Files
Nim/compiler/semfold.nim
bptato e97b0bb541 Do not directly cast int128 to uint64 in semfold (#25396)
int128 is an array of uint32s, so while this works on little-endian
CPUs, it's completely broken on big-endian. e.g. following snippet would
fail:

	const x = 0xFFFFFFFF'u32
	const y = (x shr 1)
	echo y # amd64: 2147483647, s390x: 0

That in turn broke float printing, resulting in miscompilation of any
code that used floats.

To fix this, we now call the aptly named castToUInt64 procedure which
performs the same cast portably.

(Thanks to barracuda156 for helping debug this.)
2025-12-30 23:09:01 +01:00

801 lines
31 KiB
Nim

#
#
# The Nim Compiler
# (c) Copyright 2015 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
# this module folds constants; used by semantic checking phase
# and evaluation phase
import
options, ast, trees, nimsets,
platform, msgs, idents, renderer, types,
commands, magicsys, modulegraphs, lineinfos, wordrecg
import std/[strutils, math, strtabs]
#from system/memory import nimCStrLen
when defined(nimPreviewSlimSystem):
import std/[assertions, formatfloat]
proc errorType*(g: ModuleGraph): PType =
## creates a type representing an error state
result = newType(tyError, g.idgen, g.owners[^1])
result.flagsImpl.incl tfCheckedForDestructor
proc getIntLitTypeG(g: ModuleGraph; literal: PNode; idgen: IdGenerator): PType =
# we cache some common integer literal types for performance:
let ti = getSysType(g, literal.info, tyInt)
result = copyType(ti, idgen, ti.owner)
result.n = literal
proc newIntNodeT*(intVal: Int128, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
result = newIntTypeNode(intVal, n.typ)
# See bug #6989. 'pred' et al only produce an int literal type if the
# original type was 'int', not a distinct int etc.
if n.typ.kind == tyInt:
# access cache for the int lit type
result.typ = getIntLitTypeG(g, result, idgen)
result.info = n.info
proc newFloatNodeT*(floatVal: BiggestFloat, n: PNode; g: ModuleGraph): PNode =
if n.typ.skipTypes(abstractInst).kind == tyFloat32:
result = newFloatNode(nkFloat32Lit, floatVal)
else:
result = newFloatNode(nkFloatLit, floatVal)
result.typ = n.typ
result.info = n.info
proc newStrNodeT*(strVal: string, n: PNode; g: ModuleGraph): PNode =
result = newStrNode(nkStrLit, strVal)
result.typ = n.typ
result.info = n.info
proc getConstExpr*(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode
# evaluates the constant expression or returns nil if it is no constant
# expression
proc evalOp*(m: TMagic, n, a, b, c: PNode; idgen: IdGenerator; g: ModuleGraph): PNode
proc checkInRange(conf: ConfigRef; n: PNode, res: Int128): bool =
res in firstOrd(conf, n.typ)..lastOrd(conf, n.typ)
proc foldAdd(a, b: Int128, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
let res = a + b
if checkInRange(g.config, n, res):
result = newIntNodeT(res, n, idgen, g)
else:
result = nil
proc foldSub(a, b: Int128, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
let res = a - b
if checkInRange(g.config, n, res):
result = newIntNodeT(res, n, idgen, g)
else:
result = nil
proc foldUnarySub(a: Int128, n: PNode; idgen: IdGenerator, g: ModuleGraph): PNode =
if a != firstOrd(g.config, n.typ):
result = newIntNodeT(-a, n, idgen, g)
else:
result = nil
proc foldAbs(a: Int128, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
if a != firstOrd(g.config, n.typ):
result = newIntNodeT(abs(a), n, idgen, g)
else:
result = nil
proc foldMul(a, b: Int128, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
let res = a * b
if checkInRange(g.config, n, res):
return newIntNodeT(res, n, idgen, g)
else:
result = nil
proc ordinalValToString*(a: PNode; g: ModuleGraph): string =
# because $ has the param ordinal[T], `a` is not necessarily an enum, but an
# ordinal
var x = getInt(a)
var t = skipTypes(a.typ, abstractRange)
case t.kind
of tyChar:
result = $chr(toInt64(x) and 0xff)
of tyEnum:
result = ""
var n = t.n
for i in 0..<n.len:
if n[i].kind != nkSym: internalError(g.config, a.info, "ordinalValToString")
var field = n[i].sym
if field.position == x:
if field.ast == nil:
return field.name.s
else:
return field.ast.strVal
localError(g.config, a.info,
"Cannot convert int literal to $1. The value is invalid." %
[typeToString(t)])
else:
result = $x
proc isFloatRange(t: PType): bool {.inline.} =
result = t.kind == tyRange and t.elementType.kind in {tyFloat..tyFloat128}
proc isIntRange(t: PType): bool {.inline.} =
result = t.kind == tyRange and t.elementType.kind in {
tyInt..tyInt64, tyUInt8..tyUInt32}
proc pickIntRange(a, b: PType): PType =
if isIntRange(a): result = a
elif isIntRange(b): result = b
else: result = a
proc isIntRangeOrLit(t: PType): bool =
result = isIntRange(t) or isIntLit(t)
proc evalOp(m: TMagic, n, a, b, c: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
# b and c may be nil
result = nil
case m
of mOrd: result = newIntNodeT(getOrdValue(a), n, idgen, g)
of mChr: result = newIntNodeT(getInt(a), n, idgen, g)
of mUnaryMinusI, mUnaryMinusI64: result = foldUnarySub(getInt(a), n, idgen, g)
of mUnaryMinusF64: result = newFloatNodeT(-getFloat(a), n, g)
of mNot: result = newIntNodeT(One - getInt(a), n, idgen, g)
of mCard: result = newIntNodeT(toInt128(nimsets.cardSet(g.config, a)), n, idgen, g)
of mBitnotI:
if n.typ.isUnsigned:
result = newIntNodeT(bitnot(getInt(a)).maskBytes(int(getSize(g.config, n.typ))), n, idgen, g)
else:
result = newIntNodeT(bitnot(getInt(a)), n, idgen, g)
of mLengthArray: result = newIntNodeT(lengthOrd(g.config, a.typ), n, idgen, g)
of mLengthSeq, mLengthOpenArray, mLengthStr:
if a.kind == nkNilLit:
result = newIntNodeT(Zero, n, idgen, g)
elif a.kind in {nkStrLit..nkTripleStrLit}:
if a.typ.kind == tyString:
result = newIntNodeT(toInt128(a.strVal.len), n, idgen, g)
elif a.typ.kind == tyCstring:
result = newIntNodeT(toInt128(nimCStrLen(a.strVal.cstring)), n, idgen, g)
else:
result = newIntNodeT(toInt128(a.len), n, idgen, g)
of mUnaryPlusI, mUnaryPlusF64: result = a # throw `+` away
# XXX: Hides overflow/underflow
of mAbsI: result = foldAbs(getInt(a), n, idgen, g)
of mSucc: result = foldAdd(getOrdValue(a), getInt(b), n, idgen, g)
of mPred: result = foldSub(getOrdValue(a), getInt(b), n, idgen, g)
of mAddI: result = foldAdd(getInt(a), getInt(b), n, idgen, g)
of mSubI: result = foldSub(getInt(a), getInt(b), n, idgen, g)
of mMulI: result = foldMul(getInt(a), getInt(b), n, idgen, g)
of mMinI:
let argA = getInt(a)
let argB = getInt(b)
result = newIntNodeT(if argA < argB: argA else: argB, n, idgen, g)
of mMaxI:
let argA = getInt(a)
let argB = getInt(b)
result = newIntNodeT(if argA > argB: argA else: argB, n, idgen, g)
of mShlI:
let valueB = toInt64(getInt(b)) and (n.typ.size * 8 - 1)
case skipTypes(n.typ, abstractRange).kind
of tyInt8: result = newIntNodeT(toInt128(toInt8(getInt(a)) shl valueB), n, idgen, g)
of tyInt16: result = newIntNodeT(toInt128(toInt16(getInt(a)) shl valueB), n, idgen, g)
of tyInt32: result = newIntNodeT(toInt128(toInt32(getInt(a)) shl valueB), n, idgen, g)
of tyInt64: result = newIntNodeT(toInt128(toInt64(getInt(a)) shl valueB), n, idgen, g)
of tyInt:
if g.config.target.intSize == 4:
result = newIntNodeT(toInt128(toInt32(getInt(a)) shl valueB), n, idgen, g)
else:
result = newIntNodeT(toInt128(toInt64(getInt(a)) shl valueB), n, idgen, g)
of tyUInt8: result = newIntNodeT(toInt128(toUInt8(getInt(a)) shl valueB), n, idgen, g)
of tyUInt16: result = newIntNodeT(toInt128(toUInt16(getInt(a)) shl valueB), n, idgen, g)
of tyUInt32: result = newIntNodeT(toInt128(toUInt32(getInt(a)) shl valueB), n, idgen, g)
of tyUInt64: result = newIntNodeT(toInt128(toUInt64(getInt(a)) shl valueB), n, idgen, g)
of tyUInt:
if g.config.target.intSize == 4:
result = newIntNodeT(toInt128(toUInt32(getInt(a)) shl valueB), n, idgen, g)
else:
result = newIntNodeT(toInt128(toUInt64(getInt(a)) shl valueB), n, idgen, g)
else: internalError(g.config, n.info, "constant folding for shl")
of mShrI:
var a = castToUInt64(getInt(a))
let b = castToUInt64(getInt(b)) and cast[uint64](n.typ.size * 8 - 1)
# To support the ``-d:nimOldShiftRight`` flag, we need to mask the
# signed integers to cut off the extended sign bit in the internal
# representation.
if 0'u64 < b: # do not cut off the sign extension, when there is
# no bit shifting happening.
case skipTypes(n.typ, abstractRange).kind
of tyInt8: a = a and 0xff'u64
of tyInt16: a = a and 0xffff'u64
of tyInt32: a = a and 0xffffffff'u64
of tyInt:
if g.config.target.intSize == 4:
a = a and 0xffffffff'u64
else:
# unsigned and 64 bit integers don't need masking
discard
let c = cast[BiggestInt](a shr b)
result = newIntNodeT(toInt128(c), n, idgen, g)
of mAshrI:
let valueB = toInt64(getInt(b)) and (n.typ.size * 8 - 1)
case skipTypes(n.typ, abstractRange).kind
of tyInt8: result = newIntNodeT(toInt128(ashr(toInt8(getInt(a)), valueB)), n, idgen, g)
of tyInt16: result = newIntNodeT(toInt128(ashr(toInt16(getInt(a)), valueB)), n, idgen, g)
of tyInt32: result = newIntNodeT(toInt128(ashr(toInt32(getInt(a)), valueB)), n, idgen, g)
of tyInt64, tyInt:
result = newIntNodeT(toInt128(ashr(toInt64(getInt(a)), valueB)), n, idgen, g)
else: internalError(g.config, n.info, "constant folding for ashr")
of mDivI:
let argA = getInt(a)
let argB = getInt(b)
if argB != Zero and (argA != firstOrd(g.config, n.typ) or argB != NegOne):
result = newIntNodeT(argA div argB, n, idgen, g)
of mModI:
let argA = getInt(a)
let argB = getInt(b)
if argB != Zero and (argA != firstOrd(g.config, n.typ) or argB != NegOne):
result = newIntNodeT(argA mod argB, n, idgen, g)
of mAddF64: result = newFloatNodeT(getFloat(a) + getFloat(b), n, g)
of mSubF64: result = newFloatNodeT(getFloat(a) - getFloat(b), n, g)
of mMulF64: result = newFloatNodeT(getFloat(a) * getFloat(b), n, g)
of mDivF64:
result = newFloatNodeT(getFloat(a) / getFloat(b), n, g)
of mIsNil:
let val = a.kind == nkNilLit or
# nil closures have the value (nil, nil)
(a.typ != nil and skipTypes(a.typ, abstractRange).kind == tyProc and
a.kind == nkTupleConstr and a.len == 2 and
a[0].kind == nkNilLit and a[1].kind == nkNilLit)
result = newIntNodeT(toInt128(ord(val)), n, idgen, g)
of mLtI, mLtB, mLtEnum, mLtCh:
result = newIntNodeT(toInt128(ord(getOrdValue(a) < getOrdValue(b))), n, idgen, g)
of mLeI, mLeB, mLeEnum, mLeCh:
result = newIntNodeT(toInt128(ord(getOrdValue(a) <= getOrdValue(b))), n, idgen, g)
of mEqI, mEqB, mEqEnum, mEqCh:
result = newIntNodeT(toInt128(ord(getOrdValue(a) == getOrdValue(b))), n, idgen, g)
of mLtF64: result = newIntNodeT(toInt128(ord(getFloat(a) < getFloat(b))), n, idgen, g)
of mLeF64: result = newIntNodeT(toInt128(ord(getFloat(a) <= getFloat(b))), n, idgen, g)
of mEqF64: result = newIntNodeT(toInt128(ord(getFloat(a) == getFloat(b))), n, idgen, g)
of mLtStr: result = newIntNodeT(toInt128(ord(getStr(a) < getStr(b))), n, idgen, g)
of mLeStr: result = newIntNodeT(toInt128(ord(getStr(a) <= getStr(b))), n, idgen, g)
of mEqStr: result = newIntNodeT(toInt128(ord(getStr(a) == getStr(b))), n, idgen, g)
of mLtU:
result = newIntNodeT(toInt128(ord(`<%`(toInt64(getOrdValue(a)), toInt64(getOrdValue(b))))), n, idgen, g)
of mLeU:
result = newIntNodeT(toInt128(ord(`<=%`(toInt64(getOrdValue(a)), toInt64(getOrdValue(b))))), n, idgen, g)
of mBitandI, mAnd: result = newIntNodeT(bitand(a.getInt, b.getInt), n, idgen, g)
of mBitorI, mOr: result = newIntNodeT(bitor(getInt(a), getInt(b)), n, idgen, g)
of mBitxorI, mXor: result = newIntNodeT(bitxor(getInt(a), getInt(b)), n, idgen, g)
of mAddU:
let val = maskBytes(getInt(a) + getInt(b), int(getSize(g.config, n.typ)))
result = newIntNodeT(val, n, idgen, g)
of mSubU:
let val = maskBytes(getInt(a) - getInt(b), int(getSize(g.config, n.typ)))
result = newIntNodeT(val, n, idgen, g)
# echo "subU: ", val, " n: ", n, " result: ", val
of mMulU:
let val = maskBytes(getInt(a) * getInt(b), int(getSize(g.config, n.typ)))
result = newIntNodeT(val, n, idgen, g)
of mModU:
let argA = maskBytes(getInt(a), int(getSize(g.config, a.typ)))
let argB = maskBytes(getInt(b), int(getSize(g.config, a.typ)))
if argB != Zero:
result = newIntNodeT(argA mod argB, n, idgen, g)
of mDivU:
let argA = maskBytes(getInt(a), int(getSize(g.config, a.typ)))
let argB = maskBytes(getInt(b), int(getSize(g.config, a.typ)))
if argB != Zero:
result = newIntNodeT(argA div argB, n, idgen, g)
of mLeSet: result = newIntNodeT(toInt128(ord(containsSets(g.config, a, b))), n, idgen, g)
of mEqSet: result = newIntNodeT(toInt128(ord(equalSets(g.config, a, b))), n, idgen, g)
of mLtSet:
result = newIntNodeT(toInt128(ord(
containsSets(g.config, a, b) and not equalSets(g.config, a, b))), n, idgen, g)
of mMulSet:
result = nimsets.intersectSets(g.config, a, b)
result.info = n.info
of mPlusSet:
result = nimsets.unionSets(g.config, a, b)
result.info = n.info
of mMinusSet:
result = nimsets.diffSets(g.config, a, b)
result.info = n.info
of mXorSet:
result = nimsets.symdiffSets(g.config, a, b)
result.info = n.info
of mConStrStr: result = newStrNodeT(getStrOrChar(a) & getStrOrChar(b), n, g)
of mInSet: result = newIntNodeT(toInt128(ord(inSet(a, b))), n, idgen, g)
of mRepr:
# BUGFIX: we cannot eval mRepr here for reasons that I forgot.
discard
of mBoolToStr:
if getOrdValue(a) == 0: result = newStrNodeT("false", n, g)
else: result = newStrNodeT("true", n, g)
of mCStrToStr, mCharToStr:
result = newStrNodeT(getStrOrChar(a), n, g)
of mStrToStr: result = newStrNodeT(getStrOrChar(a), n, g)
of mEnumToStr: result = newStrNodeT(ordinalValToString(a, g), n, g)
of mArrToSeq:
result = copyTree(a)
result.typ = n.typ
of mCompileOption:
result = newIntNodeT(toInt128(ord(commands.testCompileOption(g.config, a.getStr, n.info))), n, idgen, g)
of mCompileOptionArg:
result = newIntNodeT(toInt128(ord(
testCompileOptionArg(g.config, getStr(a), getStr(b), n.info))), n, idgen, g)
of mEqProc:
result = newIntNodeT(toInt128(ord(
exprStructuralEquivalent(a, b, strictSymEquality=true))), n, idgen, g)
else: discard
