from database on page load #1450
refactor(store/observers/instance): Don't fetch lists eagerly #1448

Handle revoked tokens

If you revoke a token for Pinafore in the Mastodon UI, then Pinafore should now handle that gracefully and log you out.

fix(actions/instances): Clear data when token revoked #1446

Better toot compose box input responsiveness

Responsiveness in the toot compose box should be slightly improved.

show commits

perf: be more consistent about compose input scheduling #1414
perf: improve input responsiveness in compose input #1413

Full changelog

nolanlawson/pinafore v1.12.0 on GitHub

v1.12.0

New features

Blurhash

Extract text from images

Bug fixes and perf improvements

More reliable notifications

Accessibility fixes

Don't over-fetch

Handle revoked tokens

Better toot compose box input responsiveness

nolanlawson/pinafore v1.12.0
on GitHub