PickleKit Test Report

Features

10 passed, 0 failed

Scenarios

431

431 passed, 0 failed

Steps

1117

1117 passed, 0 failed, 0 skipped

The Anzan language itself

48/48 scenarios passed · 0µs

Operator precedence and associativityoutline · 6passed

2 + 3 * 4, 14, # multiplicative over additivepassed2ms

WhenI calculate "2 + 3 * 4"2ms

Thenthe result is "14"64µs

2^3^2, 512, # ^ is right-associativepassed200µs

WhenI calculate "2^3^2"157µs

Thenthe result is "512"39µs

-2^2, -4, # unary minus binds looser than ^passed126µs

WhenI calculate "-2^2"90µs

Thenthe result is "-4"34µs

2^-2, 0.25, # the exponent may carry its own signpassed702µs

WhenI calculate "2^-2"661µs

Thenthe result is "0.25"37µs

[10, 2][0]^2, 100, # postfix indexing binds tighter than ^passed213µs

WhenI calculate "[10, 2][0]^2"177µs

Thenthe result is "100"33µs

√4 + 5, 7, # prefix √ is unary, not greedypassed1ms

WhenI calculate "√4 + 5"965µs

Thenthe result is "7"36µs

Number literalsoutline · 6passed

1_000 + 0, 1000passed145µs

WhenI calculate "1_000 + 0"109µs

Thenthe result is "1000"34µs

2.5e-3, 0.0025passed106µs

WhenI calculate "2.5e-3"71µs

Thenthe result is "0.0025"33µs

.5 + .5, 1passed113µs

WhenI calculate ".5 + .5"77µs

Thenthe result is "1"33µs

0xFF, 255passed87µs

WhenI calculate "0xFF"52µs

Thenthe result is "255"32µs

0b1010, 10passed92µs

WhenI calculate "0b1010"59µs

Thenthe result is "10"31µs

0xDEAD_BEEF, 3735928559passed89µs

WhenI calculate "0xDEAD_BEEF"54µs

Thenthe result is "3735928559"33µs

Pinned cell references evaluate like plain onesoutline · 3passed

$A:$1 * 2passed289µs

Givencell A:1 contains "21"148µs

WhenI calculate "$A:$1 * 2"105µs

Thenthe result is "42"32µs

$A:1 + A:1passed176µs

Givencell A:1 contains "21"57µs

WhenI calculate "$A:1 + A:1"82µs

Thenthe result is "42"33µs

A:$1 * 2passed155µs

Givencell A:1 contains "21"53µs

WhenI calculate "A:$1 * 2"66µs

Thenthe result is "42"33µs

A dangling $ is a loud lex erroroutline · 3passed

$passed93µs

WhenI calculate "$"53µs

Thenthe calculation fails mentioning "pins a cell reference"37µs

$xpassed73µs

WhenI calculate "$x"37µs

Thenthe calculation fails mentioning "pins a cell reference"34µs

2 + $ratepassed94µs

WhenI calculate "2 + $rate"58µs

Thenthe calculation fails mentioning "pins a cell reference"34µs

A comment-only line is a note, not an errorpassed94µs

WhenI calculate "# the calc below confirms our test"47µs

Thenthe result is "# the calc below confirms our test"40µs

A standalone note never touches anspassed268µs

WhenI calculate "21 * 2"107µs

AndI calculate "# just thinking out loud"42µs

AndI calculate "ans"74µs

Thenthe result is "42"37µs

A trailing comment is stripped before evaluationpassed125µs

WhenI calculate "5 + 3 # adds them"87µs

Thenthe result is "8"35µs

A hash inside a string is not a commentpassed113µs

WhenI calculate "len("a # b")"74µs

Thenthe result is "5"36µs

Malformed programmer literals fail loudlyoutline · 4passed

0xFG, malformed hexpassed116µs

WhenI calculate "0xFG"71µs

Thenthe calculation fails mentioning "malformed hex"41µs

0x1.5, malformed hexpassed92µs

WhenI calculate "0x1.5"53µs

Thenthe calculation fails mentioning "malformed hex"36µs

0x, needs digitspassed73µs

WhenI calculate "0x"38µs

Thenthe calculation fails mentioning "needs digits"33µs

0b12, malformed binarypassed81µs

WhenI calculate "0b12"45µs

Thenthe calculation fails mentioning "malformed binary"32µs

Implicit multiplication covers parens, names, and constantspassed311µs

WhenI calculate "x9 = 4"135µs

AndI calculate "(2)(3) + 2x9"134µs

Thenthe result is "14"36µs

Modulo is exactpassed135µs

WhenI calculate "mod(0.3, 0.1)"105µs

Thenthe result is "0"27µs

Percent literalsoutline · 5passed

3%, 0.03passed97µs

WhenI calculate "3%"62µs

Thenthe result is "0.03"32µs

1 * 3%, 0.03passed170µs

WhenI calculate "1 * 3%"130µs

Thenthe result is "0.03"38µs

100 + 5%, 100.05passed173µs

WhenI calculate "100 + 5%"125µs

Thenthe result is "100.05"44µs

50% * 2, 1passed151µs

WhenI calculate "50% * 2"113µs

Thenthe result is "1"33µs

(2 + 3)%, 0.05passed137µs

WhenI calculate "(2 + 3)%"106µs

Thenthe result is "0.05"28µs

A number can't directly follow another valueoutline · 3passed

3 4passed101µs

WhenI calculate "3 4"61µs

Thenthe calculation fails mentioning "operator"37µs

3 % 4passed100µs

WhenI calculate "3 % 4"66µs

Thenthe calculation fails mentioning "operator"33µs

(1) 2passed155µs

WhenI calculate "(1) 2"79µs

Thenthe calculation fails mentioning "operator"72µs

A string is not a conditionpassed221µs

WhenI calculate "if("a", 1, 2)"184µs

Thenthe calculation fails mentioning "number"33µs

Empty reductions yield identitiespassed329µs

WhenI calculate "∑_i=1^0(i) + ∏_i=1^0(i)"207µs

Thenthe result is "1"118µs

A reduction spanning too many terms is refusedpassed253µs

WhenI calculate "∑_i=1^200000(i)"207µs

Thenthe calculation fails mentioning "100,000"42µs

Reserved words refuse assignmentoutline · 3passed

ans = 5passed120µs

WhenI calculate "ans = 5"74µs

Thenthe calculation fails mentioning "cannot assign"42µs

sigma = 1passed105µs

WhenI calculate "sigma = 1"69µs

Thenthe calculation fails mentioning "cannot assign"33µs

true = 0passed89µs

WhenI calculate "true = 0"54µs

Thenthe calculation fails mentioning "cannot assign"33µs

Parameters shadow globals without clobbering thempassed429µs

WhenI calculate "shade = 10"76µs

AndI calculate "twice(shade) = shade * 2"166µs

AndI calculate "twice(3) + shade"148µs

Thenthe result is "16"34µs

Free variables resolve at call timepassed341µs

WhenI calculate "base = 10"67µs

AndI calculate "above(y) = base + y"83µs

AndI calculate "base = 100"72µs

AndI calculate "above(1)"80µs

Thenthe result is "101"33µs

Special forms ship their manualpassed105µs

WhenI calculate "man(if)"66µs

Thendocumentation is shown mentioning "taken branch"36µs

String escapes are honoredpassed97µs

WhenI calculate "len("a\tb")"64µs

Thenthe result is "3"31µs

The word data is still an ordinary namepassed161µs

WhenI calculate "data = 5"59µs

AndI calculate "data * 2"68µs

Thenthe result is "10"31µs

Compact named arguments versus cell referencespassed452µs

GivenI calculate "data Q { a: Number, age: Number }"111µs

WhenI calculate "Q(a: 1, age:36).age"160µs

Thenthe result is "36"29µs

WhenI calculate "sum(a:1)"113µs

Thenthe result is "0"32µs

Calculating in the log

25/25 scenarios passed · 0µs

Everyday calculations are exactoutline · 10passed

0.1 + 0.2, 0.3passed146µs

WhenI calculate "0.1 + 0.2"105µs

Thenthe result is "0.3"38µs

2(3 + 4), 14passed140µs

WhenI calculate "2(3 + 4)"104µs

Thenthe result is "14"34µs

2^10, 1024passed129µs

WhenI calculate "2^10"92µs

Thenthe result is "1024"34µs

∑(1, 2, 3), 6passed158µs

WhenI calculate "∑(1, 2, 3)"124µs

Thenthe result is "6"32µs

∑_i=1^10(i^2), 385passed425µs

WhenI calculate "∑_i=1^10(i^2)"391µs

Thenthe result is "385"31µs

pmt(0.05/12, 360, 200000), -1073.643246024277969656985158225109053609679713701passed1.94s

WhenI calculate "pmt(0.05/12, 360, 200000)"1.94s

Thenthe result is "-1073.643246024277969656985158225109053609679713701"109µs

margin(100, 80), 20passed784µs

WhenI calculate "margin(100, 80)"743µs

Thenthe result is "20"37µs

date(2026, 6, 6) - date(2026, 1, 1), 156passed275µs

WhenI calculate "date(2026, 6, 6) - date(2026, 1, 1)"238µs

Thenthe result is "156"33µs

if(2 > 1, 10, 20), 10passed178µs

WhenI calculate "if(2 > 1, 10, 20)"142µs

Thenthe result is "10"33µs

gcd(48, 36), 12passed212µs

WhenI calculate "gcd(48, 36)"173µs

Thenthe result is "12"34µs

A hundred dimes make exactly a dollar bagpassed1ms

WhenI calculate "∑_i=1^100(0.1)"1ms

Thenthe result is "10"70µs

The previous answer carries forwardpassed227µs

WhenI calculate "6 * 7"108µs

AndI calculate "ans + 8"83µs

Thenthe result is "50"32µs

Variables persist across calculationspassed209µs

WhenI calculate "rate2026 = 0.0825"80µs

AndI calculate "1200 * rate2026"95µs

Thenthe result is "99"31µs

A leading equals sign is toleratedpassed106µs

WhenI calculate "= 1 + 2"74µs

Thenthe result is "3"30µs

A failed calculation never clobbers the previous answerpassed225µs

WhenI calculate "6 * 7"69µs

AndI calculate "1 / 0"61µs

AndI calculate "ans + 0"60µs

Thenthe result is "42"30µs

Dividing by zero explains itselfpassed102µs

WhenI calculate "1 / 0"60µs

Thenthe calculation fails mentioning "division by zero"37µs

Typos are caught, not guessedpassed133µs

WhenI calculate "12 * rte"95µs

Thenthe calculation fails mentioning "unknown variable 'rte'"34µs

Comparisons answer 1 or 0outline · 7passed

2 < 3, 1passed104µs

WhenI calculate "2 < 3"70µs

Thenthe result is "1"32µs

3 < 2, 0passed96µs

WhenI calculate "3 < 2"67µs

Thenthe result is "0"26µs

3 > 2, 1passed120µs

WhenI calculate "3 > 2"83µs

Thenthe result is "1"33µs

2 <= 2, 1passed106µs

WhenI calculate "2 <= 2"72µs

Thenthe result is "1"31µs

2 >= 3, 0passed90µs

WhenI calculate "2 >= 3"63µs

Thenthe result is "0"25µs

2 == 2, 1passed97µs

WhenI calculate "2 == 2"64µs

Thenthe result is "1"30µs

2 != 2, 0passed87µs

WhenI calculate "2 != 2"59µs

Thenthe result is "0"25µs

Comparison chains are rejected, not misreadpassed114µs

WhenI calculate "1 < 2 < 3"77µs

Thenthe calculation fails mentioning "can't be chained"35µs

Data types — declared records with named construction

47/47 scenarios passed · 0µs

Declaring a type registers its constructorpassed371µs

WhenI calculate "data Person { name: String, age: Number, active: Boolean }"134µs

