* Stop the admin menu jumping when a GoDAM page loads (#2207)
* Fix: stop the admin menu jumping when a GoDAM page loads
pages/style.css restores Tailwind's border reset for every element on the
page. It is injected by page-css.min.js after the page has painted, and
core's a { transition: border .05s } then animates each admin menu link's
border from 3px to 0: every item grows from 34px to 40px and the whole
menu jumps by about 186px, then snaps back. Keep the reset off
#adminmenumain and #wpadminbar.
* Fix: open the Media Editor with the admin menu already folded
The editor folds the menu from React after the page has painted, so the
menu showed at full width and then collapsed. Add the folded body class
on the server when the editor opens a file.
* Keep the reset off the menu and admin bar containers too; test the editor fold
- pages/style.css also excludes #adminmenumain and #wpadminbar themselves,
not only their descendants.
- MediaEditorMenuFoldTest: the Media Editor with an id gets 'folded';
its file picker, other GoDAM pages, the Media Library and no screen
do not.
* Regenerate the POT: 12 references into class-pages.php moved down
Line numbers only; no strings added or changed. Built like CI: Node
22.14.0, npm ci, npm run build:prod, composer pot.
* feat(seo): fix VideoObject schema and add video sitemap (#2186)
* feat(seo): fix VideoObject schema and add video sitemap (#1622 v1)
Make GoDAM's video schema indexable by Google and add a video sitemap
that plugs into Yoast SEO, Rank Math, All in One SEO or the WordPress
core sitemap. No new public URLs are created.
- Schema contentUrl now uses the HLS (.m3u8) stream, falling back to the
original file; the DASH .mpd is never output. Resolved at output time,
so pages cached before this change are fixed without a re-save.
- Add embedUrl pointing at GoDAM's existing video-embed player page.
- Never output an empty description: fall back to a 160-char transcript
excerpt (fetched on save only, cached in _godam_seo_transcript_excerpt),
then the page excerpt, otherwise omit the property.
- Output isFamilyFriendly as a real JSON boolean (string values such as
"false" are parsed correctly) and omit an empty uploadDate.
- Add Video_Sitemap: /godam-video-sitemap.xml with one <url> per
published page and one <video:video> per video, built from the cached
video SEO post meta via the shared schema builder so both always agree.
- Exclude drafts, private and password-protected posts, pages noindexed
in Yoast / Rank Math / AIOSEO, and videos the editor excludes.
- List the sitemap in the Yoast, Rank Math and AIOSEO indexes, or via a
core sitemap provider when none is active; paginate at 1000 URLs,
cache per site with version-based invalidation.
- Add GoDAM > Settings > SEO with an "Enable video sitemap" toggle (off
by default) and a per-video "Exclude from video sitemap" toggle in
Gutenberg, Elementor and WPBakery.
- Warn editors in all three editors when a video title looks like a
raw file name.
- Prefer the HLS URL when prefilling SEO contentUrl in the editors.
- Add new filters: godam_video_seo_content_url, godam_video_seo_embed_url,
godam_video_seo_description, godam_video_sitemap_enabled,
godam_video_sitemap_entries_per_page, godam_video_sitemap_post_types,
godam_video_sitemap_exclude_post, godam_video_sitemap_entry.
- Add PHPUnit coverage (VideoSeoSchemaTest, VideoSitemapTest) and Jest
tests for the filename-title check.
* Resolve feedback
* Resolve feedbacks and regenrate POT file
* Revert the POT file
* Resolve Privacy Issue
* feat(seo): add key moments to video schema from chapters (#1597) (#2194)
Output each GoDAM video chapter as a schema.org Clip in the VideoObject's
hasPart, so Google can show "key moments" (jump-to links) for the video
in search results.
- Add Seo::get_key_moment_clips(): reads the chapters from the
attachment's rtgodam_meta at output time, so editing chapters updates
the schema without re-saving the page.
- Each Clip has name, startOffset, endOffset and a deep link of the form
<page>?t=<seconds>#godam-video-<attachment id>, which the GoDAM player
already uses to seek on page load.
- A clip ends at the chapter's own endTime, else the next chapter's
start, else the video duration (clamped to the duration when known).
- Skip chapters with no title, an unreadable start, a start at or past
the end of the video, or zero length; drop duplicate start times;
strip HTML from chapter names.
- Parse chapter times as plain seconds, mm:ss or hh:mm:ss(.ms), falling
back to originalTime when startTime is missing.
- Build clip URLs without add_query_arg(): use & on plain permalinks and
drop any existing fragment from the page URL.
