feat(codemode): native coercion parity in interpreter (#37608)

This commit is contained in:
Aiden Cline 2026-07-18 12:01:10 -05:00 committed by GitHub
commit 5f437a09b0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 323 additions and 48 deletions

View file

@ -12,7 +12,7 @@ import {
PromiseNamespace,
UriFunction,
} from "./model.js"
import { rejectCircularInsertion, typeofValue } from "./references.js"
import { containsOpaqueReference, rejectCircularInsertion, typeofValue } from "./references.js"
import { isBlockedMember, type SafeObject } from "../tool-runtime.js"
import {
CodeModeDate,
@ -137,21 +137,31 @@ export const invokeGlobalMethod = (ref: GlobalMethodReference, args: Array<unkno
return invokeJsonMethod(ref.name, args, node)
}
const requireDataArgument = (name: string, index: number, arg: unknown, node: AstNode): unknown => {
if (containsOpaqueReference(arg)) {
throw new InterpreterRuntimeError(
`String.${name} expects argument ${index + 1} to be a data value.`,
node,
"InvalidDataValue",
)
}
return arg
}
const invokeStringMethod = (value: string, name: string, args: Array<unknown>, node: AstNode): unknown => {
const str = (index: number): string => {
const arg = args[index]
if (typeof arg !== "string")
throw new InterpreterRuntimeError(`String.${name} expects argument ${index + 1} to be a string.`, node)
return arg
}
const num = (index: number): number => {
const arg = args[index]
if (typeof arg !== "number")
throw new InterpreterRuntimeError(`String.${name} expects argument ${index + 1} to be a number.`, node)
return arg
}
// Coerce arguments like native JS; opaque runtime references still reject.
const str = (index: number): string => coerceToString(requireDataArgument(name, index, args[index], node))
const num = (index: number): number => coerceToNumber(requireDataArgument(name, index, args[index], node))
const optNum = (index: number): number | undefined => (args[index] === undefined ? undefined : num(index))
const optStr = (index: number): string | undefined => (args[index] === undefined ? undefined : str(index))
const rejectRegex = (): void => {
if (args[0] instanceof CodeModeRegExp) {
throw new InterpreterRuntimeError(
`String.${name} cannot take a regular expression; use regex.test(string) or String.search instead.`,
node,
).as("TypeError")
}
}
let result: unknown
switch (name) {
@ -187,8 +197,11 @@ const invokeStringMethod = (value: string, name: string, args: Array<unknown>, n
break
}
case "split": {
if (args.length === 0) {
result = [value]
// Native: an undefined separator returns the whole string, not a split on "undefined",
// unless the limit truncates to zero.
if (args[0] === undefined) {
const requestedLimit = optNum(1)
result = requestedLimit !== undefined && requestedLimit >>> 0 === 0 ? [] : [value]
break
}
if (args[0] instanceof CodeModeRegExp) {
@ -203,12 +216,15 @@ const invokeStringMethod = (value: string, name: string, args: Array<unknown>, n
result = value.slice(optNum(0), optNum(1))
break
case "includes":
rejectRegex()
result = value.includes(str(0), optNum(1))
break
case "startsWith":
rejectRegex()
result = value.startsWith(str(0), optNum(1))
break
case "endsWith":
rejectRegex()
result = value.endsWith(str(0), optNum(1))
break
case "indexOf":
@ -263,7 +279,7 @@ const invokeStringMethod = (value: string, name: string, args: Array<unknown>, n
case "repeat": {
const count = num(0)
if (!Number.isFinite(count) || count < 0)
throw new InterpreterRuntimeError("String.repeat expects a finite non-negative count.", node)
throw new InterpreterRuntimeError("String.repeat expects a finite non-negative count.", node).as("RangeError")
result = value.repeat(count)
break
}
@ -301,6 +317,8 @@ const invokeStringMethod = (value: string, name: string, args: Array<unknown>, n
return boundedData(result, `String.${name} result`)
}
export const arrayStatics = new Set(["isArray", "of", "from"])
const invokeArrayStatic = (name: string, args: Array<unknown>, node: AstNode): unknown => {
switch (name) {
case "isArray":
@ -400,11 +418,9 @@ const invokeStringReplacer = <R>(
if (name === "replace") value.replace(pattern.regex, collect)
else value.replaceAll(pattern.regex, collect)
} else {
if (typeof pattern !== "string") {
throw new InterpreterRuntimeError(`String.${name} expects argument 1 to be a string.`, node)
}
if (name === "replace") value.replace(pattern, collect)
else value.replaceAll(pattern, collect)
const search = coerceToString(requireDataArgument(name, 0, pattern, node))
if (name === "replace") value.replace(search, collect)
else value.replaceAll(search, collect)
}
return Effect.gen(function* () {