proc getConstIfExpr(c: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
result = nil
for i in 0..<n.len:
var it = n[i]
if it.len == 2:
var e = getConstExpr(c, it[0], idgen, g)
if e == nil: return nil
if getOrdValue(e) != 0:
if result == nil:
result = getConstExpr(c, it[1], idgen, g)
if result == nil: return
elif it.len == 1:
if result == nil: result = getConstExpr(c, it[0], idgen, g)
else: internalError(g.config, it.info, "getConstIfExpr()")
proc leValueConv*(a, b: PNode): bool =
result = false
case a.kind
of nkCharLit..nkUInt64Lit:
case b.kind
of nkCharLit..nkUInt64Lit: result = a.getInt <= b.getInt
of nkFloatLit..nkFloat128Lit: result = a.intVal <= round(b.floatVal).int
else: result = false #internalError(a.info, "leValueConv")
of nkFloatLit..nkFloat128Lit:
case b.kind
of nkFloatLit..nkFloat128Lit: result = a.floatVal <= b.floatVal
of nkCharLit..nkUInt64Lit: result = a.floatVal <= toFloat64(b.getInt)
else: result = false # internalError(a.info, "leValueConv")
else: result = false # internalError(a.info, "leValueConv")
proc magicCall(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
if n.len <= 1: return
var s = n[0].sym
var a = getConstExpr(m, n[1], idgen, g)
var b, c: PNode = nil
if a == nil: return
if n.len > 2:
b = getConstExpr(m, n[2], idgen, g)
if b == nil: return
if n.len > 3:
c = getConstExpr(m, n[3], idgen, g)
if c == nil: return
result = evalOp(s.magic, n, a, b, c, idgen, g)
proc getAppType(n: PNode; g: ModuleGraph): PNode =
if g.config.globalOptions.contains(optGenDynLib):
result = newStrNodeT("lib", n, g)
elif g.config.globalOptions.contains(optGenStaticLib):
result = newStrNodeT("staticlib", n, g)
elif g.config.globalOptions.contains(optGenGuiApp):
result = newStrNodeT("gui", n, g)
else:
result = newStrNodeT("console", n, g)
proc rangeCheck(n: PNode, value: Int128; g: ModuleGraph) =
if value < firstOrd(g.config, n.typ) or value > lastOrd(g.config, n.typ):
localError(g.config, n.info, "cannot convert " & $value &
" to " & typeToString(n.typ))
proc floatRangeCheck(n: PNode, value: BiggestFloat; g: ModuleGraph) =
if value < firstFloat(n.typ) or value > lastFloat(n.typ):
localError(g.config, n.info, "cannot convert " & $value &
" to " & typeToString(n.typ))
proc foldConv(n, a: PNode; idgen: IdGenerator; g: ModuleGraph; check = false): PNode =
let dstTyp = skipTypes(n.typ, abstractRange - {tyTypeDesc})
let srcTyp = skipTypes(a.typ, abstractRange - {tyTypeDesc})
# if srcTyp.kind == tyUInt64 and "FFFFFF" in $n:
# echo "n: ", n, " a: ", a
# echo "from: ", srcTyp, " to: ", dstTyp, " check: ", check
# echo getInt(a)
# echo high(int64)
# writeStackTrace()
case dstTyp.kind
of tyBool:
case srcTyp.kind
of tyFloat..tyFloat64:
result = newIntNodeT(toInt128(getFloat(a) != 0.0), n, idgen, g)
of tyChar, tyUInt..tyUInt64, tyInt..tyInt64:
result = newIntNodeT(toInt128(a.getOrdValue != 0), n, idgen, g)
of tyBool, tyEnum: # xxx shouldn't we disallow `tyEnum`?
result = a
result.typ = n.typ
else:
raiseAssert $srcTyp.kind
of tyInt..tyInt64, tyUInt..tyUInt64:
case srcTyp.kind
of tyFloat..tyFloat64:
result = newIntNodeT(toInt128(getFloat(a)), n, idgen, g)
of tyChar, tyUInt..tyUInt64, tyInt..tyInt64:
var val = a.getOrdValue
if dstTyp.kind in {tyUInt..tyUInt64}:
result = newIntNodeT(maskBytes(val, int getSize(g.config, dstTyp)), n, idgen, g)
result.transitionIntKind(nkUIntLit)
else:
if check: rangeCheck(n, val, g)
result = newIntNodeT(val, n, idgen, g)
else:
result = a
result.typ = n.typ
if check and result.kind in {nkCharLit..nkUInt64Lit} and
dstTyp.kind notin {tyUInt..tyUInt64}:
rangeCheck(n, getInt(result), g)
of tyFloat..tyFloat64:
case srcTyp.kind
of tyInt..tyInt64, tyUInt..tyUInt64, tyEnum, tyBool, tyChar:
result = newFloatNodeT(toFloat64(getOrdValue(a)), n, g)
else:
result = a
result.typ = n.typ
of tyOpenArray, tyVarargs, tyProc, tyPointer:
result = nil
else:
result = a
result.typ = n.typ
proc getArrayConstr(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
if n.kind == nkBracket:
result = n
else:
result = getConstExpr(m, n, idgen, g)
if result == nil: result = n
proc foldArrayAccess(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
var x = getConstExpr(m, n[0], idgen, g)
if x == nil or x.typ.skipTypes({tyGenericInst, tyAlias, tySink}).kind == tyTypeDesc:
return
var y = getConstExpr(m, n[1], idgen, g)
if y == nil: return
var idx = toInt64(getOrdValue(y))
case x.kind
of nkPar, nkTupleConstr:
if idx >= 0 and idx < x.len:
result = x.sons[idx]
if result.kind == nkExprColonExpr: result = result[1]
else:
result = nil
#localError(g.config, n.info, formatErrorIndexBound(idx, x.len-1) & $n)
of nkBracket:
idx -= toInt64(firstOrd(g.config, x.typ))
if idx >= 0 and idx < x.len: result = x[int(idx)]
else:
result = nil
#localError(g.config, n.info, formatErrorIndexBound(idx, x.len-1) & $n)
of nkStrLit..nkTripleStrLit:
result = newNodeIT(nkCharLit, x.info, n.typ)
if idx >= 0 and idx < x.strVal.len:
result.intVal = ord(x.strVal[int(idx)])
else:
result = nil
#localError(g.config, n.info, formatErrorIndexBound(idx, x.strVal.len-1) & $n)
else: result = nil
proc foldFieldAccess(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
# a real field access; proc calls have already been transformed
result = nil
if n[1].kind != nkSym: return nil
var x = getConstExpr(m, n[0], idgen, g)
if x == nil or x.kind notin {nkObjConstr, nkPar, nkTupleConstr}: return
var field = n[1].sym
for i in ord(x.kind == nkObjConstr)..<x.len:
var it = x[i]
if it.kind != nkExprColonExpr:
# lookup per index:
result = x[field.position]
if result.kind == nkExprColonExpr: result = result[1]
return
if it[0].sym.name.id == field.name.id:
result = x[i][1]
return
localError(g.config, n.info, "field not found: " & field.name.s)
proc foldConStrStr(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
result = newNodeIT(nkStrLit, n.info, n.typ)
result.strVal = ""
for i in 1..<n.len:
let a = getConstExpr(m, n[i], idgen, g)
if a == nil: return nil
result.strVal.add(getStrOrChar(a))
proc newSymNodeTypeDesc*(s: PSym; idgen: IdGenerator; info: TLineInfo): PNode =
result = newSymNode(s, info)
if s.typ.kind != tyTypeDesc:
result.typ = newType(tyTypeDesc, idgen, s.owner)
result.typ.addSonSkipIntLit(s.typ, idgen)
else:
result.typ = s.typ
proc foldDefine(m, s: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
result = nil
var name = s.name.s
let prag = extractPragma(s)
if prag != nil:
for it in prag:
if it.kind in nkPragmaCallKinds and it.len == 2 and it[0].kind == nkIdent:
let word = whichKeyword(it[0].ident)
if word in {wStrDefine, wIntDefine, wBoolDefine, wDefine}:
# should be processed in pragmas.nim already
if it[1].kind in {nkStrLit, nkRStrLit, nkTripleStrLit}:
name = it[1].strVal
if isDefined(g.config, name):
let str = g.config.symbols[name]
case s.magic
of mIntDefine:
try:
result = newIntNodeT(toInt128(str.parseInt), n, idgen, g)
except ValueError:
localError(g.config, s.info,
"{.intdefine.} const was set to an invalid integer: '" &
str & "'")
of mStrDefine:
result = newStrNodeT(str, n, g)
of mBoolDefine:
try:
result = newIntNodeT(toInt128(str.parseBool.int), n, idgen, g)
except ValueError:
localError(g.config, s.info,
"{.booldefine.} const was set to an invalid bool: '" &