- Only add hasPart when the video has an attachment and a host page
(render-time output on archives/404 has nothing to deep-link to).
- Add filters godam_video_seo_key_moments_enabled and
godam_video_seo_key_moments.
- Move ISO 8601 duration parsing to Seo::iso8601_to_seconds();
Video_Sitemap::iso8601_duration_to_seconds() now delegates to it.
- Add VideoSeoKeyMomentsTest (27 tests) and a get_permalink() test stub.
* Match WPBakery SEO Settings to Gutenberg Model (#2209)
* ci: run the POT check only on PRs into main (#2188)
* ci: run the POT check only on the develop → main merge
Feature PRs into develop no longer have to regenerate languages/godam.pot.
The release PR from develop into main still fails if the committed POT is
stale. PRs into main from any other branch skip the job.
* ci: run the POT check on every PR into main
Patch releases reach main from fix and hotfix branches, not only from
develop, so they need the check too. The branches: [main] trigger
already keeps it off feature PRs into develop.
* Pass only a numeric form ID to the form layer preview (#2223)
Cast the layer id to an integer in the preview template and in the Ninja Forms
and MetForm callbacks, and add unit tests for it.
* Keep the Gravity Forms recorder working after an AJAX page change (#2217)
* Keep the Gravity Forms recorder working after an AJAX page change
Gravity Forms replaces a page's markup on every AJAX page change or failed
submit. The recorder kept pointing at the replaced container, file input and
button, and Uppy had bound its dashboard trigger once to the old button, so
Start Recording opened nothing on the new page.
On every gform_post_render, an uploader whose markup left the page now finds
the new container by its id, reads the input and button from it, binds the
new button once and shows the recording it holds again.
* Regenerate the translation template for the recorder's moved lines
* Stop sending the signed-in viewer's email with video analytics (#2213)
* Stop sending the signed-in viewer's email with video analytics
The page params that GoDAM prints on every front-end page no longer carry
the current user's email address (`emailId`), and the analytics request body
no longer has an `email` field. The WordPress user id is still sent.
Tests: the request-body helper and the page params are both checked for the
absence of the email, including a page cached by an older version.
* Remove the viewer's email from the analytics data listed in the readme
The analytics.godam.io entry under External services no longer lists Email as data sent.
* Keep the signed-in viewer's email out of the player's page data
The player script's page data (godamData.currentLoggedInUserData) carried the current user's email address, which nothing in the browser reads. It now holds the display name and whether someone is signed in. The comments routes still read the email on the server.
* Keep the fixture lines untouched in the JS test
The explanation moves into the test, so the fixture lines this branch leaves alone cannot conflict with the sibling change.
* List the page's user fields instead of removing one
rtgodam_get_current_user_data_for_page() now returns only the display name and the sign-in type, so a field added later to the server-side user data is not printed into the page by accident.
* Regenerate the POT for the new line numbers
The new helper in custom-functions.php moves five strings further down the file, so their source references change. Nothing else in the file differs from develop.
* Clean up video share popup: remove GoDAM Central/Embed links, use Page link, show share button for all videos (#2211)
* Clean up video share popup links and share button gating
- Remove the redundant GoDAM Central and Embed copy fields from the share
modal footer, keeping only the Page Link field.
- Point the social share links at the WordPress page link instead of the
GoDAM Central URL, which is private by default and shows a login prompt.
- Show the share button for all videos when sharing is enabled in the
block settings, instead of only transcoded videos.
Closes #2210
* Address review: keep share anchor and gate button on an ID
- Encode the share link with encodeURIComponent so the #godam-video-{id}
anchor and any query param travel with the social share URL instead of
splitting off (encodeURI left #, ? and & intact).
- Hide the share button for players with neither a WordPress attachment ID
nor a transcoding job ID ("Insert from URL" embeds), since their share
link could not target the video.
- Remove stale README references to the removed GoDAM Central/Embed links
and correct the ID described for share links.
---------
Co-authored-by: KMchaudhary <kuldipkumar.chaudhary@rtcamp.com>
* Save converted MP3 audio as audio/mpeg, and fix files already saved as audio/mp3 (#2208)
* Fix: store transcoded MP3 audio as audio/mpeg, not audio/mp3
WordPress registers MP3 as audio/mpeg. Attachments saved as audio/mp3
were left out of REST media_type=audio, and REST ignored a
mime_type=audio/mp3 filter. New conversions are saved as audio/mpeg,
and a one-time migration (2.4.1 batch) fixes the ones saved before.
* Test the saved type, satisfy the hook-integrity check, retry a failed SELECT
- TranscodedAudioMimeTypeTest calls add_transcoded_files() and checks
that converted audio is saved as audio/mpeg (video stays video/mp4).
