@markii/ansi 0.15.0 → 0.17.0-next.0

package/dist/measure.d.ts

@@ -1,25 +0,0 @@
-/**
- * Hand-written display-width measurement (no dependency is permitted here;
- * see AGENTS.md's Stack section and the batch brief). This is a DELIBERATE
- * APPROXIMATION of Unicode's East Asian Width property (UAX #11), not a
- * full Unicode width table: it covers the ranges most terminal content
- * actually uses (Latin text, CJK, common emoji, combining marks) and gets
- * one class of case knowingly wrong, documented below.
- *
- * KNOWN LIMITATION: a grapheme cluster joined by a zero-width joiner
- * (U+200D), such as a multi-person or multi-skin-tone emoji sequence,
- * measures as the sum of its parts rather than as the single glyph a
- * terminal actually draws. Getting this exactly right needs a real
- * grapheme-segmentation table, which this package deliberately does not
- * carry.
- */
-/** Removes every SGR and OSC 8 escape sequence this engine can itself produce, so `measure` counts only what a terminal actually draws. */
-export declare function stripAnsi(text: string): string;
-/**
- * The display width of `text`: ANSI escapes are stripped first, then every
- * remaining code point (iterated as a code point, not a UTF-16 code unit,
- * so a surrogate pair counts once) contributes its width per
- * `codePointWidth`'s rules. This is what every wrapping/padding/column
- * helper in `./box.ts` measures against.
- */
-export declare function measure(text: string): number;

package/dist/measure.js

@@ -1,108 +0,0 @@
-/**
- * Hand-written display-width measurement (no dependency is permitted here;
- * see AGENTS.md's Stack section and the batch brief). This is a DELIBERATE
- * APPROXIMATION of Unicode's East Asian Width property (UAX #11), not a
- * full Unicode width table: it covers the ranges most terminal content
- * actually uses (Latin text, CJK, common emoji, combining marks) and gets
- * one class of case knowingly wrong, documented below.
- *
- * KNOWN LIMITATION: a grapheme cluster joined by a zero-width joiner
- * (U+200D), such as a multi-person or multi-skin-tone emoji sequence,
- * measures as the sum of its parts rather than as the single glyph a
- * terminal actually draws. Getting this exactly right needs a real
- * grapheme-segmentation table, which this package deliberately does not
- * carry.
- */
-/** SGR: `ESC [ ... m`. */
-const SGR_PATTERN = /\x1b\[[0-9;]*m/g;
-/** OSC 8 hyperlink open/close: `ESC ] 8 ; ; ... BEL`. */
-const OSC8_PATTERN = /\x1b\]8;;[^\x07\x1b]*\x07/g;
-/** Removes every SGR and OSC 8 escape sequence this engine can itself produce, so `measure` counts only what a terminal actually draws. */
-export function stripAnsi(text) {
-    return text.replace(OSC8_PATTERN, '').replace(SGR_PATTERN, '');
-}
-/** Combining marks: measure 0, since a terminal draws them layered onto the preceding column rather than advancing the cursor. */
-const COMBINING_RANGES = [
-    [0x0300, 0x036f],
-    [0x1ab0, 0x1aff],
-    [0x1dc0, 0x1dff],
-    [0x20d0, 0x20f0],
-    [0xfe00, 0xfe0f],
-    [0xfe20, 0xfe2f],
-];
-/** Zero-width characters: measure 0 (zero-width space/joiner/non-joiner, word joiner, byte-order mark used as a zero-width no-break space). */
-const ZERO_WIDTH_RANGES = [
-    [0x200b, 0x200d],
-    [0x2060, 0x2060],
-    [0xfeff, 0xfeff],
-];
-/** East Asian Wide and Fullwidth: measure 2. */
-const WIDE_RANGES = [
-    [0x1100, 0x115f],
-    [0x2e80, 0x303e],
-    [0x3041, 0x33ff],
-    [0x3400, 0x4dbf],
-    [0x4e00, 0x9fff],
-    [0xa000, 0xa4cf],
-    [0xac00, 0xd7a3],
-    [0xf900, 0xfaff],
-    [0xfe30, 0xfe6f],
-    [0xff00, 0xff60],
-    [0xffe0, 0xffe6],
-    [0x20000, 0x2fffd],
-    [0x30000, 0x3fffd],
-];
-/** Common emoji ranges: measure 2. U+2600-U+27BF is included only when followed by the U+FE0F variation selector (see `measureText`). */
-const EMOJI_RANGES = [
-    [0x1f300, 0x1f64f],
-    [0x1f680, 0x1f6ff],
-    [0x1f900, 0x1f9ff],
-    [0x1fa70, 0x1faff],
-];
-const EMOJI_PRESENTATION_LOW = 0x2600;
-const EMOJI_PRESENTATION_HIGH = 0x27bf;
-const VARIATION_SELECTOR_16 = 0xfe0f;
-function inRanges(code, ranges) {
-    for (const [low, high] of ranges) {
-        if (code >= low && code <= high)
-            return true;
-    }
-    return false;
-}
-/**
- * The display width of one code point, given the NEXT code point in the
- * stream (needed only to check the emoji-variation-selector rule for the
- * dingbat range U+2600-U+27BF).
- */
-function codePointWidth(code, next) {
-    if (inRanges(code, COMBINING_RANGES))
-        return 0;
-    if (inRanges(code, ZERO_WIDTH_RANGES))
-        return 0;
-    if (inRanges(code, WIDE_RANGES))
-        return 2;
-    if (inRanges(code, EMOJI_RANGES))
-        return 2;
-    if (code >= EMOJI_PRESENTATION_LOW &&
-        code <= EMOJI_PRESENTATION_HIGH &&
-        next === VARIATION_SELECTOR_16) {
-        return 2;
-    }
-    return 1;
-}
-/**
- * The display width of `text`: ANSI escapes are stripped first, then every
- * remaining code point (iterated as a code point, not a UTF-16 code unit,
- * so a surrogate pair counts once) contributes its width per
- * `codePointWidth`'s rules. This is what every wrapping/padding/column
- * helper in `./box.ts` measures against.
- */
-export function measure(text) {
-    const stripped = stripAnsi(text);
-    const codePoints = Array.from(stripped, (char) => char.codePointAt(0) ?? 0);
-    let width = 0;
-    for (let index = 0; index < codePoints.length; index += 1) {
-        width += codePointWidth(codePoints[index], codePoints[index + 1]);
-    }
-    return width;
-}