str & "'")
of mGenericDefine:
let rawTyp = s.typ
# pretend we don't support distinct types
let typ = rawTyp.skipTypes(abstractVarRange-{tyDistinct})
try:
template intNode(value): PNode =
let val = toInt128(value)
rangeCheck(n, val, g)
newIntNodeT(val, n, idgen, g)
case typ.kind
of tyString, tyCstring:
result = newStrNodeT(str, n, g)
of tyInt..tyInt64:
result = intNode(str.parseBiggestInt)
of tyUInt..tyUInt64:
result = intNode(str.parseBiggestUInt)
of tyBool:
result = intNode(str.parseBool.int)
of tyEnum:
# compile time parseEnum
let ident = getIdent(g.cache, str)
for e in typ.n:
if e.kind != nkSym: internalError(g.config, "foldDefine for enum")
let es = e.sym
let match =
if es.ast.isNil:
es.name.id == ident.id
else:
es.ast.strVal == str
if match:
result = intNode(es.position)
break
if result.isNil:
raise newException(ValueError, "invalid enum value: " & str)
else:
localError(g.config, s.info, "unsupported type $1 for define '$2'" %
[typeToString(rawTyp), name])
except ValueError as e:
localError(g.config, s.info,
"could not process define '$1' of type $2; $3" %
[name, typeToString(rawTyp), e.msg])
else: result = copyTree(s.astdef) # unreachable
else:
result = copyTree(s.astdef)
if result != nil:
result.info = n.info
proc getConstExpr(m: PSym, n: PNode; idgen: IdGenerator; g: ModuleGraph): PNode =
result = nil
case n.kind
of nkSym:
var s = n.sym
case s.kind
of skEnumField:
result = newIntNodeT(toInt128(s.position), n, idgen, g)
of skConst:
case s.magic
of mIsMainModule: result = newIntNodeT(toInt128(ord(sfMainModule in m.flags)), n, idgen, g)
of mCompileDate: result = newStrNodeT(getDateStr(), n, g)
of mCompileTime: result = newStrNodeT(getClockStr(), n, g)
of mCpuEndian: result = newIntNodeT(toInt128(ord(CPU[g.config.target.targetCPU].endian)), n, idgen, g)
of mHostOS: result = newStrNodeT(toLowerAscii(platform.OS[g.config.target.targetOS].name), n, g)
of mHostCPU: result = newStrNodeT(platform.CPU[g.config.target.targetCPU].name.toLowerAscii, n, g)
of mBuildOS: result = newStrNodeT(toLowerAscii(platform.OS[g.config.target.hostOS].name), n, g)
of mBuildCPU: result = newStrNodeT(platform.CPU[g.config.target.hostCPU].name.toLowerAscii, n, g)
of mAppType: result = getAppType(n, g)
of mIntDefine, mStrDefine, mBoolDefine, mGenericDefine:
result = foldDefine(m, s, n, idgen, g)
else:
result = copyTree(s.astdef)
if result != nil:
result.info = n.info
of skProc, skFunc, skMethod:
result = n
of skParam:
if s.typ != nil and s.typ.kind == tyTypeDesc:
result = newSymNodeTypeDesc(s, idgen, n.info)
of skType:
# XXX gensym'ed symbols can come here and cannot be resolved. This is
# dirty, but correct.
if s.typ != nil:
result = newSymNodeTypeDesc(s, idgen, n.info)
of skGenericParam:
if s.typ.kind == tyStatic:
if s.typ.n != nil and tfUnresolved notin s.typ.flags:
result = s.typ.n
result.typ = s.typ.base
elif s.typ.isIntLit:
result = s.typ.n
else:
result = newSymNodeTypeDesc(s, idgen, n.info)
else: discard
of nkCharLit..nkNilLit:
result = copyNode(n)
of nkIfExpr:
result = getConstIfExpr(m, n, idgen, g)
of nkCallKinds:
if n[0].kind != nkSym: return
var s = n[0].sym
if s.kind != skProc and s.kind != skFunc: return
try:
case s.magic
of mNone:
# If it has no sideEffect, it should be evaluated. But not here.
return
of mLow:
if skipTypes(n[1].typ, abstractVarRange).kind in tyFloat..tyFloat64:
result = newFloatNodeT(firstFloat(n[1].typ), n, g)
else:
result = newIntNodeT(firstOrd(g.config, n[1].typ), n, idgen, g)
of mHigh:
if skipTypes(n[1].typ, abstractVar+{tyUserTypeClassInst}).kind notin
{tySequence, tyString, tyCstring, tyOpenArray, tyVarargs}:
if skipTypes(n[1].typ, abstractVarRange).kind in tyFloat..tyFloat64:
result = newFloatNodeT(lastFloat(n[1].typ), n, g)
else:
result = newIntNodeT(lastOrd(g.config, skipTypes(n[1].typ, abstractVar)), n, idgen, g)
else:
var a = getArrayConstr(m, n[1], idgen, g)
if a.kind == nkBracket:
# we can optimize it away:
result = newIntNodeT(toInt128(a.len-1), n, idgen, g)
of mLengthOpenArray:
var a = getArrayConstr(m, n[1], idgen, g)
if a.kind == nkBracket:
# we can optimize it away! This fixes the bug ``len(134)``.
result = newIntNodeT(toInt128(a.len), n, idgen, g)
else:
result = magicCall(m, n, idgen, g)
of mLengthArray:
