Repeated components — Kanji Morphology

These are counts in one KanjiVG snapshot. They are not a rule that the first copy must sit on the left, the top, or the upper left, and not a rule that the first copy must change its stroke types.

Copy
Name: The name is kvg:original when the copy’s groups record exactly one value, and kvg:element otherwise. 亻 with original 人 is named 人. In this snapshot no copy records two different originals.
Sibling copies: Two or more copies of one name whose outermost groups are children of the same parent group. 林 has one sibling set: the two 木 under the character group. In 品 the two lower 口 share a parent. The upper 口 does not share that parent. A character is counted once, however many sibling sets it has.
Copies anywhere in the tree: Two or more copies of one name anywhere under the character group. One character and one name are one set. 品 is one set of three 口. 森 is one set of three 木. The group named 林 is not a fourth 木. Arrangement, the place of the first copy, and the sequence comparison use this set. A sibling set can be a proper part of it.
Arrangement
Written first
Sequence: The kvg:type of each stroke in the copy, in stroke order. A letter stays. A slash stays one string: ㇔/㇏ is not ㇏. Same means the tuples are equal. When every copy after the first shares one tuple and the first differs, the first is changed if the length is the same, fewer if it is shorter, and more if it is longer. Mixed means those later copies do not share one tuple.

KanjiVG · CC BY-SA 3.0

林 · two copies of 木. The left copy is written first.

Cinnabar is the copy written first, indigo the second, ochre the third, and celadon the fourth. Click a plate to replay. With reduced motion the finished character stays still, and the colors stay.

First copy
Second copy
Third copy
Fourth copy

How many characters

One row is one scope. Characters counts each jōyō character once. Sets counts each component name inside a character once. A character can contain more than one repeated name. 競 repeats 立, 兄, 口, 亠, 丿, and 八. The file writes that last pair as 儿 with original 八.

Sibling, and anywhere

Scope	Characters	Sets

How many copies

How the copies are arranged

Tree sets are all the copies of one name. Sibling sets are the copies that share a parent, so a triangle of three can contain a sibling pair of two. The one sibling triangle is the three 丿 in 派.

Left–right, top–bottom, triangle, other

Arrangement	Sibling sets	Tree sets

Which copy is written first

Place of the first copy

Place	Sets

Outside left, top, and upper left

One row is one tree set whose first copy is not on the left, the top, or the upper left. Strokes run from the copy written first to the copy written last.

Character	Component	Copies	Arrangement	Place	Strokes	Types

Whether the sequence stays

Same sequence, or the first copy differs

Sequence	Sets

The first copy has fewer strokes

Every tree set in which the copies after the first share one sequence, and the first copy is shorter.

Character	Component	Copies	Arrangement	Place	Strokes	Types

The later copies do not agree

Every tree set in which the copies after the first do not share one sequence.

Character	Component	Copies	Arrangement	Place	Strokes	Types

Where the file renames a copy

The name is not the element alone

Character	Component	Copies	Arrangement	Place	Strokes	Types

What this shows

Stroke diagrams: KanjiVG, Ulrich Apel et al., CC BY-SA 3.0, from commit 70a0b7ae0c18ceb5cb358274b029cce0234a43bc of github.com/KanjiVG/kanjivg. The plate files are unmodified copies in site/assets/kanjivg/. The sequences are the kvg:type values in those files. Jōyō index: the list used on the Gallery.