Skip to content

fix: preserve BigInt precision above Number.MAX_SAFE_INTEGER #355

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
avoidwork merged 3 commits into from
Sep 27, 2026
Merged

fix: preserve BigInt precision above Number.MAX_SAFE_INTEGER #355

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e43b17c
feat: add OpenSpec change for BigInt precision fix
avoidwork Sep 26, 2026
4dd4f4a
fix: preserve BigInt precision above Number.MAX_SAFE_INTEGER
avoidwork Sep 27, 2026
f025ba9
chore: archive fix-bigint-precision OpenSpec change
avoidwork Sep 27, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
125 changes: 107 additions & 18 deletions dist/filesize.cjs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,6 +228,77 @@ function calculateOptimizedValue(num, e, isDecimal, bits, ceil, autoExponent = t
return { result, e };
}

/**
* Calculates the unit exponent for a bigint input using bigint comparisons
* @param {bigint} num - Input file size in bytes
* @param {number} e - Current exponent value
* @param {number} exponent - Original user-provided exponent option (-1 for auto)
* @param {boolean} isDecimal - Whether to use decimal (SI) base
* @param {number} precision - Current precision value (modified when e > 8)
* @returns {Object} Object with computed e value and possibly adjusted precision
*/
function calculateBigIntExponent(num, e, exponent, isDecimal, precision) {
if (typeof e === "string") {
e = Number(e);
}

if (e === -1 || isNaN(e)) {
e = 0;
if (isDecimal) {
while (e < 8 && num >= 10n ** BigInt(3 * (e + 1))) {
e++;
}
} else {
while (e < 8 && num >= 1024n ** BigInt(e + 1)) {
e++;
}
}
} else if (e < 0) {
e = 0;
} else {
e = Math.floor(e);
}

if (e > 8) {
if (precision > 0) {
precision += 8 - e;
}
return { e: 8, precision };
}

return { e, precision };
}

/**
* Calculates the value for a bigint input using bigint arithmetic
* @param {bigint} num - Input file size in bytes
* @param {number} e - Current exponent
* @param {boolean} isDecimal - Whether using decimal base
* @param {boolean} bits - Whether to calculate bits
* @param {number} ceil - Ceiling value for auto-increment
* @param {boolean} autoExponent - Whether exponent is auto (-1 or NaN)
* @returns {Object} Object with result and e properties
*/
function calculateBigIntValue(num, e, isDecimal, bits, ceil, autoExponent = true) {
const power = isDecimal ? 10n ** BigInt(3 * e) : 1024n ** BigInt(e);
// Scaled division preserves precision above Number.MAX_SAFE_INTEGER.
// Multiply by 10^16 before dividing, then scale back down, so the
// quotient keeps ~16 significant digits instead of collapsing to a float.
const SHIFT = 10n ** 16n;
let result = Number((num * SHIFT) / power) / Number(SHIFT);

if (bits) {
result *= 8;
// Handle auto-increment for bits (only when exponent is auto)
if (autoExponent && result >= ceil && e < 8) {
result /= ceil;
e++;
}
}

return { result, e };
}

/**
* Optimized precision handling with scientific notation correction
* @param {number} value - Current value
Expand Down Expand Up @@ -667,6 +738,8 @@ function filesize(
val = 0,
u = EMPTY;

const isBigInt = typeof arg === "bigint";

num = Number(arg);

if (isNaN(num)) {
Expand Down Expand Up @@ -706,27 +779,43 @@ function filesize(
);
}

// BigInt inputs use bigint arithmetic to preserve precision above
// Number.MAX_SAFE_INTEGER and detect unit boundaries accurately.
const bigNum = isBigInt ? (neg ? -BigInt(arg) : BigInt(arg)) : null;

// Exponent calculation + clamp + precision adjustment
const { e: calculatedE, precision: precisionAdjusted } = calculateExponent(
num,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
let precisionAdjusted = precision;
if (isBigInt) {
const { e: calculatedE, precision: pa } = calculateBigIntExponent(
bigNum,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
precisionAdjusted = pa;
} else {
const { e: calculatedE, precision: pa } = calculateExponent(
num,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
precisionAdjusted = pa;
}
const autoExponent = exponent === -1 || isNaN(exponent);

const { result: valueResult, e: valueExponent } = calculateOptimizedValue(
num,
e,
isDecimal,
bits,
ceil,
autoExponent,
);
val = valueResult;
e = valueExponent;
let valueResult;
if (isBigInt) {
valueResult = calculateBigIntValue(bigNum, e, isDecimal, bits, ceil, autoExponent);
} else {
valueResult = calculateOptimizedValue(num, e, isDecimal, bits, ceil, autoExponent);
}
val = valueResult.result;
e = valueResult.e;