View file

@ -105,7 +105,7 @@ export class GlobalMethodReference {
}
export class CoercionFunction {
constructor(readonly name: "Number" | "String" | "Boolean" | "parseInt" | "parseFloat") {}
constructor(readonly name: "Number" | "String" | "Boolean" | "parseInt" | "parseFloat" | "isFinite" | "isNaN") {}
}
export class UriFunction {

View file

@ -10,6 +10,7 @@ import {
ErrorConstructorReference,
GlobalMethodReference,
GlobalNamespace,
type GlobalNamespaceName,
getArray,
getBoolean,
getNode,
@ -34,7 +35,7 @@ import {
UriFunction,
} from "./model.js"
import { caughtErrorValue, constructErrorValue } from "./errors.js"
import { type CallbackRunner, invokeArrayFrom, invokeGlobalMethod, invokeIntrinsic } from "./methods.js"
import { arrayStatics, type CallbackRunner, invokeArrayFrom, invokeGlobalMethod, invokeIntrinsic } from "./methods.js"
import {
constructPromise,
invokePromiseInstanceMethod,
@ -46,10 +47,11 @@ import { containsOpaqueReference, isRuntimeReference, rejectCircularInsertion, t
import { ScopeStack } from "./scope.js"
import { arrayMethods, mapMethods, setMethods, spreadItems } from "../stdlib/collections.js"
import { consoleMethods, formatConsoleMessage } from "../stdlib/console.js"
import { dateMethods } from "../stdlib/date.js"
import { mathConstants } from "../stdlib/math.js"
import { dateMethods, dateStatics } from "../stdlib/date.js"
import { jsonStatics } from "../stdlib/json.js"
import { mathConstants, mathMethods } from "../stdlib/math.js"
import { numberConstants, numberMethods, numberStatics } from "../stdlib/number.js"
import { objectMethodsPreservingIdentity } from "../stdlib/object.js"
import { objectMethodsPreservingIdentity, objectStatics } from "../stdlib/object.js"
import { promiseStatics } from "../stdlib/promise.js"
import { escapeRegexHint, regexpMethods, regexpProperties, regexFailureReason } from "../stdlib/regexp.js"
import { stringMethods, stringStatics } from "../stdlib/string.js"
@ -57,6 +59,7 @@ import {
urlMethods,
urlProperties,
urlSearchParamsMethods,
urlStatics,
urlWritableProperties,
invokeUriFunction,
uriArgument,
@ -83,6 +86,32 @@ import {
CodeModeURLSearchParams,
} from "../values.js"
const globalStaticMembers: Partial<Record<GlobalNamespaceName, Set<string>>> = {
Object: objectStatics,
Math: mathMethods,
JSON: jsonStatics,
Array: arrayStatics,
console: consoleMethods,
Date: dateStatics,
URL: urlStatics,
}
const calleeDescription = (callee: AstNode): string => {
if (callee.type === "Identifier") return getString(callee, "name")
if (callee.type === "MemberExpression") {
const object = getNode(callee, "object")
const property = getNode(callee, "property")
const key =
callee.computed !== true && property.type === "Identifier"
? getString(property, "name")
: property.type === "Literal" && typeof property.value === "string"
? property.value
: undefined
if (object.type === "Identifier" && key !== undefined) return `${getString(object, "name")}.${key}`
}
return "The called value"
}
const instanceofValue = (lhs: unknown, rhs: unknown, node: AstNode): boolean => {
if (rhs instanceof ErrorConstructorReference) {
const brand = errorBrandName(lhs)
@ -199,6 +228,8 @@ export class Interpreter<R> {
globalScope.set("console", { mutable: false, value: new GlobalNamespace("console") })
globalScope.set("parseInt", { mutable: false, value: new CoercionFunction("parseInt") })
globalScope.set("parseFloat", { mutable: false, value: new CoercionFunction("parseFloat") })
globalScope.set("isFinite", { mutable: false, value: new CoercionFunction("isFinite") })
globalScope.set("isNaN", { mutable: false, value: new CoercionFunction("isNaN") })
globalScope.set("Date", { mutable: false, value: new GlobalNamespace("Date") })
globalScope.set("RegExp", { mutable: false, value: new GlobalNamespace("RegExp") })