Thenthe result is "data Person { name: String, age: Number, active: Boolean }"28µs

WhenI calculate "Person(name: "Ada", age: 36, active: true)"165µs

Thenthe result is "Person(name: "Ada", age: 36, active: true)"38µs

Construction is named fields or one map — never positionalpassed604µs

GivenI calculate "data Pt { x: Number, y: Number }"97µs

WhenI calculate "Pt(y: 4, x: 3)"105µs

Thenthe result is "Pt(x: 3, y: 4)"35µs

WhenI calculate "m = {x: 3, y: 4}"98µs

AndI calculate "Pt(m)"72µs

Thenthe result is "Pt(x: 3, y: 4)"46µs

WhenI calculate "Pt(3, 4)"106µs

Thenthe calculation fails mentioning "takes named fields"34µs

Field types accept any casing and Boolean fields render true/falsepassed255µs

WhenI calculate "data Flag { on: boolean }"85µs

Thenthe result is "data Flag { on: Boolean }"28µs

WhenI calculate "Flag(on: 1 < 2)"109µs

Thenthe result is "Flag(on: true)"28µs

Fields read like map members, by dot or by keypassed2ms

GivenI calculate "data Pt { x: Number, y: Number }"95µs

AndI calculate "p = Pt(x: 3, y: 4)"117µs

WhenI calculate "sqrt(p.x^2 + p.y^2)"1ms

Thenthe result is "5"54µs

WhenI calculate "p["y"]"70µs

Thenthe result is "4"31µs

WhenI calculate "keys(p)"67µs

Thenthe result is "["x", "y"]"60µs

WhenI calculate "len(p)"66µs

Thenthe result is "2"29µs

Records collect into arrays and flow through higher-order functionspassed1ms

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"132µs

AndI calculate "team = [Person(name: "Ada", age: 36, active: true), Person(name: "Grace", age: 30, active: false)]"266µs

WhenI calculate "map(x -> x.age, team)"102µs

Thenthe result is "[36, 30]"37µs

WhenI calculate "sum(map(x -> x.age, team))"126µs

Thenthe result is "66"29µs

WhenI calculate "filter(x -> x.active, team)"93µs

Thenthe result is "[Person(name: "Ada", age: 36, active: true)]"38µs

WhenI calculate "map(Person, [{name: "Bo", age: 1, active: true}])"161µs

Thenthe result is "[Person(name: "Bo", age: 1, active: true)]"36µs

Instances canonicalize to declaration order and compare deeplypassed527µs

GivenI calculate "data Pt { x: Number, y: Number }"95µs

WhenI calculate "Pt(y: 2, x: 1) == Pt(x: 1, y: 2)"178µs

Thenthe result is "1"33µs

WhenI calculate "Pt(x: 1, y: 2) == {x: 1, y: 2}"186µs

Thenthe result is "0"26µs

toJson is pretty by default, compact on request, honest about Booleanspassed628µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"128µs

AndI calculate "p = Person(name: "Ada", age: 36, active: true)"147µs

WhenI calculate "toJson(p)"71µs

Thenthe result is ""{\n \"name\": \"Ada\",\n \"age\": 36,\n \"active\": true\n}""34µs

WhenI calculate "toJson(p, Json.Compact)"87µs

Thenthe result is ""{\"name\":\"Ada\",\"age\":36,\"active\":true}""31µs

WhenI calculate "toJson([1, 2])"95µs

Thenthe result is ""[\n 1,\n 2\n]""28µs

fromJson parses JSON into values, exactlypassed811µs

WhenI calculate "fromJson("[1, 2, 3]")"106µs

Thenthe result is "[1, 2, 3]"41µs

WhenI calculate "fromJson("{\"name\": \"Ada\", \"age\": 36}").age"99µs

Thenthe result is "36"30µs

WhenI calculate "fromJson("0.30000000000000004") == 0.30000000000000004"101µs

Thenthe result is "1"29µs

WhenI calculate "fromJson("123456789012345678901234567890.5") * 2"113µs

Thenthe result is "246913578024691357802469135781"37µs

WhenI calculate "fromJson("true") + fromJson("false")"127µs

Thenthe result is "1"29µs

WhenI calculate "fromJson("\"\\u0041da\"")"62µs

Thenthe result is ""Ada""26µs

fromJson and a constructor re-type a serialized recordpassed707µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"122µs

AndI calculate "p = Person(name: "Ada", age: 36, active: true)"145µs

WhenI calculate "Person(fromJson(toJson(p))) == p"140µs

Thenthe result is "1"29µs

WhenI calculate "map(Person, fromJson(toJson([p, p])))[1].age"232µs

Thenthe result is "36"29µs

fromJson mistakes explain themselvesoutline · 5passed

fromJson("null"), has no Anzan valuepassed112µs

WhenI calculate "fromJson("null")"76µs

Thenthe calculation fails mentioning "has no Anzan value"33µs

fromJson("[1,"), unexpected endpassed191µs

WhenI calculate "fromJson("[1,")"155µs

Thenthe calculation fails mentioning "unexpected end"33µs

fromJson("[1] junk"), trailing contentpassed115µs

WhenI calculate "fromJson("[1] junk")"81µs

Thenthe calculation fails mentioning "trailing content"31µs

fromJson("{\"a\":1,\"a\":2}"), duplicate keypassed121µs

WhenI calculate "fromJson("{\"a\":1,\"a\":2}")"88µs

Thenthe calculation fails mentioning "duplicate key"31µs

fromJson(5), wants JSON textpassed106µs

WhenI calculate "fromJson(5)"74µs

Thenthe calculation fails mentioning "wants JSON text"30µs

Json options are named constants, not magic flagspassed551µs

WhenI calculate "Json.Pretty"58µs

Thenthe result is ""pretty""27µs

WhenI calculate "toJson([1, 2], Json.Compact)"128µs

Thenthe result is ""[1,2]""25µs

WhenI calculate "toJson([1, 2], "compact")"105µs

Thenthe result is ""[1,2]""26µs

WhenI calculate "Json = 5"53µs

Thenthe calculation fails mentioning "cannot assign"31µs

WhenI calculate "man(Json)"57µs

Thendocumentation is shown mentioning "Formatting options"31µs

Construction mistakes explain themselvesoutline · 9passed

Person(name: "Ada", age: 36), missing 'active'passed284µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"135µs

WhenI calculate "Person(name: "Ada", age: 36)"117µs

Thenthe calculation fails mentioning "missing 'active'"29µs

Person(name: 7, age: 36, active: true), 'name' of Person is a Stringpassed393µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"123µs

WhenI calculate "Person(name: 7, age: 36, active: true)"220µs

Thenthe calculation fails mentioning "'name' of Person is a String"44µs

Person(name: "A", age: "x", active: true), 'age' of Person is a Numberpassed318µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"133µs

WhenI calculate "Person(name: "A", age: "x", active: true)"138µs

Thenthe calculation fails mentioning "'age' of Person is a Number"42µs

Person(name: "A", age: 36, active: 7), use true or falsepassed320µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"133µs

WhenI calculate "Person(name: "A", age: 36, active: 7)"150µs

Thenthe calculation fails mentioning "use true or false"32µs

Person(name: "A", age: 36, active: true, pet: 1), no field 'pet'passed343µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"125µs

WhenI calculate "Person(name: "A", age: 36, active: true, pet: 1)"180µs

Thenthe calculation fails mentioning "no field 'pet'"32µs

Person(name: "A", name: "B", age: 1, active: 1), duplicate fieldpassed307µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"124µs

WhenI calculate "Person(name: "A", name: "B", age: 1, active: 1)"147µs

Thenthe calculation fails mentioning "duplicate field"32µs

toJson(sqrt), can't serialize a functionpassed252µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"129µs

WhenI calculate "toJson(sqrt)"72µs

Thenthe calculation fails mentioning "can't serialize a function"40µs

toJson(1, "wat"), unknown toJson optionpassed268µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"136µs

WhenI calculate "toJson(1, "wat")"97µs

Thenthe calculation fails mentioning "unknown toJson option"32µs

toJson(1, true), Json.Pretty or Json.Compactpassed251µs

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"124µs

WhenI calculate "toJson(1, true)"92µs

Thenthe calculation fails mentioning "Json.Pretty or Json.Compact"32µs

Declaration mistakes explain themselvesoutline · 5passed

data person { a: Number }, capital letterpassed105µs

WhenI calculate "data person { a: Number }"71µs

Thenthe calculation fails mentioning "capital letter"31µs

data Bad { a: truthy }, declared data typepassed104µs

WhenI calculate "data Bad { a: truthy }"70µs

Thenthe calculation fails mentioning "declared data type"32µs

data Empty {}, at least one fieldpassed96µs

WhenI calculate "data Empty {}"61µs

Thenthe calculation fails mentioning "at least one field"32µs

data Dup { a: Number, A: Number }, duplicate fieldpassed154µs

WhenI calculate "data Dup { a: Number, A: Number }"109µs

Thenthe calculation fails mentioning "duplicate field"42µs

data Abs { a: Number }, built-inpassed130µs

WhenI calculate "data Abs { a: Number }"94µs

Thenthe calculation fails mentioning "built-in"33µs

Types and functions can't share a name, but redeclaring your own type workspassed725µs

GivenI calculate "f(x) = x + 1"126µs

WhenI calculate "data F { a: Number }"82µs

Thenthe calculation fails mentioning "already a function"34µs

WhenI calculate "data G { a: Number }"83µs

AndI calculate "g(x) = x"82µs

Thenthe calculation fails mentioning "is a data type"43µs

WhenI calculate "data G { a: Number, b: Number }"100µs

AndI calculate "G(a: 1, b: 2).b"123µs

Thenthe result is "2"40µs

A type documents itself through its trailing commentpassed253µs

WhenI calculate "data Invoice { total: Number, paid: Boolean } # one customer invoice"111µs

AndI calculate "man(Invoice)"63µs

Thendocumentation is shown mentioning "one customer invoice"75µs

toJson escapes strings and renders empty containerspassed295µs

WhenI calculate "toJson("a\tb")"81µs

Thenthe result is ""\"a\\tb\"""34µs

WhenI calculate "toJson([])"63µs

Thenthe result is ""[]""26µs

WhenI calculate "toJson({})"57µs

Thenthe result is ""{}""26µs

A constructor call works in tail position of a recursive functionpassed492µs

GivenI calculate "data Box { v: Number }"79µs

AndI calculate "wrap(n) = if(n <= 0, Box(v: n), wrap(n - 1))"196µs

WhenI calculate "wrap(5).v"188µs

Thenthe result is "0"25µs

The explicit formula marker rejects declarations in cellspassed153µs

Givencell A:1 contains "=data Pt { x: Number }"105µs

Thencell A:1 shows an error mentioning "drop the leading '='"45µs

A plain declaration in a cell is a sheet-scoped 𝑫 typepassed522µs

Givencell A:1 contains "data Pt { x: Number, y: Number }"153µs

Andcell B:1 contains "Pt(x: 3, y: 4).x"118µs

Thencell A:1 shows "𝑫 Pt"46µs

Andcell B:1 shows "3"74µs

WhenI calculate "data Pt { x: Number }"79µs

Thenthe calculation fails mentioning "defined in cell Sheet 1!A:1"41µs

Types and record variables survive save and reopenpassed1ms

GivenI calculate "data Person { name: String, age: Number, active: Boolean }"145µs

AndI calculate "p = Person(name: "Ada", age: 36, active: true)"165µs

Whenthe workbook is saved and reopened857µs

AndI calculate "p.age + Person(name: "B", age: 4, active: false).age"207µs

Thenthe result is "40"36µs

A typed parameter dispatches a function by argument typepassed464µs

GivenI calculate "kind(n: Number) = "number""100µs

AndI calculate "kind(s: String) = "string""143µs

WhenI calculate "kind(42)"99µs

Thenthe result is ""number""27µs

WhenI calculate "kind("hi")"62µs

Thenthe result is ""string""26µs

A function can be written for a data typepassed2ms

GivenI calculate "data Point { x: Number, y: Number }"97µs

AndI calculate "midpoint(a: Point, b: Point) = Point(x: (a.x + b.x) / 2, y: (a.y + b.y) / 2)"255µs

WhenI calculate "midpoint(Point(x: 0, y: 0), Point(x: 4, y: 10))"1ms

Thenthe result is "Point(x: 2, y: 5)"56µs

An operator can be overloaded for a data typepassed611µs

GivenI calculate "data Point { x: Number, y: Number }"117µs

AndI calculate "+(a: Point, b: Point) = Point(x: a.x + b.x, y: a.y + b.y)"194µs

WhenI calculate "Point(x: 1, y: 2) + Point(x: 10, y: 20)"254µs

Thenthe result is "Point(x: 11, y: 22)"40µs

An overloaded operator can mix a data type and a scalarpassed493µs

GivenI calculate "data Point { x: Number, y: Number }"104µs

AndI calculate "*(a: Point, s: Number) = Point(x: a.x * s, y: a.y * s)"178µs

WhenI calculate "Point(x: 1, y: 2) * 3"167µs

Thenthe result is "Point(x: 3, y: 6)"38µs

Built-in arithmetic is untouched by an overloadpassed558µs

GivenI calculate "data Point { x: Number, y: Number }"96µs

AndI calculate "+(a: Point, b: Point) = Point(x: a.x + b.x, y: a.y + b.y)"180µs

WhenI calculate "1 + 2"111µs

Thenthe result is "3"74µs

WhenI calculate ""Q" + 1"63µs

Thenthe result is ""Q1""26µs

An operator overload must involve a data typepassed182µs

WhenI calculate "+(a: Number, b: Number) = 5"141µs

Thenthe calculation fails mentioning "must involve a data type"37µs

Records compare equal by all of their statepassed737µs

GivenI calculate "data Point { x: Number, y: Number }"98µs

WhenI calculate "Point(x: 1, y: 2) == Point(x: 1, y: 2)"180µs

Thenthe result is "1"32µs

WhenI calculate "Point(x: 1, y: 2) == Point(x: 9, y: 9)"169µs

Thenthe result is "0"31µs

WhenI calculate "Point(x: 1, y: 2) != Point(x: 9, y: 9)"187µs

Thenthe result is "1"33µs

Operator overloads survive save and reopenpassed2ms

GivenI calculate "data Point { x: Number, y: Number }"105µs

AndI calculate "+(a: Point, b: Point) = Point(x: a.x + b.x, y: a.y + b.y)"194µs

AndI calculate "*(a: Point, s: Number) = Point(x: a.x * s, y: a.y * s)"173µs

Whenthe workbook is saved and reopened557µs

AndI calculate "(Point(x: 1, y: 1) + Point(x: 2, y: 3)).y"275µs

Thenthe result is "4"32µs

AndI calculate "(Point(x: 2, y: 3) * 2).x"158µs

Thenthe result is "4"30µs

A data type can have fields of another data typepassed810µs

GivenI calculate "data Point { x: Number, y: Number }"111µs

AndI calculate "data Line { a: Point, b: Point }"101µs

AndI calculate "l = Line(a: Point(x: 1, y: 2), b: Point(x: 3, y: 4))"246µs

WhenI calculate "l.b.y"56µs

Thenthe result is "4"33µs

WhenI calculate "l == Line(a: Point(x: 1, y: 2), b: Point(x: 3, y: 4))"226µs

Thenthe result is "1"30µs

A nested field rejects the wrong typepassed440µs

GivenI calculate "data Point { x: Number, y: Number }"100µs

AndI calculate "data Line { a: Point, b: Point }"90µs

WhenI calculate "Line(a: 5, b: Point(x: 3, y: 4))"191µs

Thenthe calculation fails mentioning "is a Point"51µs

Nested records survive save and reopenpassed1ms

GivenI calculate "data Point { x: Number, y: Number }"110µs

AndI calculate "data Line { a: Point, b: Point }"93µs

AndI calculate "seg = Line(a: Point(x: 1, y: 1), b: Point(x: 4, y: 5))"240µs

Whenthe workbook is saved and reopened695µs

AndI calculate "seg.b.x - seg.a.x"106µs

Thenthe result is "3"34µs

Number formats are presentation, never value

15/15 scenarios passed · 0µs

A formatted cell displays its value in costumeoutline · 13passed

1234567.5, number, 1,234,567.50passed261µs

Givencell B:1 contains "1234567.5"81µs

Andcell B:1 is formatted as "number"76µs

Thencell B:1 displays "1,234,567.50"99µs

1234.5, dollars, $1,234.50passed150µs

Givencell B:1 contains "1234.5"70µs

Andcell B:1 is formatted as "dollars"34µs

Thencell B:1 displays "$1,234.50"42µs

-1234.5, dollars, -$1,234.50passed169µs

Givencell B:1 contains "-1234.5"73µs

Andcell B:1 is formatted as "dollars"35µs

Thencell B:1 displays "-$1,234.50"56µs

-2, euros, -€2.00passed143µs

Givencell B:1 contains "-2"56µs

Andcell B:1 is formatted as "euros"33µs

Thencell B:1 displays "-€2.00"51µs

0.0825, percent, 8.25%passed137µs

Givencell B:1 contains "0.0825"58µs

Andcell B:1 is formatted as "percent"33µs

Thencell B:1 displays "8.25%"42µs

1, percent, 100.00%passed127µs

Givencell B:1 contains "1"50µs

Andcell B:1 is formatted as "percent"32µs

Thencell B:1 displays "100.00%"42µs

20610, a date, 2026-06-06passed185µs

Givencell B:1 contains "20610"54µs

Andcell B:1 is formatted as "a date"69µs

Thencell B:1 displays "2026-06-06"57µs

0, a date, 1970-01-01passed108µs

Givencell B:1 contains "0"49µs

Andcell B:1 is formatted as "a date"27µs

Thencell B:1 displays "1970-01-01"29µs

195, hex, 0xC3passed175µs

Givencell B:1 contains "195"78µs

Andcell B:1 is formatted as "hex"44µs

Thencell B:1 displays "0xC3"50µs

-255, hex, -0xFFpassed145µs

Givencell B:1 contains "-255"64µs

Andcell B:1 is formatted as "hex"34µs

Thencell B:1 displays "-0xFF"43µs

195, binary, 0b1100_0011passed162µs

Givencell B:1 contains "195"49µs

Andcell B:1 is formatted as "binary"34µs

Thencell B:1 displays "0b1100_0011"75µs

5, binary, 0b101passed131µs

Givencell B:1 contains "5"54µs

Andcell B:1 is formatted as "binary"33µs

Thencell B:1 displays "0b101"41µs

1.5, hex, 1.5passed127µs

Givencell B:1 contains "1.5"55µs

Andcell B:1 is formatted as "hex"31µs

Thencell B:1 displays "1.5"37µs

Formatting never touches the underlying mathpassed338µs

Givencell B:1 contains "0.0825"55µs

Andcell B:1 is formatted as "percent"32µs

Andcell B:2 contains "=B:1 * 200"79µs

Thencell B:1 displays "8.25%"83µs

Andcell B:2 shows "16.5"84µs

Hex format is display-only, like every formatpassed342µs

Givencell A:1 contains "=bitOr(0xC0, 0x03)"99µs

Andcell A:1 is formatted as "hex"32µs

Andcell A:2 contains "=A:1 + 1"62µs

Thencell A:1 displays "0xC3"89µs

Andcell A:2 shows "196"54µs

Defining your own functions

16/16 scenarios passed · 0µs

Define a function and use itpassed274µs

WhenI calculate "tax(x) = x * 1.0825 # TX sales tax"133µs

AndI calculate "tax(100)"102µs

Thenthe result is "108.25"35µs

Functions compose regardless of definition orderpassed311µs

WhenI calculate "g(x) = f(x) + 1"104µs

AndI calculate "f(x) = x * 2"81µs

AndI calculate "g(20)"93µs

Thenthe result is "41"30µs

Recursion workspassed326µs

WhenI calculate "fact(n) = if(n <= 1, 1, n * fact(n - 1))"189µs

AndI calculate "fact(10)"103µs

Thenthe result is "3628800"30µs

Lambdas are valuespassed244µs

WhenI calculate "map(x -> x * 2, [1, 2, 3])"199µs

Thenthe result is "[2, 4, 6]"41µs

Structures carry datapassed318µs

WhenI calculate "people = [{name: "Ada", age: 36}, {name: "Bob", age: 32}]"195µs

AndI calculate "people[0].age"80µs

Thenthe result is "36"37µs

Machin's 1706 formula recovers pipassed12ms

WhenI calculate "arctanInv(x, n) = ∑_k=0^(n)((-1)^k / ((2k + 1) * x^(2k + 1)))"353µs

AndI calculate "16 * arctanInv(5, 40) - 4 * arctanInv(239, 15) - pi"11ms

Thenthe result is within "1e-45" of zero119µs

Comments are for humans and ignored by the mathpassed132µs

WhenI calculate "6 * 7 # the answer"98µs

Thenthe result is "42"32µs

A trailing comment becomes the function's documentationpassed230µs

WhenI calculate "tax(x) = x * 1.0825 # TX sales tax"131µs

AndI calculate "man(tax)"63µs

Thendocumentation is shown mentioning "TX sales tax"32µs

Built-in functions ship their manualpassed97µs

WhenI calculate "man(pmt)"60µs

Thendocumentation is shown mentioning "payment"33µs

Only the taken branch of if() runspassed153µs

WhenI calculate "if(1, 2, 1/0)"119µs

Thenthe result is "2"32µs

Deep recursion is bounded by memory, not a counterpassed57ms

WhenI calculate "countdown(n) = if(n <= 0, 0, countdown(n - 1) + 1)"177µs

AndI calculate "countdown(2000)"57ms

Thenthe result is "2000"76µs

Tail-recursive functions run at constant stack — any depthpassed12.63s

WhenI calculate "sumTo(n, acc) = if(n <= 0, acc, sumTo(n - 1, acc + n))"234µs

AndI calculate "sumTo(500000, 0)"12.63s

Thenthe result is "125000250000"85µs

Forgetting a base case fails politely, with a hintpassed185ms

WhenI calculate "F(n) = F(n - 1) + F(n - 2)"206µs

AndI calculate "f(3)"184ms

Thenthe calculation fails mentioning "base case"72µs

Runaway recursion is cut off cleanly, not a crashpassed11.01s

WhenI calculate "loop(n) = loop(n + 1)"153µs

AndI calculate "loop(1)"11.01s

Thenthe calculation fails mentioning "nested too deeply"91µs

Built-in names are protectedpassed146µs

WhenI calculate "abs(x) = x"107µs

Thenthe calculation fails mentioning "built-in"35µs

Function calls are case-insensitivepassed998µs

WhenI calculate "MIN(1, 2) + Sqrt(16)"957µs

Thenthe result is "5"38µs

The expanded function library

114/114 scenarios passed · 0µs

Exact combinatorics — BigInt keeps every digitoutline · 7passed

choose(5, 2), 10passed150µs

WhenI calculate "choose(5, 2)"114µs

Thenthe result is "10"33µs

choose(52, 5), 2598960passed185µs

WhenI calculate "choose(52, 5)"144µs

Thenthe result is "2598960"38µs

choose(100, 50), 100891344545564193334812497256passed366µs

WhenI calculate "choose(100, 50)"321µs

Thenthe result is "100891344545564193334812497256"42µs

choose(2, 5), 0passed129µs

WhenI calculate "choose(2, 5)"100µs

Thenthe result is "0"27µs

perm(5, 2), 20passed138µs

WhenI calculate "perm(5, 2)"104µs

Thenthe result is "20"32µs

perm(10, 3), 720passed144µs

WhenI calculate "perm(10, 3)"109µs

Thenthe result is "720"32µs

perm(10, 10), 3628800passed162µs

WhenI calculate "perm(10, 10)"127µs

Thenthe result is "3628800"33µs

Array plumbingoutline · 11passed

sort([3, 1, 2]), [1, 2, 3]passed169µs

WhenI calculate "sort([3, 1, 2])"126µs

Thenthe result is "[1, 2, 3]"41µs

sort(["pear", "fig", "kiwi"]), ["fig", "kiwi", "pear"]passed135µs

WhenI calculate "sort(["pear", "fig", "kiwi"])"94µs

Thenthe result is "["fig", "kiwi", "pear"]"38µs

unique([3, 1, 3, 2, 1]), [3, 1, 2]passed186µs

WhenI calculate "unique([3, 1, 3, 2, 1])"145µs

Thenthe result is "[3, 1, 2]"39µs

reverse([1, 2, 3]), [3, 2, 1]passed147µs

WhenI calculate "reverse([1, 2, 3])"107µs

Thenthe result is "[3, 2, 1]"38µs

reverse("abc"), "cba"passed88µs

WhenI calculate "reverse("abc")"59µs

Thenthe result is ""cba""27µs

seq(1, 5), [1, 2, 3, 4, 5]passed165µs

WhenI calculate "seq(1, 5)"118µs

Thenthe result is "[1, 2, 3, 4, 5]"45µs

seq(10, 0, -2), [10, 8, 6, 4, 2, 0]passed255µs

WhenI calculate "seq(10, 0, -2)"156µs

Thenthe result is "[10, 8, 6, 4, 2, 0]"82µs

sum(map(x -> x^2, seq(1, 10))), 385passed504µs

WhenI calculate "sum(map(x -> x^2, seq(1, 10)))"467µs

Thenthe result is "385"35µs

list(1, 2, 3), [1, 2, 3]passed143µs

WhenI calculate "list(1, 2, 3)"100µs

Thenthe result is "[1, 2, 3]"40µs

sumproduct([2, 3], [10, 100]), 320passed259µs

WhenI calculate "sumproduct([2, 3], [10, 100])"223µs

Thenthe result is "320"33µs

sumproduct(1, 2, 3, 4, 5, 6), 32passed223µs

WhenI calculate "sumproduct(1, 2, 3, 4, 5, 6)"188µs

Thenthe result is "32"32µs

list() turns a range into an array — higher-order functions over cellspassed568µs

Giventhe sheet contains:152µs

WhenI calculate "sum(filter(x -> x > 10, list(A:1..A:4)))"200µs

Thenthe result is "42"32µs

WhenI calculate "map(x -> x * 2, list(A:1..A:2))"141µs

Thenthe result is "[10, 24]"37µs

Statistics depth (sample conventions, full precision)outline · 11passed

variance(2, 4, 4, 4, 5, 5, 7, 9), 4.5714285714285714285714285714285714285714285714286passed1ms

WhenI calculate "variance(2, 4, 4, 4, 5, 5, 7, 9)"1ms

Thenthe result is "4.5714285714285714285714285714285714285714285714286"49µs

mode(1, 2, 2, 3, 3, 3), 3passed203µs

WhenI calculate "mode(1, 2, 2, 3, 3, 3)"164µs

Thenthe result is "3"37µs

mode(4, 9, 4, 9), 4passed151µs

WhenI calculate "mode(4, 9, 4, 9)"118µs

Thenthe result is "4"31µs

percentile(15, 20, 35, 40, 50, 0.4), 29passed270µs

WhenI calculate "percentile(15, 20, 35, 40, 50, 0.4)"236µs

Thenthe result is "29"31µs

percentile(1, 2, 3, 4, 0.75), 3.25passed230µs

WhenI calculate "percentile(1, 2, 3, 4, 0.75)"194µs

Thenthe result is "3.25"34µs

percentile(1, 2, 3, 1), 3passed171µs

WhenI calculate "percentile(1, 2, 3, 1)"137µs

Thenthe result is "3"31µs

geomean(4, 9), 6passed898µs

WhenI calculate "geomean(4, 9)"862µs

Thenthe result is "6"34µs

correl(1, 2, 3, 2, 4, 6), 1passed3ms

WhenI calculate "correl(1, 2, 3, 2, 4, 6)"3ms

Thenthe result is "1"34µs

slope(3, 5, 7, 1, 2, 3), 2passed2ms

WhenI calculate "slope(3, 5, 7, 1, 2, 3)"2ms

Thenthe result is "2"35µs

intercept(3, 5, 7, 1, 2, 3), 1passed2ms

WhenI calculate "intercept(3, 5, 7, 1, 2, 3)"2ms

Thenthe result is "1"33µs

forecast(4, 3, 5, 7, 1, 2, 3), 9passed2ms

WhenI calculate "forecast(4, 3, 5, 7, 1, 2, 3)"2ms

Thenthe result is "9"38µs

Amortization and depreciationoutline · 10passed

round(ipmt(0.05/12, 1, 360, 200000), 10), -833.3333333333passed1.80s

WhenI calculate "round(ipmt(0.05/12, 1, 360, 200000), 10)"1.80s

Thenthe result is "-833.3333333333"82µs

round(ipmt(0.05/12, 360, 360, 200000), 10), -4.4549512283passed3.77s

WhenI calculate "round(ipmt(0.05/12, 360, 360, 200000), 10)"3.77s

Thenthe result is "-4.4549512283"83µs

round(cumipmt(0.05/12, 360, 200000, 1, 12), 2), -9932.99passed1.78s

WhenI calculate "round(cumipmt(0.05/12, 360, 200000, 1, 12), 2)"1.78s

Thenthe result is "-9932.99"95µs

round(cumprinc(0.05/12, 360, 200000, 1, 12), 2), -2950.73passed1.78s

WhenI calculate "round(cumprinc(0.05/12, 360, 200000, 1, 12), 2)"1.78s

Thenthe result is "-2950.73"70µs

ipmt(0.05/12, 7, 360, 200000) + ppmt(0.05/12, 7, 360, 200000) == pmt(0.05/12, 360, 200000), 1passed5.32s

WhenI calculate "ipmt(0.05/12, 7, 360, 200000) + ppmt(0.05/12, 7, 360, 200000) == pmt(0.05/12, 360, 200000)"5.32s

Thenthe result is "1"102µs

sln(30000, 7500, 10), 2250passed897µs

WhenI calculate "sln(30000, 7500, 10)"854µs

Thenthe result is "2250"38µs

round(syd(30000, 7500, 10, 1), 2), 4090.91passed897µs

WhenI calculate "round(syd(30000, 7500, 10, 1), 2)"856µs

Thenthe result is "4090.91"38µs

ddb(30000, 7500, 10, 1), 6000passed775µs

WhenI calculate "ddb(30000, 7500, 10, 1)"735µs

Thenthe result is "6000"36µs

ddb(30000, 7500, 10, 10), 0passed1ms

WhenI calculate "ddb(30000, 7500, 10, 10)"1ms

Thenthe result is "0"37µs

round(nominal(effectiveRate(0.06, 12), 12), 10), 0.06passed972µs

WhenI calculate "round(nominal(effectiveRate(0.06, 12), 12), 10)"931µs

Thenthe result is "0.06"38µs

Business days and calendar positionsoutline · 9passed

quarter(date(2026, 6, 6)), 2passed192µs

WhenI calculate "quarter(date(2026, 6, 6))"156µs

Thenthe result is "2"34µs

quarter(date(2026, 11, 1)), 4passed179µs

WhenI calculate "quarter(date(2026, 11, 1))"144µs

Thenthe result is "4"33µs

weeknum(date(2026, 1, 1)), 1passed176µs

WhenI calculate "weeknum(date(2026, 1, 1))"141µs

Thenthe result is "1"33µs

weeknum(date(2026, 1, 4)), 2passed175µs

WhenI calculate "weeknum(date(2026, 1, 4))"139µs

Thenthe result is "2"34µs

workday(date(2026, 6, 5), 1) == date(2026, 6, 8), 1passed268µs

WhenI calculate "workday(date(2026, 6, 5), 1) == date(2026, 6, 8)"233µs

Thenthe result is "1"32µs

workday(date(2026, 6, 8), -1) == date(2026, 6, 5), 1passed271µs

WhenI calculate "workday(date(2026, 6, 8), -1) == date(2026, 6, 5)"237µs

Thenthe result is "1"32µs

networkdays(date(2026, 6, 1), date(2026, 6, 30)), 22passed265µs

WhenI calculate "networkdays(date(2026, 6, 1), date(2026, 6, 30))"231µs

Thenthe result is "22"32µs

networkdays(date(2026, 6, 1), date(2026, 6, 30), date(2026, 6, 19)), 21passed350µs