With the old audio/mp3 it fails.
- add_transcoded_files() starts $transcoded_files as array(), not false;
PHP 8.1+ reported every conversion as a deprecated false-to-array use.
- The migration's two direct queries carry godam-coverage-ignore: the
runner already runs it once per site.
- A failed SELECT now returns false so the runner retries, instead of
marking the site done with nothing changed.
- The runner test checks batch order instead of which batch is last.
* Run the MP3 type migration in the 2.5.0 batch
2.4.1 ships without this change but still stores 2.4.1 as each site's
migration version, and the runner skips batches at or below the stored
version. Keyed 2.4.1, the migration would never run on sites that update
through 2.4.1; keyed 2.5.0 it runs for both 2.4.0 and 2.4.1 sites.
* Release prep v2.5.0: version bump, changelog and POT refresh (#2231)
- Version 2.5.0 in godam.php, readme.txt, README.md, package.json and package-lock.json.
- v2.5.0 changelog in readme.txt and CHANGELOG.md; v2.3.2 removed from readme.txt to keep the last three.
- Every @SInCE n.e.x.t replaced with 2.5.0 (51 tags in 6 files).
- languages/godam.pot regenerated from a production build.
* Fix the two new Plugin Check warnings in the Video SEO sitemap code (#2232)
- class-video-sitemap.php: the phpcs:ignore named slow_query_meta_key, which is not a sniff, so it ignored nothing. Use slow_db_query_meta_key.
- class-video-sitemap-core-provider.php: ignore NonPrefixedHooknameFound on WordPress core's own wp_sitemaps_index_entry filter.
* Keep page-builder shortcode text out of the video sitemap description (#2235)
* Keep page-builder shortcode text out of the video sitemap description
The sitemap is served at parse_request priority 1, before WPBakery and Divi
register their shortcodes, so core's strip_shortcodes() left tags such as
[vc_row] in the fallback description taken from the page text. Add the tag
names found in the text to strip_shortcodes()'s own list so unregistered
shortcodes are removed the same way registered ones are, which gives the same
description in the sitemap and on the front end.
The strip_shortcodes() test stub removed every [tag], which hid the bug. It now
follows core: only registered shortcodes are removed.
* Only strip builder-style unregistered shortcode names from the fallback description
Review of the first version found that it also removed ordinary bracketed text
([sic], [1], [b]bold[/b]) from the front-end description, that one very long
bracketed word made core's pattern too large and the description empty, that a
tag another filter had taken off the strip list was re-added, and that the
filter stayed registered if stripping threw.
Only names that look like a builder tag are added now (a letter first, an
underscore or hyphen in the name, 64 characters at most), tags that are
registered are left to core, and the filter is removed in a finally block.
New tests pin the cap, the length limit, the filter removal and the ordinary
bracketed text.
* Set the 2.5.0 changelog date to October 22, 2026 (#2236)
2.5.0 releases on 22 Oct (release ticket rtCamp/godam-plugin-wp#60), but readme.txt and CHANGELOG.md dated it October 21.
---------
Co-authored-by: Kuldip Chaudhary <64731232+KMchaudhary@users.noreply.github.com>
Co-authored-by: Rudrakshi Gupta <109859631+rudrakshi-gupta@users.noreply.github.com>
Co-authored-by: Vishal Kumar <vishnshiv3@gmail.com>
Co-authored-by: KMchaudhary <kuldipkumar.chaudhary@rtcamp.com>
Review packet
In one line: The plugin no longer puts the signed-in viewer's email address into the front-end page (in the
videoAnalyticsParamsobject and in the player'sgodamData.currentLoggedInUserData) or into the analytics events the player sends. No issue; planned in rtCamp/godam-analytics#364 (item 4).1. Criteria and proof
emailId(before: it carries the viewer's address)test_signed_in_viewers_email_is_not_in_the_page_params; capture S01, block 1email(before:godamData.currentLoggedInUserData.emailholds the address)test_player_page_data_for_a_signed_in_viewer_has_no_email; capture S01, block 2emailkey (before:emailholds the viewer's address)does not send the viewer email, even when the page still carries one; capture S01, block 3emailId, does not make the player send itemailId: 'a@b.test'on purpose)test_the_server_side_user_data_still_has_the_emailtest_page_params_keep_what_analytics_still_needs; JS teststill sends the WordPress user id2. Before and after
A signed-in subscriber opens a page with videos on godam-dev in Chrome and plays one for 6 seconds. Before is
develop(2.4.0); after is this branch. Captures are verbatim from the browser and from the analytics service's database.emailIdinvideoAnalyticsParamsemailingodamData.currentLoggedInUserDatanameandtypestay)emailin the play requestemailstored for the 6 events of the run3. Tests