globalScope.set("Map", { mutable: false, value: new GlobalNamespace("Map") })
@ -1454,10 +1485,23 @@ export class Interpreter<R> {
throw new InterpreterRuntimeError(`Unsupported update operator '${operator}'.`, node)
}
// CodeMode numeric coercion, not host Number(): null-prototype data objects would make
// the host throw during ToPrimitive, and opaque runtime references must reject clearly.
const operand = (current: unknown): number => {
if (containsOpaqueReference(current)) {
throw new InterpreterRuntimeError(
`'${operator}' requires a data value in CodeMode.`,
argument,
"InvalidDataValue",
)
}
return coerceToNumber(current)
}
if (argument.type === "Identifier") {
return Effect.sync(() => {
const name = getString(argument, "name")
const current = Number(this.scopes.get(name, argument))
const current = operand(this.scopes.get(name, argument))
const next = current + increment
this.scopes.set(name, next, argument)
return prefix ? next : current
@ -1466,7 +1510,7 @@ export class Interpreter<R> {
if (argument.type === "MemberExpression") {
return this.modifyMember(argument, (current) => {
const value = Number(current)
const value = operand(current)
const next = value + increment
return Effect.succeed({ write: true, next, result: prefix ? next : value })
})
@ -1563,6 +1607,9 @@ export class Interpreter<R> {
callable.settle(args[0])
return undefined
}
if (callable === undefined || callable === null) {
throw new InterpreterRuntimeError(`${calleeDescription(callee)} is not a function.`, callee).as("TypeError")
}
throw new InterpreterRuntimeError("Only tools are callable in CodeMode.", callee)
})
}
@ -1833,16 +1880,18 @@ export class Interpreter<R> {
}
if (objectValue instanceof GlobalNamespace) {
if (typeof key !== "string" || isBlockedMember(key)) {
throw new InterpreterRuntimeError(
`${objectValue.name}.${String(key)} is not available in CodeMode.`,
propertyNode,
)
if (typeof key === "string" && isBlockedMember(key)) {
throw new InterpreterRuntimeError(`${objectValue.name}.${key} is not available in CodeMode.`, propertyNode)
}
if (typeof key !== "string") return new ComputedValue(undefined)
if (objectValue.name === "Math" && mathConstants.has(key)) {
return new ComputedValue((Math as unknown as Record<string, number>)[key])
}
return new GlobalMethodReference(objectValue.name, key)
if (globalStaticMembers[objectValue.name]?.has(key)) {
return new GlobalMethodReference(objectValue.name, key)
}
// Unknown static members read as undefined so feature detection works like native JS.
return new ComputedValue(undefined)
}
if (typeof objectValue === "string") {
@ -1858,12 +1907,21 @@ export class Interpreter<R> {
return new ComputedValue(undefined)
}
if (objectValue instanceof CoercionFunction && typeof key === "string" && !isBlockedMember(key)) {
if (objectValue instanceof CoercionFunction) {
if (typeof key === "string" && isBlockedMember(key)) {
throw new InterpreterRuntimeError(`${objectValue.name}.${key} is not available in CodeMode.`, propertyNode)
}
if (typeof key !== "string") return new ComputedValue(undefined)
if (objectValue.name === "Number" && numberConstants.has(key)) {
return new ComputedValue((Number as unknown as Record<string, number>)[key])
}
if (objectValue.name === "Number" && numberStatics.has(key)) return new GlobalMethodReference("Number", key)
if (objectValue.name === "String" && stringStatics.has(key)) return new GlobalMethodReference("String", key)
if (objectValue.name === "Number" && numberStatics.has(key)) {
return new GlobalMethodReference("Number", key)
}
if (objectValue.name === "String" && stringStatics.has(key)) {
return new GlobalMethodReference("String", key)
}
return new ComputedValue(undefined)
}
if (objectValue instanceof CodeModeDate) {