WhenI calculate "networkdays(date(2026, 6, 1), date(2026, 6, 30), date(2026, 6, 19))"317µs

Thenthe result is "21"31µs

networkdays(date(2026, 6, 5), date(2026, 6, 1)), -5passed263µs

WhenI calculate "networkdays(date(2026, 6, 5), date(2026, 6, 1))"225µs

Thenthe result is "-5"36µs

Scientific completionsoutline · 9passed

deg(pi), 180passed711µs

WhenI calculate "deg(pi)"674µs

Thenthe result is "180"34µs

deg(pi / 4), 45passed807µs

WhenI calculate "deg(pi / 4)"770µs

Thenthe result is "45"34µs

sin(rad(90)), 1passed320µs

WhenI calculate "sin(rad(90))"286µs

Thenthe result is "1"32µs

sinh(0), 0passed110µs

WhenI calculate "sinh(0)"82µs

Thenthe result is "0"26µs

cosh(0), 1passed105µs

WhenI calculate "cosh(0)"72µs

Thenthe result is "1"30µs

tanh(0), 0passed135µs

WhenI calculate "tanh(0)"85µs

Thenthe result is "0"44µs

asinh(0), 0passed144µs

WhenI calculate "asinh(0)"99µs

Thenthe result is "0"40µs

acosh(1), 0passed111µs

WhenI calculate "acosh(1)"81µs

Thenthe result is "0"27µs

atanh(0), 0passed112µs

WhenI calculate "atanh(0)"75µs

Thenthe result is "0"33µs

atan2 knows its quadrantpassed707µs

WhenI calculate "atan2(1, 1) - pi/4"272µs

Thenthe result is within "1e-15" of zero63µs

WhenI calculate "atan2(-1, -1) + 3 * pi/4"298µs

Thenthe result is within "1e-15" of zero70µs

Programmer tools — exact at any widthoutline · 14passed

toBase(255, 16), "FF"passed125µs

WhenI calculate "toBase(255, 16)"96µs

Thenthe result is ""FF""27µs

toBase(10, 2), "1010"passed117µs

WhenI calculate "toBase(10, 2)"93µs

Thenthe result is ""1010""22µs

fromBase("ff", 16), 255passed154µs

WhenI calculate "fromBase("ff", 16)"114µs

Thenthe result is "255"36µs

fromBase("1010", 2), 10passed131µs

WhenI calculate "fromBase("1010", 2)"101µs

Thenthe result is "10"28µs

toBase(-255, 16), "-FF"passed125µs

WhenI calculate "toBase(-255, 16)"98µs

Thenthe result is ""-FF""25µs

fromBase("-ff", 16), -255passed125µs

WhenI calculate "fromBase("-ff", 16)"90µs

Thenthe result is "-255"32µs

fromBase(toBase(123456789012345678901234567890, 36), 36), 123456789012345678901234567890passed233µs

WhenI calculate "fromBase(toBase(123456789012345678901234567890, 36), 36)"188µs

Thenthe result is "123456789012345678901234567890"41µs

bitAnd(12, 10), 8passed198µs

WhenI calculate "bitAnd(12, 10)"152µs

Thenthe result is "8"43µs

bitOr(12, 10), 14passed164µs

WhenI calculate "bitOr(12, 10)"118µs

Thenthe result is "14"43µs

bitXor(12, 10), 6passed149µs

WhenI calculate "bitXor(12, 10)"111µs

Thenthe result is "6"35µs

bitShift(1, 100), 1.267650600228229401496703205376e+30passed164µs

WhenI calculate "bitShift(1, 100)"117µs

Thenthe result is "1.267650600228229401496703205376e+30"44µs

bitShift(256, -4), 16passed140µs

WhenI calculate "bitShift(256, -4)"108µs

Thenthe result is "16"29µs

bitAnd(0xFF, 0x0F), 15passed180µs

WhenI calculate "bitAnd(0xFF, 0x0F)"122µs

Thenthe result is "15"52µs

toBase(0b1111, 16), "F"passed164µs

WhenI calculate "toBase(0b1111, 16)"122µs

Thenthe result is ""F""38µs

solve() is goal seek in a formulapassed253ms

WhenI calculate "solve(x -> x^2, 2) - sqrt(2)"1ms

Thenthe result is within "1e-12" of zero53µs

WhenI calculate "solve(cos, 0, 1) - pi/2"555µs

Thenthe result is within "1e-12" of zero57µs

WhenI calculate "solve(x -> if(x < 1, -1, 1), 0) - 1"1ms

Thenthe result is within "1e-12" of zero50µs

WhenI calculate "solve(r -> fv(r, 120, -100), 20000, 0.01) * 12"250ms

Thenthe result is within "1e-9" of "0.0958092381724"119µs

Derived builtins equal their definitionsoutline · 16passed

avg(2, 4, 9) == sum(2, 4, 9) / count(2, 4, 9)passed1ms

WhenI calculate "avg(2, 4, 9) == sum(2, 4, 9) / count(2, 4, 9)"1ms

Thenthe result is "1"44µs

stdev(2, 4, 4, 7) == sqrt(variance(2, 4, 4, 7))passed3ms

WhenI calculate "stdev(2, 4, 4, 7) == sqrt(variance(2, 4, 4, 7))"3ms

Thenthe result is "1"34µs

variance(2, 4) == ((2 - 3)^2 + (4 - 3)^2) / 1passed2ms

WhenI calculate "variance(2, 4) == ((2 - 3)^2 + (4 - 3)^2) / 1"2ms

Thenthe result is "1"37µs

cbrt(27) == root(27, 3)passed192µs

WhenI calculate "cbrt(27) == root(27, 3)"160µs

Thenthe result is "1"30µs

geomean(2, 4, 8) == root(product(2, 4, 8), 3)passed289µs

WhenI calculate "geomean(2, 4, 8) == root(product(2, 4, 8), 3)"246µs

Thenthe result is "1"37µs

percent(8.25) == 8.25 / 100passed1ms

WhenI calculate "percent(8.25) == 8.25 / 100"1ms

Thenthe result is "1"35µs

deg(2) == 2 * 180 / pipassed384µs

WhenI calculate "deg(2) == 2 * 180 / pi"350µs

Thenthe result is "1"32µs

choose(10, 4) == fact(10) / (fact(4) * fact(6))passed831µs

WhenI calculate "choose(10, 4) == fact(10) / (fact(4) * fact(6))"796µs

Thenthe result is "1"33µs

perm(10, 4) == fact(10) / fact(6)passed759µs

WhenI calculate "perm(10, 4) == fact(10) / fact(6)"726µs

Thenthe result is "1"30µs

lcm(12, 18) == 12 * 18 / gcd(12, 18)passed708µs

WhenI calculate "lcm(12, 18) == 12 * 18 / gcd(12, 18)"675µs

Thenthe result is "1"30µs

median(4, 1, 3) == sort([4, 1, 3])[1]passed246µs

WhenI calculate "median(4, 1, 3) == sort([4, 1, 3])[1]"202µs

Thenthe result is "1"38µs

sumproduct([2, 3], [10, 100]) == 2 * 10 + 3 * 100passed316µs

WhenI calculate "sumproduct([2, 3], [10, 100]) == 2 * 10 + 3 * 100"283µs

Thenthe result is "1"31µs

npv(0.1, 300, 420) == 300 / 1.1 + 420 / 1.1^2passed580µs

WhenI calculate "npv(0.1, 300, 420) == 300 / 1.1 + 420 / 1.1^2"545µs

Thenthe result is "1"32µs

forecast(4, 3, 5, 7, 1, 2, 3) == intercept(3, 5, 7, 1, 2, 3) + slope(3, 5, 7, 1, 2, 3) * 4passed6ms

WhenI calculate "forecast(4, 3, 5, 7, 1, 2, 3) == intercept(3, 5, 7, 1, 2, 3) + slope(3, 5, 7, 1, 2, 3) * 4"6ms

Thenthe result is "1"31µs

quarter(date(2026, 11, 1)) == trunc((month(date(2026, 11, 1)) - 1) / 3) + 1passed431µs

WhenI calculate "quarter(date(2026, 11, 1)) == trunc((month(date(2026, 11, 1)) - 1) / 3) + 1"399µs

