Reclaim Type List row width and read counts as a fraction - #1870
Merged
Conversation
PaulHax
force-pushed
the
type-list-density
branch
3 times, most recently
from
August 28, 2026 22:21
f7048aa to
857d470
Compare
PaulHax
marked this pull request as ready for review
August 28, 2026 22:22
PaulHax
force-pushed
the
type-list-density
branch
from
August 28, 2026 22:43
857d470 to
e0d5743
Compare
PaulHax
marked this pull request as draft
August 28, 2026 22:43
PaulHax
force-pushed
the
type-list-density
branch
from
August 28, 2026 22:49
e0d5743 to
46ee700
Compare
PaulHax
marked this pull request as ready for review
August 28, 2026 22:49
Tree indentation, `total : frame` counts, and the row buttons share one 30px row, and at the four and five levels a real taxonomy reaches, the type name is the part that loses. Indentation. The step drops from 16px to 12px and the disclosure column from 24px to 20px, which still clears the 16px chevron. Tree rows also give up the checkbox's left padding, which the disclosure column already supplies, and half of Vuetify's 8px gap between the checkbox and its label. A row at depth d gets 4d + 16 pixels back. The step and the column width each had two hard-coded copies, one driving the margin and one driving the label's ellipsis budget; a constant now feeds each so truncation cannot drift from the visible indent. Flat rows keep their padding and gap and have no disclosure column, so datasets without a hierarchy render exactly as before. Counts. A row read `12 : 3 red-snapper` - total, colon, frame count - where the colon reads as a ratio or a label separator and nothing tells the reader which side is which. They are now the fraction they are, part over whole: `3 / 12 red-snapper`. The frame count is the total narrowed by the frame's interval tree, per-frame suppression, and a keyframe test, so it is always a subset; on multicam it is selected-camera scoped while the total counts each logical track once, also a subset. The smaller number is therefore always the numerator and the fraction needs no legend. A test asserts that invariant, since it is an emergent property of two independently computed maps and a break would print `13 / 12`. Either count can be hidden from Type Settings, both switches defaulting on and following `timelineCountSettings.totalCount` for naming and hydration. The divider belongs to the fraction, so it survives only while both counts do; a lone number is ambiguous, so the row tooltip names which count it is and the header tooltip describes the format on screen. Sorting by either count is independent of showing it. Building the rows also dereferenced frame counts unconditionally, running an interval-tree search plus a suppression pass on every frame advance; they now share the guard the model already used, so hiding the frame count stops paying for it during playback. Compaction. Compact Parents walked down the unused chain above the list and stopped at the branch point without absorbing it, leaving the node whose subtree is the whole list as a row that indents everything beneath it. That node is where compaction is worth the most: unused and unconfigured, its checkbox does what the header checkbox does, its rolled-up count is the list total, and styling it changes no annotation because nothing displays as it. It is absorbed now, with the guard that a used or configured ancestor keeps its row, and the single-node threshold standing. Search still expands the breadcrumb wholesale. Subtrees and check state are computed before compaction; header actionability is derived from the final visible rows afterward. The breadcrumb itself now truncates at the front. Its deepest entry is its most specific and the one worth reading, and that was the end being cut. An inner span holds character order steady, because the `direction: rtl` that moves the ellipsis also runs the bidi algorithm over the names and would reorder digit and punctuation runs.
PaulHax
force-pushed
the
type-list-density
branch
from
August 29, 2026 01:15
46ee700 to
c5fa3f4
Compare
BryonLewis
approved these changes
Aug 29, 2026
BryonLewis
left a comment
BryonLewis
left a comment
Collaborator
There was a problem hiding this comment.
I'll handle in another PR the one comment I have about the changes from the PR with testing.
Comment on lines
+1
to
+2
| // @vitest-environment jsdom | ||
| /// <reference types="vitest" /> |
Collaborator
There was a problem hiding this comment.
I think after this merges they may need to be updated to be more in line with #1872
I may just doa secondary PR.
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.
Reclaim Type List row width and read counts as a fraction
Follow-up to Bryon's #1867 review.
Deep hierarchy rows run out of room for the type name.
Changes
3 / 12 red-snapperinstead of12 : 3 red-snapper.character order for names containing digits or punctuation.