[string-lerp.git] / string-lerp.js

/* string-lerp - progressively turn one string into another
   Copyright 2014 Joe Wreschnig
  
   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.
*/

/* @license Copyright 2014 Joe Wreschnig - GNU GPL v2 or later */

(function (exports) {
    "use strict";

    var MAX_MATRIX_SIZE = 256 * 256;

    function costMatrix(source, target, ins, del, sub) {
        /** Calculate the Levenshtein cost matrix for source and target

            If source and target are strings, they cannot contain any
            astral or combining codepoints. Such data must be passed
            as arrays of strings with one element per glyph.

            ins, del, and sub are the costs for insertion, deletion,
            and substition respectively. Their default value is 1. If
            only ins is passed, del and sub are set to the same cost.
            If ins and del are passed, sub is set to the more
            expensive of the two.

            The matrix is returned as a flat typed array.

            Following http://en.wikipedia.org/wiki/Levenshtein_distance
        */
        ins = ins === undefined ? 1 : (ins | 0);
        del = (del | 0) || ins;
        sub = (sub | 0) || Math.max(ins, del);
        var m = source.length + 1;
        var n = target.length + 1;
        var d = new Uint32Array(m * n);
        var i, j;
        for (i = 1; i < m; ++i)
            d[n * i] = i;
        for (j = 1; j < n; ++j)
            d[j] = j;
        for (j = 1; j < n; ++j)
            for (i = 1; i < m; ++i)
                if (source[i - 1] === target[j - 1])
                    d[n * i + j] = d[n * (i - 1) + j - 1];
                else
                    d[n * i + j] = Math.min(del + d[n * (i - 1) + j    ],
                                            ins + d[n *    i    + j - 1],
                                            sub + d[n * (i - 1) + j - 1]);
        return d;
    }

    // First, note that deletion is just substition with nothing, so
    // any DEL operation can be replaced by a SUB. Second, the
    // operation code *is* the necessary slice offset for applying the
    // diff.
    var INS = 0, SUB = 1;

    function editPath(costs, target) {
        /** Given a cost matrix and a target, create an edit list */
        var path = [];
        var j = target.length;
        var n = j + 1;
        var i = costs.length / n - 1;
        while (i || j) {
            var sub = (i && j) ? costs[n * (i - 1) + j - 1] : Infinity;
            var del = i ? costs[n * (i - 1) + j] : Infinity;
            var ins = j ? costs[n * i + j - 1] : Infinity;
            if (sub <= ins && sub <= del) {
                if (costs[n * i + j] !== costs[n * (i - 1) + j - 1])
                    path.push([SUB, i - 1, target[j - 1]]);
                --i; --j;
            } else if (ins <= del) {
                path.push([INS, i, target[j - 1]]);
                --j;
            } else {
                path.push([SUB, i - 1, ""]);
                --i;
            }
        }
        return path;
    }

    function diff(source, target, ins, del, sub) {
        /** Create a list of edits to turn source into target

            ins, del, and sub are as passed to costMatrix.
        */
        return editPath(costMatrix(source, target, ins, del, sub), target);
    }

    function patchArray(diff, source) {
        for (var i = 0; i < diff.length; ++i) {
            var edit = diff[i];
            source.splice(edit[1], edit[0], edit[2]);
        }
        return source;
    }

    function patchString(diff, source) {
        for (var i = 0; i < diff.length; ++i) {
            var edit = diff[i];
            var head = source.slice(0, edit[1]);
            var tail = source.slice(edit[1] + edit[0]);
            source = head + edit[2] + tail;
        }
        return source;
    }

    function patch(diff, source) {
        /** Apply a list of edits to source */
        var patcher = Array.isArray(source) ? patchArray : patchString;
        return patcher(diff, source);
    }

    var MULTI = /[\uD800-\uDBFF][\uDC00-\uDFFF]|[\u0300-\u036F\u1DC0-\u1DFF\u20D0-\u20FF\uFE20-\uFE2F]/;

    var GLYPH = /([\0-\u02FF\u0370-\u1DBF\u1E00-\u20CF\u2100-\uD7FF\uDC00-\uFE1F\uFE30-\uFFFF]|[\uD800-\uDBFF][\uDC00-\uDFFF]|[\uD800-\uDBFF])([\u0300-\u036F\u1DC0-\u1DFF\u20D0-\u20FF\uFE20-\uFE2F]*)/g;

    function diffLerp(source, target, amount ) {
        /** Interpolate between two strings using edit operations

            This interpolation algorithm applys a partial edit of one
            string into the other. This produces nice looking results,
            but can take a significant amount of time and memory to
            compute the edits. It is not recommended for strings
            longer than a few hundred characters.
        */

        // If given strings with astral codepoints or combining
        // characters, split them into arrays of "glyphs" first,
        // do the edit on the list of "glyphs", and rejoin them.
        //
        // This split is not perfect for all languages, but at least
        // it won't create invalid surrogate pairs or orphaned
        // combining characters.
        //
        // TODO: The way this is called is a hack.
        if (source.match && (source.match(MULTI) || target.match(MULTI))) {
            var sourceGlyphs = source.match(GLYPH) || [];
            var targetGlyphs = target.match(GLYPH) || [];
            return diffLerp(sourceGlyphs, targetGlyphs, amount).join("");
        }