Thenthe result is "1"29µs

bitShift(5, 3) == 5 * 2^3passed190µs

WhenI calculate "bitShift(5, 3) == 5 * 2^3"159µs

Thenthe result is "1"29µs

The new functions explain their mistakesoutline · 24passed

choose(-1, 2), non-negativepassed143µs

WhenI calculate "choose(-1, 2)"106µs

Thenthe calculation fails mentioning "non-negative"35µs

sort([1, "a"]), all numbers or all stringspassed118µs

WhenI calculate "sort([1, "a"])"85µs

Thenthe calculation fails mentioning "all numbers or all strings"31µs

seq(1, 10, 0), can't be 0passed152µs

WhenI calculate "seq(1, 10, 0)"121µs

Thenthe calculation fails mentioning "can't be 0"29µs

mode(1, 2, 3), no value repeatspassed149µs

WhenI calculate "mode(1, 2, 3)"113µs

Thenthe calculation fails mentioning "no value repeats"33µs

percentile(1, 2, 3, 1.5), between 0 and 1passed160µs

WhenI calculate "percentile(1, 2, 3, 1.5)"127µs

Thenthe calculation fails mentioning "between 0 and 1"31µs

geomean(4, -9), positivepassed124µs

WhenI calculate "geomean(4, -9)"93µs

Thenthe calculation fails mentioning "positive"29µs

correl(1, 2, 3, 4, 5), equal-lengthpassed161µs

WhenI calculate "correl(1, 2, 3, 4, 5)"129µs

Thenthe calculation fails mentioning "equal-length"30µs

ipmt(0.05, 0, 12, 1000), 1 ≤ per ≤ nperpassed187µs

WhenI calculate "ipmt(0.05, 0, 12, 1000)"136µs

Thenthe calculation fails mentioning "1 ≤ per ≤ nper"49µs

syd(30000, 7500, 10, 11), 1 ≤ per ≤ lifepassed210µs

WhenI calculate "syd(30000, 7500, 10, 11)"175µs

Thenthe calculation fails mentioning "1 ≤ per ≤ life"33µs

toBase(1.5, 16), integerpassed117µs

WhenI calculate "toBase(1.5, 16)"87µs

Thenthe calculation fails mentioning "integer"27µs

fromBase("xyz", 16), not a base-16passed115µs

WhenI calculate "fromBase("xyz", 16)"85µs

Thenthe calculation fails mentioning "not a base-16"27µs

bitAnd(-1, 2), non-negativepassed124µs

WhenI calculate "bitAnd(-1, 2)"93µs

Thenthe calculation fails mentioning "non-negative"28µs

solve(x -> x^2 + 1, 0), did not convergepassed1ms

WhenI calculate "solve(x -> x^2 + 1, 0)"1ms

Thenthe calculation fails mentioning "did not converge"29µs

ipmt(0.05, 1, 0, 1000), 1 ≤ per ≤ nperpassed175µs

WhenI calculate "ipmt(0.05, 1, 0, 1000)"139µs

Thenthe calculation fails mentioning "1 ≤ per ≤ nper"33µs

cumipmt(0.05/12, 12, 1000, 5, 2), 1 ≤ start ≤ endpassed273µs

WhenI calculate "cumipmt(0.05/12, 12, 1000, 5, 2)"238µs

Thenthe calculation fails mentioning "1 ≤ start ≤ end"33µs

nominal(-2, 12), above -100%passed137µs

WhenI calculate "nominal(-2, 12)"102µs

Thenthe calculation fails mentioning "above -100%"32µs

ddb(30000, 7500, 10, 11), 1 ≤ per ≤ lifepassed208µs

WhenI calculate "ddb(30000, 7500, 10, 11)"173µs

Thenthe calculation fails mentioning "1 ≤ per ≤ life"32µs

sln(1, 1, 0), can't be 0passed122µs

WhenI calculate "sln(1, 1, 0)"92µs

Thenthe calculation fails mentioning "can't be 0"27µs

toBase(255, 50), 2–36passed124µs

WhenI calculate "toBase(255, 50)"90µs

Thenthe calculation fails mentioning "2–36"32µs

fromBase("", 16), at least one digitpassed109µs

WhenI calculate "fromBase("", 16)"80µs

Thenthe calculation fails mentioning "at least one digit"27µs

bitShift(1, 1.5), integerpassed116µs

WhenI calculate "bitShift(1, 1.5)"87µs

Thenthe calculation fails mentioning "integer"27µs

seq(1, 1000000), 100,000passed931ms

WhenI calculate "seq(1, 1000000)"930ms

Thenthe calculation fails mentioning "100,000"84µs

workday(date(2026, 1, 1), 200001), too manypassed267µs

WhenI calculate "workday(date(2026, 1, 1), 200001)"221µs

Thenthe calculation fails mentioning "too many"42µs

networkdays(date(1, 1, 1), date(9999, 1, 1)), too manypassed301µs

WhenI calculate "networkdays(date(1, 1, 1), date(9999, 1, 1))"262µs

Thenthe calculation fails mentioning "too many"35µs

The mathematics a user can reach

105/105 scenarios passed · 0µs

Core arithmetic and algebraoutline · 28passed

1 + 2 * 3, 7passed133µs

WhenI calculate "1 + 2 * 3"95µs

Thenthe result is "7"35µs

(1 + 2) * 3, 9passed150µs

WhenI calculate "(1 + 2) * 3"106µs

Thenthe result is "9"40µs

10 / 4, 2.5passed658µs

WhenI calculate "10 / 4"614µs

Thenthe result is "2.5"41µs

mod(7, 3), 1passed137µs

WhenI calculate "mod(7, 3)"99µs

Thenthe result is "1"35µs

-2^2, -4passed136µs

WhenI calculate "-2^2"98µs

Thenthe result is "-4"35µs

2^-1, 0.5passed644µs

WhenI calculate "2^-1"606µs

Thenthe result is "0.5"35µs

abs(-5), 5passed119µs

WhenI calculate "abs(-5)"84µs

Thenthe result is "5"33µs

min(3, 1, 2), 1passed144µs

WhenI calculate "min(3, 1, 2)"109µs

Thenthe result is "1"32µs

max(3, 1, 2), 3passed138µs

WhenI calculate "max(3, 1, 2)"105µs

Thenthe result is "3"31µs

floor(2.7), 2passed121µs

WhenI calculate "floor(2.7)"77µs

Thenthe result is "2"39µs

floor(-1.5), -2passed153µs

WhenI calculate "floor(-1.5)"113µs

Thenthe result is "-2"37µs

ceil(2.1), 3passed119µs

WhenI calculate "ceil(2.1)"84µs

Thenthe result is "3"32µs

ceil(-1.5), -1passed143µs

WhenI calculate "ceil(-1.5)"108µs

Thenthe result is "-1"32µs

trunc(-2.7), -2passed119µs

WhenI calculate "trunc(-2.7)"86µs

Thenthe result is "-2"31µs

round(2.345, 2), 2.34passed140µs

WhenI calculate "round(2.345, 2)"103µs

Thenthe result is "2.34"34µs

round(2.567, 2), 2.57passed143µs

WhenI calculate "round(2.567, 2)"98µs

Thenthe result is "2.57"43µs

round(2.5), 2passed112µs

WhenI calculate "round(2.5)"79µs

Thenthe result is "2"30µs

mod(7, 3), 1passed130µs

WhenI calculate "mod(7, 3)"98µs

Thenthe result is "1"30µs

fact(5), 120passed119µs

WhenI calculate "fact(5)"86µs

Thenthe result is "120"31µs

fact(20), 2432902008176640000passed160µs

WhenI calculate "fact(20)"123µs

Thenthe result is "2432902008176640000"35µs

gcd(12, 18), 6passed153µs

WhenI calculate "gcd(12, 18)"115µs

Thenthe result is "6"35µs

lcm(4, 6), 12passed135µs

WhenI calculate "lcm(4, 6)"99µs

Thenthe result is "12"33µs

percent(8.25), 0.0825passed646µs

WhenI calculate "percent(8.25)"610µs

Thenthe result is "0.0825"33µs

sqrt(144), 12passed878µs

WhenI calculate "sqrt(144)"843µs

Thenthe result is "12"32µs

cbrt(27), 3passed137µs

WhenI calculate "cbrt(27)"103µs

Thenthe result is "3"32µs

cbrt(-27), -3passed131µs

WhenI calculate "cbrt(-27)"97µs

Thenthe result is "-3"31µs

root(32, 5), 2passed133µs

WhenI calculate "root(32, 5)"100µs

Thenthe result is "2"31µs

pow(4, 0.5), 2passed134µs

WhenI calculate "pow(4, 0.5)"100µs

Thenthe result is "2"30µs

Mathematical symbols are first-classoutline · 11passed

√16, 4passed899µs

WhenI calculate "√16"856µs

Thenthe result is "4"39µs

√(2 + 2), 2passed963µs

WhenI calculate "√(2 + 2)"921µs

Thenthe result is "2"40µs

√2^2, 2passed916µs

WhenI calculate "√2^2"862µs

Thenthe result is "2"51µs

6 × 7 ÷ 2, 21passed668µs

WhenI calculate "6 × 7 ÷ 2"630µs

Thenthe result is "21"36µs

6 ÷ 2 − 1, 2passed664µs

WhenI calculate "6 ÷ 2 − 1"628µs

Thenthe result is "2"34µs

3 · 4, 12passed112µs

WhenI calculate "3 · 4"78µs

Thenthe result is "12"32µs

π − pi, 0passed109µs

WhenI calculate "π − pi"79µs

Thenthe result is "0"27µs

τ ÷ π, 2passed761µs

WhenI calculate "τ ÷ π"720µs

Thenthe result is "2"38µs

pi, 3.14159265358979323846264338327950288419716939937510582097494passed104µs

WhenI calculate "pi"46µs

Thenthe result is "3.14159265358979323846264338327950288419716939937510582097494"55µs

3 ≠ 2, 1passed107µs

WhenI calculate "3 ≠ 2"73µs

Thenthe result is "1"32µs

2 ≤ 2, 1passed102µs

WhenI calculate "2 ≤ 2"68µs

Thenthe result is "1"31µs

Constants are protectedpassed123µs

WhenI calculate "π = 3"72µs

Thenthe calculation fails mentioning "cannot assign to 'π'"48µs

Trigonometry (radians)outline · 6passed

sin(pi / 2), 1passed235µs

WhenI calculate "sin(pi / 2)"200µs

Thenthe result is "1"33µs

cos(pi), -1passed154µs

WhenI calculate "cos(pi)"100µs

Thenthe result is "-1"51µs

tan(pi / 4), 0.9999999999999999passed252µs

WhenI calculate "tan(pi / 4)"212µs

Thenthe result is "0.9999999999999999"37µs

asin(1) * 2, 3.1415926535897932passed142µs

WhenI calculate "asin(1) * 2"103µs

Thenthe result is "3.1415926535897932"37µs

acos(0) * 2, 3.1415926535897932passed132µs

WhenI calculate "acos(0) * 2"94µs

Thenthe result is "3.1415926535897932"36µs

atan(1) * 4, 3.1415926535897932passed151µs

WhenI calculate "atan(1) * 4"113µs

Thenthe result is "3.1415926535897932"36µs

Exponentials and logarithmsoutline · 7passed

exp(0), 1passed119µs

WhenI calculate "exp(0)"85µs

Thenthe result is "1"32µs

ln(e), 1passed161µs

WhenI calculate "ln(e)"126µs

Thenthe result is "1"32µs

exp(ln(7)), 6.999999999999999passed172µs

WhenI calculate "exp(ln(7))"129µs

Thenthe result is "6.999999999999999"40µs

log10(1000), 3passed149µs

WhenI calculate "log10(1000)"115µs

Thenthe result is "3"31µs

log(2, 8), 3passed132µs

WhenI calculate "log(2, 8)"100µs

Thenthe result is "3"29µs

sin(0), 0passed111µs

WhenI calculate "sin(0)"68µs

Thenthe result is "0"39µs

cos(0), 1passed118µs

WhenI calculate "cos(0)"83µs

Thenthe result is "1"32µs

Statistics over lists, arrays, and rangesoutline · 8passed

avg(2, 4, 9), 5passed695µs

WhenI calculate "avg(2, 4, 9)"658µs

Thenthe result is "5"34µs

median(1, 9, 5), 5passed146µs

WhenI calculate "median(1, 9, 5)"112µs

Thenthe result is "5"31µs

median(1, 2, 3, 4), 2.5passed695µs

WhenI calculate "median(1, 2, 3, 4)"658µs

Thenthe result is "2.5"35µs

count(1, 2, 3), 3passed135µs

WhenI calculate "count(1, 2, 3)"102µs

Thenthe result is "3"31µs

product(2, 3, 4), 24passed153µs

WhenI calculate "product(2, 3, 4)"112µs

Thenthe result is "24"37µs

stdev(2, 4, 4, 4, 5, 5, 7, 9), 2.1380899352993950774764278470380281724320113187307passed2ms