tests/php/ViewerEmailNotInPageTest.php. It builds the page params with a signed-in user and asserts there is noemailIdkey and the address appears nowhere in the JSON; that the player's page data has noemailbut keepsnameandtype; that the server-side user data still has the email; and thatuserId,siteId,endpoint,tokenandpostIdare still in the page params. Each test runs in its own process, and loads its stubs there, because loadinggodam-transcoder-functions.phpregisters hooks andEarlyTranslationHooksTestchecks exactly that on first load.assets/src/js/godam-player/analytics-helpers.test.js(noemailkey even when the page still carriesemailId;wp_user_idstill sent).tests/stubs/template-tags.php, the page-tag stubs the PHP test needs (is_single,get_the_titleand similar).tests/stubs/engagement-functions.php. The current-user stub now also returnsIDanddata->user_email(an emptydatawhen nobody is signed in), as WordPress does. The Engagement tests that use it still pass: 428 PHP tests, 1414 assertions, in normal, reversed and random order.assets/src/js/godam-player/engagement.test.js. Its page-data fixture no longer has anemail, matching what the page now prints.test_signed_in_viewers_email_is_not_in_the_page_paramsfails with "Failed asserting that an array does not have the key 'emailId'",test_player_page_data_for_a_signed_in_viewer_has_no_emailfails with "Failed asserting that an array does not have the key 'email'", and the JS test fails withExpected path: not "email",Received value: "a@b.test".4. Risk flags
No risk flags fired on #2213. Nothing in the diff touched the watched areas.
5. Independent review
A separate review pass saw only the task and the diff, first on the analytics change and then again after the player-data and readme commits.
RTGODAM_ANALYTICS_BASEconstant reached every other testsetUp(), in the separate process. The suite passes in normal, reversed and random order.email, until those caches expire (raised twice)godamData.currentLoggedInUserData.emailin the player script's page data, seen on godam-dev. Fixed in 5732543. Static search of the source:emailIdis read only by the analytics sender and thegodam-for-wooadd-on;rtgodam_get_current_logged_in_user_data()is called by the page data (now filtered) and by the comments routes, which run on the server.assets/buildis not committed. A scan of 12 front-end pages as a signed-in viewer finds 0 copies from GoDAM.dataeven when nobody was signed in, and its header comment overstated what it guardsemailkey where it had''; a required column could reject itemail. The service tested isdevelop, which per rtCamp/godam-analytics#364 holds the same code as production.readme.txtlists "Email" among the data sent to analyticsREADME.mdhas the same table row; that file is changed by the readme PR of this release.class-godam-player.phpthat buildsgodamData, and removing one field is a blocklistnameandtype(240c9ba), so a later field is not printed by accident. No unit test for the call itself: the method registers about 20 scripts and styles and needs the built files, so the stubs would outweigh the check. The scan of 12 pages in S01 covers it on a real site.composer.jsonpins PHPUnit to^9.6, which honours them. The fixed values are the stub's, not the plugin's.6. Not verified
email, until they expire. Site owners can purge their page cache after updating. Not tested.godam-for-wooadd-on readsemailIdand sends it asemailtoo. After this PR it reads nothing there, so its Reel Pop events carry an emptyemail(seen in capture S02, type 4, stored empty). No change is needed in that repo for this PR.emailuntil they update. The analytics service already treats the field as optional; dropping it on the service side is item 4 of rtCamp/godam-analytics#364.7. Scope
Beyond the analytics code: the player's page data lost its
email(found when the page source was checked on godam-dev; no script reads it), and one line ofreadme.txtchanged, because the External services text for analytics.godam.io no longer lists Email as data sent (the later items are renumbered).languages/godam.potis regenerated (5 source line numbers moved, because the new helper sits above those strings incustom-functions.php).wp_user_idis kept for now: replacing it with an opaque id is a larger change and is not part of this PR.What changed
rtgodam_get_localize_array()no longer setsemailIdfrom the current user.buildAnalyticsRequestBody()no longer readsemailIdor sendsemail.rtgodam_get_current_user_data_for_page()returns the current user's data without the email; the player script'sgodamData.currentLoggedInUserDatauses it. The comments routes keep callingrtgodam_get_current_logged_in_user_data(), which still has the email.How to check by hand
videoAnalyticsParamshasuserIdand noemailId.analytics, play the video. The request payload has noemailkey.