        // The edit path works from the string end, forwards, because
        // that's how Levenshtein edits work. To match LTR reading
        // direction (and the behavior of fastLerp), swap the strings
        // and invert the parameter when editing.
        var edits = diff(target, source, 2, 2, 3);
        var partial = edits.slice(0, Math.round((1 - amount) * edits.length));
        return patch(partial, target);
    }

    var NUMBERS = /(-?\d{1,20}(?:\.\d{1,20})?)/g;

    function areNumericTwins(source, target) {
        /** Check if a and b differ only in numerals */
        return source.replace(NUMBERS, "0") === target.replace(NUMBERS, "0");
    }

    function nlerp(source, target, amount) {
        return source + (target - source) * amount;
    }

    function numericLerp(source, target, amount) {
        /** Interpolate numerically between strings containing numbers

            Numbers may have a leading "-" and a single "." to mark
            the decimal point, but something must be after the ".".
            No other floating point syntax (e.g. 1e6) is supported.
            If both of the numbers in a pair are integers, the result
            is clamped to an integer.

            For example, numericLerp("0.0", "100", 0.123) === "12.3"
            because the "." in "0.0" is interpreted as a decimal
            point. But numericLerp("0.", "100.", 0.123) === "12."
            because the strings are interpreted as integers followed
            by a full stop.

            Calling this functions on strings that differ in more than
            numerals gives undefined results.
        */

        var sourceParts = source.split(NUMBERS);
        var targetParts = target.split(NUMBERS);
        var destParts = targetParts;
        for (var i = 1; i < sourceParts.length; i += 2) {
            var sourcePart = sourceParts[i];
            var targetPart = targetParts[i];
            var part = nlerp(+sourcePart, +targetPart, amount);
            var sourcePoint = sourcePart.indexOf(".");
            var targetPoint = targetPart.indexOf(".");
            var point = Math.round(nlerp(
                sourcePoint >= 0 ? (sourcePart.length - 1) - sourcePoint : 0,
                targetPoint >= 0 ? (targetPart.length - 1) - targetPoint : 0,
                amount));
            targetParts[i] = part.toFixed(point);
        }
        return targetParts.join("");
    }

    function fastLerpAstral(source, target, amount) {
        var sourceGlyphs = source.match(GLYPH) || [];
        var targetGlyphs = target.match(GLYPH) || [];
        var sourceLength = Math.round(sourceGlyphs.length * amount);
        var targetLength = Math.round(targetGlyphs.length * amount);
        var head = targetGlyphs.slice(0, targetLength);
        var tail = sourceGlyphs.slice(sourceLength, sourceGlyphs.length);
        head.push.apply(head, tail);
        return head.join("");
    }

    function fastLerpBasic(source, target, amount) {
        var sourceLength = Math.round(source.length * amount);
        var targetLength = Math.round(target.length * amount);
        var head = target.substring(0, targetLength);
        var tail = source.substring(sourceLength, source.length);
        return head + tail;
    }

    function fastLerp(source, target, amount) {
        /** Interpolate between two strings based on length

            This interpolation algorithm progressively replaces the
            front of one string with another. This approach is fast
            but does not look good when the strings are similar.
        */

        // TODO: Consider fast-pathing this even more for very large
        // strings, e.g. in the megabyte range. These are large enough
        // that it should be fine to just pick a codepoint and search
        // for the nearest glyph start.
        if (source.match(MULTI) || target.match(MULTI))
            return fastLerpAstral(source, target, amount);
        else
            return fastLerpBasic(source, target, amount);
    }

    function lerp(source, target, amount) {
        /** Interpolate between two strings as best as possible

            If the strings are identical aside from numbers in them,
            they are passed through numericLerp.

            If the strings are not numbers and short, they are passed
            through diffLerp.

            Otherwise, they are passed through fastLerp.
        */
        source = source.toString();
        target = target.toString();

        // Fast path for boundary cases.
        if (amount === 0) return source;
        if (amount === 1) return target;

        if (areNumericTwins(source, target))
            return numericLerp(source, target, amount);

        // Numeric lerps should over- and under-shoot when fed numbers
        // outside 0 to 1, but other types cannot.
        if (amount < 0) return source;
        if (amount > 1) return target;

        var n = source.length * target.length;
        var appropriate = (n && n < MAX_MATRIX_SIZE) ? diffLerp : fastLerp;
        return appropriate(source, target, amount);
    }

    exports.costMatrix = costMatrix;
    exports.patch = patch;
    exports.diff = diff;
    exports.fastLerp = fastLerp;
    exports.diffLerp = diffLerp;
    exports.numericLerp = numericLerp;
    exports.lerp = lerp;

})(typeof exports === "undefined" ? (this.stringLerp = {}) : exports);