// Rounding + auto-increment ceiling
const rounded = applyRounding(val, ceil, e, round, roundingFunc, autoExponent);
Expand Down
125 changes: 107 additions & 18 deletions dist/filesize.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,6 +224,77 @@ function calculateOptimizedValue(num, e, isDecimal, bits, ceil, autoExponent = t
return { result, e };
}

/**
* Calculates the unit exponent for a bigint input using bigint comparisons
* @param {bigint} num - Input file size in bytes
* @param {number} e - Current exponent value
* @param {number} exponent - Original user-provided exponent option (-1 for auto)
* @param {boolean} isDecimal - Whether to use decimal (SI) base
* @param {number} precision - Current precision value (modified when e > 8)
* @returns {Object} Object with computed e value and possibly adjusted precision
*/
function calculateBigIntExponent(num, e, exponent, isDecimal, precision) {
if (typeof e === "string") {
e = Number(e);
}

if (e === -1 || isNaN(e)) {
e = 0;
if (isDecimal) {
while (e < 8 && num >= 10n ** BigInt(3 * (e + 1))) {
e++;
}
} else {
while (e < 8 && num >= 1024n ** BigInt(e + 1)) {
e++;
}
}
} else if (e < 0) {
e = 0;
} else {
e = Math.floor(e);
}

if (e > 8) {
if (precision > 0) {
precision += 8 - e;
}
return { e: 8, precision };
}

return { e, precision };
}

/**
* Calculates the value for a bigint input using bigint arithmetic
* @param {bigint} num - Input file size in bytes
* @param {number} e - Current exponent
* @param {boolean} isDecimal - Whether using decimal base
* @param {boolean} bits - Whether to calculate bits
* @param {number} ceil - Ceiling value for auto-increment
* @param {boolean} autoExponent - Whether exponent is auto (-1 or NaN)
* @returns {Object} Object with result and e properties
*/
function calculateBigIntValue(num, e, isDecimal, bits, ceil, autoExponent = true) {
const power = isDecimal ? 10n ** BigInt(3 * e) : 1024n ** BigInt(e);
// Scaled division preserves precision above Number.MAX_SAFE_INTEGER.
// Multiply by 10^16 before dividing, then scale back down, so the
// quotient keeps ~16 significant digits instead of collapsing to a float.
const SHIFT = 10n ** 16n;
let result = Number((num * SHIFT) / power) / Number(SHIFT);

if (bits) {
result *= 8;
// Handle auto-increment for bits (only when exponent is auto)
if (autoExponent && result >= ceil && e < 8) {
result /= ceil;
e++;
}
}

return { result, e };
}

/**
* Optimized precision handling with scientific notation correction
* @param {number} value - Current value
Expand Down Expand Up @@ -661,6 +732,8 @@ function filesize(
val = 0,
u = EMPTY;

const isBigInt = typeof arg === "bigint";

num = Number(arg);

if (isNaN(num)) {
Expand Down Expand Up @@ -700,27 +773,43 @@ function filesize(
);
}

// BigInt inputs use bigint arithmetic to preserve precision above
// Number.MAX_SAFE_INTEGER and detect unit boundaries accurately.
const bigNum = isBigInt ? (neg ? -BigInt(arg) : BigInt(arg)) : null;

// Exponent calculation + clamp + precision adjustment
const { e: calculatedE, precision: precisionAdjusted } = calculateExponent(
num,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
let precisionAdjusted = precision;
if (isBigInt) {
const { e: calculatedE, precision: pa } = calculateBigIntExponent(
bigNum,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
precisionAdjusted = pa;
} else {
const { e: calculatedE, precision: pa } = calculateExponent(
num,
e,
exponent,
isDecimal,
precision,
);
e = calculatedE;
precisionAdjusted = pa;
}
const autoExponent = exponent === -1 || isNaN(exponent);

const { result: valueResult, e: valueExponent } = calculateOptimizedValue(
num,
e,
isDecimal,
bits,
ceil,
autoExponent,
);
val = valueResult;
e = valueExponent;
let valueResult;
if (isBigInt) {
valueResult = calculateBigIntValue(bigNum, e, isDecimal, bits, ceil, autoExponent);
} else {
valueResult = calculateOptimizedValue(num, e, isDecimal, bits, ceil, autoExponent);
}
val = valueResult.result;
e = valueResult.e;

// Rounding + auto-increment ceiling
const rounded = applyRounding(val, ceil, e, round, roundingFunc, autoExponent);
Expand Down
2 changes: 1 addition & 1 deletion dist/filesize.min.js
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion dist/filesize.min.js.map
View file
Open in desktop

Large diffs are not rendered by default.

Loading
Loading