v6.2.1 is a focused reliability patch for the 6.2.x line.
It strengthens several areas that matter directly during play: combat-context resolution,
combat stance initialization, Actor sheet persistence, localization and visual consistency.
It also continues the architectural cleanup introduced in v6.2.0 by aligning more closely
with Foundry VTT 14 lifecycle behavior and protecting source data from Active Effect contamination.
⚔️
Combat reliability
Tokens now resolve the encounter they actually belong to, with stronger stance initialization and cleanup.
◆
Data integrity
Sheet submissions rebase values against Actor source data before persistence.
🌍
Localization
Weapon groups, language names, Spanish content and French regression coverage were improved.
🖥️
Sheet polish
Resource terminology, layouts, separators and editor presentation received targeted refinements.
✦
⚔️ Combat reliability
The main gameplay-facing fix in v6.2.1 is a more explicit relationship between a token
and the encounter it actually belongs to.
- Combat context is resolved from the owning TokenDocument, not from whichever encounter happens to be displayed.
- Token HUD actions remain correct when a different encounter is currently viewed.
- HUD elevation rendering is preserved.
- Cleanup behavior remains safe when multiple encounters are involved.
- Character tokens added to combat through the HUD now have regression coverage ensuring a valid TOR2E stance is initialized.
- The custom token-layer combat toggle was removed in favor of a simpler Foundry VTT 14 lifecycle path.
✦
◆ Safer Actor sheet persistence
v6.2.1 further hardens the separation between effective values shown on a sheet
and the source values stored on the Actor.
- Submitted sheet values are rebased against
_source before they are persisted.
- This prevents Active Effect contributions from being accidentally saved as if they were base Actor values.
- Submission error reporting was improved to make unexpected Foundry form failures easier to diagnose.
- Actor image persistence now has dedicated end-to-end regression coverage.
This is primarily a data-integrity fix: the goal is that an Active Effect can change what the user sees
without silently changing the underlying base value when an unrelated field is edited.
✦
🌍 Localization improvements
- Weapon-group labels are localized correctly on NPC and Adversary sheets.
- Catalan and Brazilian Portuguese language names were corrected in the system manifest.
- Spanish translations received a broad update across several TOR2E terms and phrases.
- A new French E2E regression test opens Actor and Item sheets and checks that translations render instead of raw localization keys.
✦
🖥️ Sheet and interface polish
- Character Sheet resource labels now better match TOR2E terminology, explicitly distinguishing Load / Fatigue and Shadow / Shadow Points.
- Related accessibility labels and resource identifiers were updated.
- Resource diamonds are aligned more consistently beneath their headings.
- Character Sheet visual separators were restored.
- Adversary and Lore sheet layout, dimensions and editor presentation received targeted refinements.
- Several related CSS and template regressions are now covered automatically.
✦
🧪 Regression and engineering work
A significant portion of this patch is dedicated to preventing these fixes from regressing later.
- Combat stance E2E scenarios were expanded to cover entering combat, leaving combat and scene reloads.
- Reusable Playwright fixtures now centralize scene and combat setup.
- Foundry UI helpers were expanded to reduce duplicated test logic.
- French localization and Actor image persistence gained dedicated E2E tests.
- CSS guardrail analysis and duplicate-selector checks were strengthened.
- A manual GitLab CI preview workflow was added for upcoming releases from
develop.
✦
Upgrade
- Make sure you are running Foundry VTT 14.
- Back up your world before updating.
- Update TOR2E to v6.2.1 through Foundry's package manager.
- Open your world and verify the sheets and combat workflows you use most often.
✦
In short
v6.2.1 is not a feature-heavy release. It is a patch aimed at making the 6.2.x line safer:
more predictable combat behavior, cleaner Actor data persistence, better localization,
and stronger automated protection around the workflows that changed.