Module glyphs 

Terminal glyphs, with an ASCII fallback for terminals not doing UTF-8 (SSH, a C locale), where Unicode would render as replacement boxes.

No Nerd Font glyphs. Every Unicode character passes the font-coverage audit (scripts/code/audit_glyphs.py): present in JetBrainsMono Nerd Font, and never an Emoji=Yes, Emoji_Presentation=No codepoint missing from Liberation Mono or Noto Sans Mono (those fall back to the color-emoji font and draw blank or clipped). Nerd Font icons appear only in the Omarchy menu and in users’ own [glyphs] overrides.

Structs

Glyphs

Symbols used by the UI, in whichever alphabet the terminal can render.

PlotMarks

The marks ratatui’s Chart, Canvas and BarChart draw, and their axis and legend lines; ratatui ignores the locale, so each set names its own.

Enums

SlotOverride

One [glyphs] override from the config: a single glyph, or a list for the slots that hold one (spinner, score_marks, mini_bars, bar_eighths).

UnicodeMode

What the user asked for, from [display] unicode.

Functions

active_is_unicode

Whether the active set is Unicode (copies of consts have no identifying address).

ascii

The ASCII set.

asciify_instructions

Instructional text mapped to ASCII for the help overlay on a non-UTF-8 terminal, at the render boundary only (user data is never transliterated). Coverage is checked by every_help_screen_is_ascii_clean.

cell_width

Cells text takes in a table cell, grapheme by grapheme at ratatui’s widths: unlike display_width, a joined emoji sequence counts once and control characters zero.

display_width

Display columns text occupies; scalar counts get CJK and combining marks wrong.

dotted

Text written with · between its parts, in the glyph set’s middot: - on an ASCII terminal. A · in a UI string is this template, drawn through here.

environment_is_utf8

Whether the terminal can be trusted with UTF-8: LC_ALL over LC_CTYPE over LANG, as POSIX; with none set, only Windows counts. The locale, not terminal capability, decides whether multi-byte characters render.

fit

text in width columns, cut at its end and marked with the ellipsis.

fit_cells

text fitted to width cells: whole if it fits, else cut at a grapheme boundary and closed with marker (never half a wide character). Control characters are dropped, as ratatui drops them. If marker itself does not fit, as much as does.

fit_middle

text in width columns, cut from the middle: weather/…/daily.

fit_start

text in width columns, cut at its start: a path keeps its leaf. The ellipsis is measured, since it is three columns on an ASCII terminal.

get

The active glyph set. Falls back to detection when init_with_overrides was never called, so library users and tests get sensible symbols without ceremony.

init_with_overrides

Choose the glyph set for this run, with [glyphs] overrides over the Unicode set. Never over the ASCII set: it is the tested floor, where rich-font overrides would garble. Later calls are ignored.

instructions_in_ascii

text with every instructional character replaced by its (wider) ASCII twin. Help rows (key, two+ spaces, description) are re-padded per section, so descriptions stay aligned and a section shifts right together when a key outgrows its column.

take_columns

The longest prefix of text that fits width display columns, never splitting a wide character.

take_columns_end

The longest suffix of text that fits width display columns, never splitting a wide character.

unicode

The Unicode set, for tests and for callers that know their output is UTF-8.

validate_overrides

Check a [glyphs] override map at load time, naming the bad slot. Each override must keep its glyph’s display width, so every layout width invariant holds.