WhenI calculate "stdev(2, 4, 4, 4, 5, 5, 7, 9)"1ms

Thenthe result is "2.1380899352993950774764278470380281724320113187307"55µs

avg([2, 4, 9]), 5passed703µs

WhenI calculate "avg([2, 4, 9])"665µs

Thenthe result is "5"35µs

stdev(1, 3) - sqrt(2), 0passed2ms

WhenI calculate "stdev(1, 3) - sqrt(2)"2ms

Thenthe result is "0"30µs

Products mirror summationpassed345µs

WhenI calculate "∏(2, 3, 4)"119µs

Thenthe result is "24"32µs

WhenI calculate "∏_i=1^5(i)"158µs

Thenthe result is "120"31µs

Euler's number from its factorial seriespassed10ms

WhenI calculate "∑_k=0^45(1 / fact(k)) - e"9ms

Thenthe result is within "1e-45" of zero61µs

Nicomachus — the sum of cubes is the square of the sumpassed862µs

WhenI calculate "reduce((a, b) -> a + b, map(x -> x^3, [1,2,3,4,5,6,7,8,9,10]), 0) == (∑_i=1^10(i))^2"822µs

Thenthe result is "1"37µs

Three roads to twenty factorial agree exactlypassed1ms

WhenI calculate "rec(n) = if(n <= 1, 1, n * rec(n - 1))"174µs

AndI calculate "∏_i=1^20(i) == fact(20)"400µs

Thenthe result is "1"32µs

WhenI calculate "rec(20) == fact(20)"594µs

Thenthe result is "1"32µs

Compound growth's product form equals its closed form, exactlypassed966µs

WhenI calculate "grow(r, n) = ∏_i=1^(n)(1 + r)"155µs

AndI calculate "grow(0.05, 30) == 1.05^30"769µs

Thenthe result is "1"38µs

A mortgage payment round-trips the principal to 30 decimal placespassed4.35s

WhenI calculate "pv(0.05/12, 360, pmt(0.05/12, 360, 200000)) - 200000"4.35s

Thenthe result is within "1e-30" of zero95µs

Finance agrees with Excel (independent cross-validation)outline · 14passed

pmt(0.05/12, 360, 200000), -1073.64, 0.01passed1.77s

WhenI calculate "pmt(0.05/12, 360, 200000)"1.77s

Thenthe result is within "0.01" of "-1073.64"169µs

pmt(0, 12, 1200), -100, 0.000001passed831µs

WhenI calculate "pmt(0, 12, 1200)"740µs

Thenthe result is within "0.000001" of "-100"87µs

fv(0.06/12, 120, -100), 16387.93, 0.01passed2ms

WhenI calculate "fv(0.06/12, 120, -100)"2ms

Thenthe result is within "0.01" of "16387.93"157µs

fv(0, 12, -100), 1200, 0.000001passed300µs

WhenI calculate "fv(0, 12, -100)"213µs

Thenthe result is within "0.000001" of "1200"83µs

pv(0.04/12, 60, -500), 27149.53, 0.01passed75ms

WhenI calculate "pv(0.04/12, 60, -500)"75ms

Thenthe result is within "0.01" of "27149.53"167µs

nper(0.05/12, -1073.64, 200000), 360, 0.01passed400µs

WhenI calculate "nper(0.05/12, -1073.64, 200000)"317µs

Thenthe result is within "0.01" of "360"78µs

nper(0, -100, 1200), 12, 0.000001passed243µs

WhenI calculate "nper(0, -100, 1200)"178µs

Thenthe result is within "0.000001" of "12"62µs

rate(360, -1073.64, 200000) * 12, 0.05, 0.0001passed309µs

WhenI calculate "rate(360, -1073.64, 200000) * 12"235µs

Thenthe result is within "0.0001" of "0.05"71µs

npv(0.1, 3000, 4200, 6800), 11307.29, 0.01passed517µs

WhenI calculate "npv(0.1, 3000, 4200, 6800)"424µs

Thenthe result is within "0.01" of "11307.29"90µs

irr(-70000, 12000, 15000, 18000, 21000, 26000), 0.0866, 0.0001passed731µs

WhenI calculate "irr(-70000, 12000, 15000, 18000, 21000, 26000)"659µs

Thenthe result is within "0.0001" of "0.0866"69µs

effectiveRate(0.06, 12), 0.061678, 0.000001passed775µs

WhenI calculate "effectiveRate(0.06, 12)"684µs

Thenthe result is within "0.000001" of "0.061678"87µs

markup(80, 25), 100, 0.000001passed745µs

WhenI calculate "markup(80, 25)"666µs

Thenthe result is within "0.000001" of "100"76µs

percentOf(30, 120), 25, 0.000001passed711µs

WhenI calculate "percentOf(30, 120)"647µs

Thenthe result is within "0.000001" of "25"62µs

percentChange(80, 100), 25, 0.000001passed992µs

WhenI calculate "percentChange(80, 100)"869µs

Thenthe result is within "0.000001" of "25"111µs

A mortgage's lifetime cost, built up across linespassed1.77s

WhenI calculate "r = 0.05 / 12"175µs

AndI calculate "payment = pmt(r, 360, 200000)"1.77s

AndI calculate "payment * 360"164µs

Thenthe result is within "0.01" of "-386511.57"100µs

IRR refuses an unsolvable streampassed216µs

WhenI calculate "irr(1000, 2000)"159µs

Thenthe calculation fails mentioning "both positive and negative"52µs

Finance (spreadsheet sign convention)outline · 6passed

fv(0.1, 2, -100), 210passed770µs

WhenI calculate "fv(0.1, 2, -100)"729µs

Thenthe result is "210"38µs

npv(0, 10, 20, 30), 60passed2ms

WhenI calculate "npv(0, 10, 20, 30)"2ms

Thenthe result is "60"44µs

npv(0.1, 110), 100passed719µs

WhenI calculate "npv(0.1, 110)"682µs

Thenthe result is "100"35µs

nper(0, -10, 100), 10passed241µs

WhenI calculate "nper(0, -10, 100)"201µs

Thenthe result is "10"36µs

irr(-100, 110), 0.1passed205µs

WhenI calculate "irr(-100, 110)"167µs

Thenthe result is "0.1"35µs

effectiveRate(0.12, 12), 0.126825030131969720661201passed719µs

WhenI calculate "effectiveRate(0.12, 12)"676µs

Thenthe result is "0.126825030131969720661201"40µs

Accounting shorthandsoutline · 4passed

margin(100, 80), 20passed703µs

WhenI calculate "margin(100, 80)"666µs

Thenthe result is "20"34µs

markup(80, 25), 100passed689µs

WhenI calculate "markup(80, 25)"654µs

Thenthe result is "100"32µs

percentOf(50, 200), 25passed685µs

WhenI calculate "percentOf(50, 200)"650µs

Thenthe result is "25"32µs

percentChange(100, 150), 50passed741µs

WhenI calculate "percentChange(100, 150)"697µs

Thenthe result is "50"41µs

Date arithmetic on exact day serialsoutline · 11passed

weekday(date(2026, 6, 6)), 6passed217µs

WhenI calculate "weekday(date(2026, 6, 6))"165µs

Thenthe result is "6"49µs

weekday(date(2026, 6, 8)), 1passed198µs

WhenI calculate "weekday(date(2026, 6, 8))"160µs

Thenthe result is "1"36µs

year(date(2026, 6, 6)), 2026passed185µs

WhenI calculate "year(date(2026, 6, 6))"150µs

Thenthe result is "2026"32µs

month(date(2026, 6, 6)), 6passed180µs

WhenI calculate "month(date(2026, 6, 6))"146µs

Thenthe result is "6"31µs

day(date(2026, 6, 6)), 6passed176µs

WhenI calculate "day(date(2026, 6, 6))"143µs

Thenthe result is "6"31µs

days(date(2026, 3, 1), date(2026, 2, 1)), 28passed249µs

WhenI calculate "days(date(2026, 3, 1), date(2026, 2, 1))"216µs

Thenthe result is "28"30µs

edate(date(2026, 1, 31), 1) == date(2026, 2, 28), 1passed263µs

WhenI calculate "edate(date(2026, 1, 31), 1) == date(2026, 2, 28)"230µs

Thenthe result is "1"31µs

eomonth(date(2024, 2, 1), 0) == date(2024, 2, 29), 1passed263µs

WhenI calculate "eomonth(date(2024, 2, 1), 0) == date(2024, 2, 29)"228µs

Thenthe result is "1"32µs

day(eomonth(date(1900, 2, 1), 0)), 28passed270µs

WhenI calculate "day(eomonth(date(1900, 2, 1), 0))"230µs

Thenthe result is "28"37µs

day(eomonth(date(2000, 2, 1), 0)), 29passed258µs

WhenI calculate "day(eomonth(date(2000, 2, 1), 0))"221µs

Thenthe result is "29"34µs

days(date(2001, 1, 1), date(2000, 1, 1)), 366passed264µs

WhenI calculate "days(date(2001, 1, 1), date(2000, 1, 1))"224µs

Thenthe result is "366"36µs

Logic composes with everythingpassed191µs

WhenI calculate "and(1 < 2, or(0, not(0)))"152µs

Thenthe result is "1"36µs

Inspecting the workbook

13/13 scenarios passed · 0µs

Workbook reports its sheetspassed956µs

Givena sheet named "Budget"817µs

WhenI calculate "Workbook.count"102µs

Thenthe result is "2"33µs

Worksheet names are readablepassed185µs

Givena sheet named "Budget"40µs

WhenI calculate "Workbook.worksheets[1].name == "Budget""110µs

Thenthe result is "1"31µs

A worksheet is reachable by namepassed161µs

Givena sheet named "Budget"38µs

WhenI calculate "Workbook.worksheets["Budget"].name == "Budget""87µs

Thenthe result is "1"32µs

A negative index counts from the endpassed198µs

Givena sheet named "Budget"38µs

WhenI calculate "Workbook.worksheets[-1].name == "Budget""126µs

Thenthe result is "1"31µs

Reading a cell's value through the object graphpassed404µs

Givena sheet named "Budget"77µs

Andcell B:1 on "Budget" contains "1200"117µs

WhenI calculate "Workbook.worksheets["Budget"].cell("B", 1).value * 2"168µs

Thenthe result is "2400"35µs

A cell value is reachable by positionpassed247µs

Givencell A:1 contains "=2 + 3"74µs

WhenI calculate "Workbook.worksheets[0].cell("A", 1).value"138µs

Thenthe result is "5"31µs

The flat cell() accessor reads the active sheetpassed184µs

Givencell A:1 contains "42"55µs

WhenI calculate "cell("A", 1).value"94µs

Thenthe result is "42"31µs

The flat cell() accessor reaches another sheet by namepassed269µs

Givena sheet named "Budget"46µs

Andcell A:1 on "Budget" contains "99"60µs

WhenI calculate "cell("Budget", "A", 1).value"122µs

Thenthe result is "99"34µs

sheetNames() lists every sheetpassed158µs

Givena sheet named "Budget"40µs

WhenI calculate "len(sheetNames())"83µs

Thenthe result is "2"31µs

rowCount() reports the grid sizepassed113µs

WhenI calculate "rowCount()"78µs

