fix(input): stop dividing touch/tablet bridge coords by output scale (#47) #93

Merged

albert merged 1 commit from u/albert/47/touch-compat-scale-offset into main 2026-07-12 18:35:03 +00:00

Conversation 0 Commits 1 Files changed 8 +132 -139

albert commented 2026-07-12 13:39:38 +00:00

Owner

Copy link

The absolute pointer/touchscreen the input bridges synthesize declares a
0..=65535 range that libinput maps onto the output's logical layout
space, so a full-pad sweep already tracks the finger 1:1 at any output
scale. The output-scale division added in #58 was therefore wrong in
principle; it only ever looked correct because it was solely exercised
with the XWayland HiDPI workaround forcing scale to 1.0, where the divide
is a no-op. On a scaled output without that workaround (Scratch at output
scale 1.5) it offset the synthesized cursor — issue #47.

Remove the correction from both bridges: rescale_abs no longer takes a
scale, new_absolute()/new_touchscreen() drop the output_scale parameter,
the --output-scale CLI flag is gone from the touch and tablet bridges,
and the host no longer queries the sway scale or passes it through.
Confirmed 1:1 end-to-end against the live Scratch repro (edge-to-edge
drag plus corner/center taps). The tablet bridge fix has no live repro
(no tablet on hand) but follows the same reasoning and needs re-testing
on real hardware.

Co-Authored-By: Claude Opus 4.8 (1M context) noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_01MYSRycxV3iG9PHjHqArLjg

Fixes #47 (for real this time, validated with the XWayland hack both on and off)

The absolute pointer/touchscreen the input bridges synthesize declares a 0..=65535 range that libinput maps onto the output's *logical* layout space, so a full-pad sweep already tracks the finger 1:1 at any output scale. The output-scale division added in #58 was therefore wrong in principle; it only ever looked correct because it was solely exercised with the XWayland HiDPI workaround forcing scale to 1.0, where the divide is a no-op. On a scaled output without that workaround (Scratch at output scale 1.5) it offset the synthesized cursor — issue #47. Remove the correction from both bridges: rescale_abs no longer takes a scale, new_absolute()/new_touchscreen() drop the output_scale parameter, the --output-scale CLI flag is gone from the touch and tablet bridges, and the host no longer queries the sway scale or passes it through. Confirmed 1:1 end-to-end against the live Scratch repro (edge-to-edge drag plus corner/center taps). The tablet bridge fix has no live repro (no tablet on hand) but follows the same reasoning and needs re-testing on real hardware. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MYSRycxV3iG9PHjHqArLjg Fixes #47 (for real this time, validated with the XWayland hack both on and off)

albert added 1 commit 2026-07-12 13:39:38 +00:00

fix(input): stop dividing touch/tablet bridge coords by output scale (#47) ... 78e69bd4af

The absolute pointer/touchscreen the input bridges synthesize declares a
0..=65535 range that libinput maps onto the output's *logical* layout
space, so a full-pad sweep already tracks the finger 1:1 at any output
scale. The output-scale division added in #58 was therefore wrong in
principle; it only ever looked correct because it was solely exercised
with the XWayland HiDPI workaround forcing scale to 1.0, where the divide
is a no-op. On a scaled output without that workaround (Scratch at output
scale 1.5) it offset the synthesized cursor — issue #47.

Remove the correction from both bridges: rescale_abs no longer takes a
scale, new_absolute()/new_touchscreen() drop the output_scale parameter,
the --output-scale CLI flag is gone from the touch and tablet bridges,
and the host no longer queries the sway scale or passes it through.
Confirmed 1:1 end-to-end against the live Scratch repro (edge-to-edge
drag plus corner/center taps). The tablet bridge fix has no live repro
(no tablet on hand) but follows the same reasoning and needs re-testing
on real hardware.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MYSRycxV3iG9PHjHqArLjg

albert merged commit f8b238719a into main 2026-07-12 18:35:03 +00:00

albert referenced this pull request from a commit 2026-07-12 18:35:04 +00:00

Merge pull request 'fix(input): stop dividing touch/tablet bridge coords by output scale (#47)' (#93) from u/albert/47/touch-compat-scale-offset into main

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

bug

Something is not working

  

duplicate

This issue or pull request already exists

  

enhancement

New feature

  

future

  

help wanted

Need some help

  

invalid

Something is wrong

  

question

More information is needed

  

wontfix

This won't be fixed

No labels

bug

duplicate

enhancement

future

help wanted

invalid

question

wontfix

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

albert/shepherd-launcher!93

No description provided.