View file

@ -23,6 +23,8 @@ export const dateMethods = new Set([
"getTimezoneOffset",
])
export const dateStatics = new Set(["now", "parse", "UTC"])
export const invokeDateStatic = (name: string, args: Array<unknown>, node: AstNode): number => {
switch (name) {
case "now":

View file

@ -2,6 +2,8 @@ import { type AstNode, InterpreterRuntimeError, supportedSyntaxMessage } from ".
import { typeofValue } from "../interpreter/references.js"
import { copyIn, copyOut } from "../tool-runtime.js"
export const jsonStatics = new Set(["parse", "stringify"])
export const invokeJsonMethod = (name: string, args: Array<unknown>, node: AstNode): unknown => {
switch (name) {
case "stringify": {

View file

@ -6,6 +6,8 @@ import { boundedData, coerceToString } from "./value.js"
export const objectMethodsPreservingIdentity = new Set(["assign", "values", "entries", "fromEntries"])
export const objectStatics = new Set(["keys", "values", "entries", "hasOwn", "is", "assign", "fromEntries"])
export const invokeObjectMethod = (name: string, args: Array<unknown>, node: AstNode): unknown => {
const requireObject = (): Record<string, unknown> => {
const input = args[0]

View file

@ -19,6 +19,8 @@ export const escapeRegexHint =
'To match special characters like ( ) [ ] { } + * ? . literally, escape them with a backslash (e.g. "\\\\(") or test for them with String.includes instead.'
export const toHostRegex = (arg: unknown, method: string, node: AstNode, extraFlags = ""): RegExp => {
// Native parity: an undefined pattern behaves as an empty pattern.
if (arg === undefined) return new RegExp("", extraFlags)
if (arg instanceof CodeModeRegExp) return arg.regex
if (typeof arg === "string") {
try {

View file

@ -61,10 +61,19 @@ export const coerceToString = (value: unknown): string => {
export const coerceToNumber = (value: unknown): number => {
if (value instanceof CodeModeDate) return value.time
if (isCodeModeValue(value)) return Number.NaN
return value !== null && typeof value === "object" && !Array.isArray(value) ? Number.NaN : Number(value)
// Arrays coerce through our own string coercion: host Number(array) joins with host
// ToPrimitive, which throws on the null-prototype objects the interpreter produces.
if (Array.isArray(value)) return Number(coerceToString(value))
return value !== null && typeof value === "object" ? Number.NaN : Number(value)
}
export const invokeCoercion = (ref: CoercionFunction, args: Array<unknown>, node: AstNode): unknown => {
// Native: Number() is 0 and String() is "", unlike their undefined-argument forms; the
// other coercers match native through the undefined-argument path below.
if (args.length === 0) {
if (ref.name === "Number") return 0
if (ref.name === "String") return ""
}
const raw = args[0]
// Error values are plain SafeObjects; the boundedData path below would strip their brand.
if (ref.name === "String" && errorBrandName(raw) !== undefined) return coerceToString(raw)
@ -72,12 +81,16 @@ export const invokeCoercion = (ref: CoercionFunction, args: Array<unknown>, node
if (ref.name === "Boolean") return true
if (ref.name === "Number") return coerceToNumber(raw)
if (ref.name === "String") return coerceToString(raw)
if (ref.name === "isFinite") return Number.isFinite(coerceToNumber(raw))
if (ref.name === "isNaN") return Number.isNaN(coerceToNumber(raw))
if (ref.name === "parseInt") return parseInt(coerceToString(raw))
return parseFloat(coerceToString(raw))
}
const value = boundedData(raw, `${ref.name} input`)
if (ref.name === "Number") return coerceToNumber(value)
if (ref.name === "Boolean") return Boolean(value)
if (ref.name === "isFinite") return Number.isFinite(coerceToNumber(value))
if (ref.name === "isNaN") return Number.isNaN(coerceToNumber(value))
if (ref.name === "parseInt") {
const radix = args[1]
if (radix !== undefined && typeof radix !== "number") {