Thenthe result is "1000"32µs

A formula reading a cell through reflection recomputes livepassed236µs

Givencell A:1 contains "10"61µs

Andcell B:1 contains "=cell("A", 1).value + 1"109µs

Thencell B:1 shows "11"63µs

A user's own function shadows a reflection accessorpassed228µs

WhenI calculate "cell(x) = x * 10"114µs

AndI calculate "cell(5)"81µs

Thenthe result is "50"28µs

An unknown sheet is reported clearlypassed179µs

WhenI calculate "cell("Nope", "A", 1).value"139µs

Thenthe calculation fails mentioning "unknown sheet"37µs

Working in the grid

21/21 scenarios passed · 0µs

Cells reference each otherpassed206µs

Givencell B:1 contains "1200"78µs

Andcell B:2 contains "=B:1 * 2"69µs

Thencell B:2 shows "2400"55µs

Labels never become errorspassed110µs

Givencell A:1 contains "Q1 revenue"61µs

Thencell A:1 shows "Q1 revenue"46µs

A comment cell is a note that holds no valuepassed93µs

Givencell A:1 contains "# tentative — revisit in Q4"52µs

Thencell A:1 shows "# tentative — revisit in Q4"39µs

A note cell is skipped in ranges and errors when referencedpassed459µs

Giventhe sheet contains:109µs

Andcell B:4 contains "=sum(B:1..B:3)"72µs

Thencell B:4 shows "150"127µs

Andcell B:5 contains "=B:2 + 1"88µs

Thencell B:5 shows an error mentioning "not a number"55µs

A trailing comment on a formula is kept and ignored by evaluationpassed201µs

Givencell B:1 contains "200"72µs

Andcell B:2 contains "=B:1 * 1.08 # with sales tax"73µs

Thencell B:2 shows "216"52µs

Ranges aggregate a columnpassed305µs

Giventhe sheet contains:115µs

Andcell B:4 contains "=sum(B:1..B:3)"75µs

Thencell B:4 shows "400"111µs

A named cell reads like prosepassed238µs

Givencell B:7 contains "0.08"57µs

Andcell B:7 is named "Projected Rate"48µs

Andcell A:1 contains "='Projected Rate' * 100"77µs

Thencell A:1 shows "8"51µs

Sheet definitions belong to their cellspassed333µs

Givencell A:1 contains "rate = 0.1"72µs

Andcell A:2 contains "=100 * rate"76µs

Thencell A:1 shows "𝑖 rate"35µs

Andcell A:2 shows "10"52µs

WhenI calculate "rate = 0.5"61µs

Thenthe calculation fails mentioning "defined in cell"32µs

A function defined in a cell is callablepassed270µs

Givencell A:1 contains "tax(x) = x * 1.0825"95µs

Andcell A:2 contains "=tax(200)"78µs

Thencell A:1 shows "λ tax(x)"34µs

Andcell A:2 shows "216.5"59µs

A slider is a value with a rangepassed335µs

Givencell A:1 contains "r = slider(0.11, 0, 0.2)"112µs

Andcell A:2 contains "=r * 100"75µs

Thencell A:1 is a slider set to "0.11"88µs

Andcell A:2 shows "11"56µs

The log sees the sheetpassed165µs

Givencell B:1 contains "1200"64µs

WhenI calculate "B:1 * 2"67µs

Thenthe result is "2400"29µs

Explicit markers override auto-detectionpassed220µs

Givencell A:1 contains ""123""37µs

Andcell A:2 contains "=12 * rte"61µs

Thencell A:1 shows "123"36µs

Andcell A:2 shows an error mentioning "unknown variable"78µs

A formula mistake is an error, not a guesspassed203µs

Givencell B:1 contains "100"83µs

Andcell B:2 contains "B:1 / 0"64µs

Thencell B:2 shows an error mentioning "division by zero"52µs

Empty cells read as zeropassed125µs

Givencell A:1 contains "=B:9 + 5"70µs

Thencell A:1 shows "5"53µs

Circular references are caught, not infinitepassed216µs

Givencell A:1 contains "=A:2 + 1"65µs

Andcell A:2 contains "=A:1 + 1"78µs

Thencell A:1 shows an error mentioning "circular reference"67µs

Referencing a text cell is an errorpassed222µs

Givencell A:1 contains "Q1 revenue"64µs

Andcell A:2 contains "=A:1 * 2"82µs

Thencell A:2 shows an error mentioning "not a number"71µs

The controls family holds valuespassed703µs

Givencell A:1 contains "flag = checkbox(true)"87µs

Andcell A:2 contains "n = stepper(5, 1, 20)"139µs

Andcell A:3 contains "region = dropdown("EU", ["EU", "US"])"118µs

Andcell B:1 contains "=if(flag, n * 2, 0)"108µs

Andcell B:2 contains "=if(region == "EU", 1, 2)"100µs

Thencell B:1 shows "10"76µs

Andcell B:2 shows "1"66µs

A column of checkboxes counts its checked onespassed310µs

Giventhe sheet contains:141µs

Andcell C:4 contains "=sum(C:1..C:3)"73µs

Thencell C:4 shows "2"92µs

Formulas reach across worksheets by namepassed419µs

Givena sheet named "Budget"47µs

Andcell B:1 on "Budget" contains "1200"77µs

Andcell A:1 contains "=Budget!B:1 * 2"75µs

Thencell A:1 shows "2400"52µs

WhenI calculate "sum(Budget!B:1..B:1) + 1"129µs

Thenthe result is "1201"32µs

A renamed-away sheet breaks loudly, not silentlypassed138µs

Givencell A:1 contains "=Nowhere!B:1"58µs

Thencell A:1 shows an error mentioning "unknown sheet"77µs

A workbook round-trips through its file formatpassed2ms

Givena sheet named "Loan"44µs

Andcell B:1 on "Loan" contains "350000"91µs

Andcell A:1 contains "=Loan!B:1 / 100"110µs

Andcell B:7 contains "0.08"62µs

Andcell B:7 is named "Projected Rate"47µs

Andcell A:2 contains "='Projected Rate' * 100"91µs

WhenI calculate "growth = 1.5"66µs

Andthe workbook is saved and reopened373µs

Thencell A:1 shows "3500"586µs

Andcell A:2 shows "8"50µs

WhenI calculate "growth * 2"72µs

Thenthe result is "3"32µs

Strings, arrays, maps, and the functions that work them

27/27 scenarios passed · 0µs

Text and structure functionsoutline · 18passed

len([1, 2, 3]), 3passed146µs

WhenI calculate "len([1, 2, 3])"113µs

Thenthe result is "3"31µs

len("hello"), 5passed110µs

WhenI calculate "len("hello")"75µs

Thenthe result is "5"32µs

len({a: 1, b: 2}), 2passed158µs

WhenI calculate "len({a: 1, b: 2})"123µs

Thenthe result is "2"32µs

first([5, 6, 7]), 5passed142µs

WhenI calculate "first([5, 6, 7])"109µs

Thenthe result is "5"31µs

last([5, 6, 7]), 7passed138µs

WhenI calculate "last([5, 6, 7])"105µs

Thenthe result is "7"31µs

keys({name: "Ada", age: 36}), ["name", "age"]passed150µs

WhenI calculate "keys({name: "Ada", age: 36})"107µs

Thenthe result is "["name", "age"]"39µs

values({a: 1, b: 2}), [1, 2]passed183µs

WhenI calculate "values({a: 1, b: 2})"143µs

Thenthe result is "[1, 2]"37µs

sum(values({a: 1, b: 2})), 3passed194µs

WhenI calculate "sum(values({a: 1, b: 2}))"145µs

Thenthe result is "3"46µs

concat("Q", 1, "-", 2026), "Q1-2026"passed151µs

WhenI calculate "concat("Q", 1, "-", 2026)"120µs

Thenthe result is ""Q1-2026""28µs

concat([1, 2], [3]), [1, 2, 3]passed161µs

WhenI calculate "concat([1, 2], [3])"119µs

Thenthe result is "[1, 2, 3]"40µs

"Q" + 1, "Q1"passed89µs

WhenI calculate ""Q" + 1"60µs

Thenthe result is ""Q1""26µs

"a" + "b" + "c", "abc"passed93µs

WhenI calculate ""a" + "b" + "c""63µs

Thenthe result is ""abc""27µs

"abc"[0], "a"passed93µs

WhenI calculate ""abc"[0]"57µs

Thenthe result is ""a""33µs

[[1, 2], [3, 4]][1][0], 3passed203µs

WhenI calculate "[[1, 2], [3, 4]][1][0]"160µs

Thenthe result is "3"41µs

{a: [1, 2]}.a[1], 2passed153µs

WhenI calculate "{a: [1, 2]}.a[1]"120µs

Thenthe result is "2"31µs

[1, 2] == [1, 2], 1passed144µs

WhenI calculate "[1, 2] == [1, 2]"112µs

Thenthe result is "1"30µs

{a: 1, b: 2} == {b: 2, a: 1}, 1passed178µs

WhenI calculate "{a: 1, b: 2} == {b: 2, a: 1}"146µs

Thenthe result is "1"30µs

"x" != 5, 1passed84µs

WhenI calculate ""x" != 5"54µs

Thenthe result is "1"28µs

Structure mistakes explain themselvesoutline · 6passed

[1, 2][5], out of rangepassed141µs

WhenI calculate "[1, 2][5]"100µs

Thenthe calculation fails mentioning "out of range"38µs

{a: 1}.b, no key 'b'passed107µs

WhenI calculate "{a: 1}.b"75µs

Thenthe calculation fails mentioning "no key 'b'"30µs

first([]), empty arraypassed102µs

WhenI calculate "first([])"63µs

Thenthe calculation fails mentioning "empty array"37µs

sum(["a"]), works on numberspassed100µs

WhenI calculate "sum(["a"])"66µs

Thenthe calculation fails mentioning "works on numbers"31µs

"a" * 2, expected a numberpassed90µs

WhenI calculate ""a" * 2"57µs

Thenthe calculation fails mentioning "expected a number"31µs

{a: 1, a: 2}, duplicate keypassed153µs

WhenI calculate "{a: 1, a: 2}"88µs

Thenthe calculation fails mentioning "duplicate key"62µs

Filter and reduce shape data like a user thinkspassed369µs

WhenI calculate "filter(x -> x > 1, [1, 2, 3])"171µs

Thenthe result is "[2, 3]"38µs

WhenI calculate "reduce((a, b) -> a + b, [], 42)"126µs

Thenthe result is "42"31µs

Lambdas live in variables and close over parameterspassed544µs

WhenI calculate "f = x -> x * 2"103µs

AndI calculate "f(21)"76µs

Thenthe result is "42"31µs

WhenI calculate "scale(arr, n) = map(x -> x * n, arr)"127µs

AndI calculate "scale([1, 2, 3], 10)"161µs

Thenthe result is "[10, 20, 30]"40µs

A named function reference follows redefinitionpassed346µs

WhenI calculate "h(x) = x + 1"84µs

AndI calculate "alias = h"51µs

AndI calculate "h(x) = x + 100"96µs

AndI calculate "alias(1)"79µs

Thenthe result is "101"31µs