# It doesn't matter if the argument is const or not for mLengthArray.
# This fixes bug #544.
result = newIntNodeT(lengthOrd(g.config, n[1].typ), n, idgen, g)
of mSizeOf:
result = foldSizeOf(g.config, n, nil)
of mAlignOf:
result = foldAlignOf(g.config, n, nil)
of mOffsetOf:
result = foldOffsetOf(g.config, n, nil)
of mAstToStr:
result = newStrNodeT(renderTree(n[1], {renderNoComments}), n, g)
of mConStrStr:
result = foldConStrStr(m, n, idgen, g)
of mIs:
# The only kind of mIs node that comes here is one depending on some
# generic parameter and that's (hopefully) handled at instantiation time
discard
else:
result = magicCall(m, n, idgen, g)
except OverflowDefect:
localError(g.config, n.info, "over- or underflow")
except DivByZeroDefect:
localError(g.config, n.info, "division by zero")
of nkAddr:
result = nil # don't fold paths containing nkAddr
of nkBracket, nkCurly:
result = copyNode(n)
for son in n.items:
var a = getConstExpr(m, son, idgen, g)
if a == nil: return nil
result.add a
incl(result.flags, nfAllConst)
of nkRange:
var a = getConstExpr(m, n[0], idgen, g)
if a == nil: return
var b = getConstExpr(m, n[1], idgen, g)
if b == nil: return
result = copyNode(n)
result.add a
result.add b
#of nkObjConstr:
# result = copyTree(n)
# for i in 1..<n.len:
# var a = getConstExpr(m, n[i][1])
# if a == nil: return nil
# result[i][1] = a
# incl(result.flags, nfAllConst)
of nkPar, nkTupleConstr:
# tuple constructor
result = copyNode(n)
if (n.len > 0) and (n[0].kind == nkExprColonExpr):
for expr in n.items:
let exprNew = copyNode(expr) # nkExprColonExpr
exprNew.add expr[0]
let a = getConstExpr(m, expr[1], idgen, g)
if a == nil: return nil
exprNew.add a
result.add exprNew
else:
for expr in n.items:
let a = getConstExpr(m, expr, idgen, g)
if a == nil: return nil
result.add a
incl(result.flags, nfAllConst)
of nkChckRangeF, nkChckRange64, nkChckRange:
var a = getConstExpr(m, n[0], idgen, g)
if a == nil: return
if leValueConv(n[1], a) and leValueConv(a, n[2]):
result = a # a <= x and x <= b
result.typ = n.typ
elif n.typ.kind in {tyUInt..tyUInt64}:
discard "don't check uints"
else:
localError(g.config, n.info,
"conversion from $1 to $2 is invalid" %
[typeToString(n[0].typ), typeToString(n.typ)])
of nkStringToCString, nkCStringToString:
var a = getConstExpr(m, n[0], idgen, g)
if a == nil: return
result = a
result.typ = n.typ
of nkHiddenStdConv, nkHiddenSubConv, nkConv:
var a = getConstExpr(m, n[1], idgen, g)
if a == nil: return
result = foldConv(n, a, idgen, g, check=true)
of nkDerefExpr, nkHiddenDeref:
let a = getConstExpr(m, n[0], idgen, g)
if a != nil and a.kind == nkNilLit:
result = nil
#localError(g.config, n.info, "nil dereference is not allowed")
of nkCast:
var a = getConstExpr(m, n[1], idgen, g)
if a == nil: return
if n.typ != nil and n.typ.kind in NilableTypes and
not (n.typ.kind == tyProc and a.typ.kind == tyProc):
# we allow compile-time 'cast' for pointer types:
result = a
result.typ = n.typ
of nkBracketExpr: result = foldArrayAccess(m, n, idgen, g)
of nkDotExpr: result = foldFieldAccess(m, n, idgen, g)
of nkCheckedFieldExpr:
assert n[0].kind == nkDotExpr
result = foldFieldAccess(m, n[0], idgen, g)
of nkStmtListExpr:
var i = 0
while i <= n.len - 2:
if n[i].kind in {nkComesFrom, nkCommentStmt, nkEmpty}: i.inc
else: break
if i == n.len - 1:
result = getConstExpr(m, n[i], idgen, g)
else:
discard