GitHub

Character Structures

IDS files use Unicode ideographic description characters to describe how components of characters are put together (they occupy the range from U+2FF0 to U+2FFB). In IDSGraphs.jl, these relationships are represented as character structures, similar to a parse tree.

IDSGraphs.parse — Method
IDSGraphs.parse(::AbstractString)

Parse a single IDS string that describes a character/component's decomposition. (This function does not do any input validation to verify that the string can actually represent a decomposition; in this case, the method will fail silently.)

This is a relatively low-level method; you typically do not need to call this directly, but it is available if you are parsing your own IDS strings.

Examples

We can take the IDS string itself and parse it as a nested structure.

julia> using IDSGraphs: parse

julia> parse("⿱此二")
CharStructure{TopBottomStructure}(TopBottomStructure(Component('此'), Component('二')))
source

Structure Hierarchy

IDSGraphs.NestedStructure — Type

Supertype for structures/relationships that are "nested" (one component within the other, touching at least 2 sides).

source

Basic Relationships

These are the simplest component relationships: components are all either horizontally or vertically stacked on top of each other, in a set of 2 or 3 components.

Nested Relationships

Most relationships have one component in the "middle" with another component "around" the middle component. The names reflect where the middle component is placed.

Note that not all of the expected positions are covered. There is no

  • LeftNestedStructure
  • TopLeftNestedStructure

Overlap Relationship

There's one more relationship different from all the others: