1936 lines
67 KiB
C++

//
//Copyright (C) 2002-2005 3Dlabs Inc. Ltd.
//Copyright (C) 2012-2013 LunarG, Inc.
//
//All rights reserved.
//
//Redistribution and use in source and binary forms, with or without
//modification, are permitted provided that the following conditions
//are met:
//
// Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
//
// Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following
// disclaimer in the documentation and/or other materials provided
// with the distribution.
//
// Neither the name of 3Dlabs Inc. Ltd. nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
//THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
//"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
//LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
//FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
//COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
//INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
//BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
//LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
//CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
//LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
//ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
//POSSIBILITY OF SUCH DAMAGE.
//
#include "ParseHelper.h"
#include "Include/InitializeParseContext.h"
#include "osinclude.h"
#include <stdarg.h>
#include <algorithm>
TParseContext::TParseContext(TSymbolTable& symt, TIntermediate& interm, int v, EProfile p, EShLanguage L, TInfoSink& is,
bool fc, EShMessages m) :
intermediate(interm), symbolTable(symt), infoSink(is), language(L), treeRoot(0),
numErrors(0), lexAfterType(false), loopNestingLevel(0),
structNestingLevel(0), inTypeParen(false),
version(v), profile(p), forwardCompatible(fc), messages(m),
contextPragma(true, false)
{
for (int type = 0; type < EbtNumTypes; ++type)
defaultPrecision[type] = EpqNone;
for (int type = 0; type < maxSamplerIndex; ++type)
defaultSamplerPrecision[type] = EpqNone;
if (profile == EEsProfile) {
TSampler sampler;
sampler.set(EbtFloat, Esd2D);
defaultSamplerPrecision[computeSamplerTypeIndex(sampler)] = EpqLow;
sampler.set(EbtFloat, EsdCube);
defaultSamplerPrecision[computeSamplerTypeIndex(sampler)] = EpqLow;
switch (language) {
case EShLangVertex:
defaultPrecision[EbtInt] = EpqHigh;
defaultPrecision[EbtUint] = EpqHigh;
defaultPrecision[EbtFloat] = EpqHigh;
defaultPrecision[EbtSampler] = EpqLow;
// TODO: functionality: need default precisions per sampler type
break;
case EShLangFragment:
defaultPrecision[EbtInt] = EpqMedium;
defaultPrecision[EbtUint] = EpqMedium;
defaultPrecision[EbtSampler] = EpqLow;
// TODO: semantics: give error when using float in frag shader without default precision
break;
default:
error(1, "INTERNAL ERROR", "unexpected language", "");
}
}
defaultGlobalQualification.clear();
defaultGlobalQualification.layoutMatrix = ElmColumnMajor;
defaultGlobalQualification.layoutPacking = ElpShared;
defaultGlobalQualification.layoutSlotLocation = 0;
}
// Get code that is not part of a shared symbol table, specific to this shader
// or needed by CPP (which does not have a shared symbol table).
const char* TParseContext::getPreamble()
{
if (profile == EEsProfile)
return "#define GL_ES 1\n";
else
return 0;
}
///////////////////////////////////////////////////////////////////////
//
// Sub- vector and matrix fields
//
////////////////////////////////////////////////////////////////////////
//
// Look at a '.' field selector string and change it into offsets
// for a vector.
//
// Returns true if there is no error.
//
bool TParseContext::parseVectorFields(const TString& compString, int vecSize, TVectorFields& fields, int line)
{
fields.num = (int) compString.size();
if (fields.num > 4) {
error(line, "illegal vector field selection", compString.c_str(), "");
return false;
}
enum {
exyzw,
ergba,
estpq,
} fieldSet[4];
for (int i = 0; i < fields.num; ++i) {
switch (compString[i]) {
case 'x':
fields.offsets[i] = 0;
fieldSet[i] = exyzw;
break;
case 'r':
fields.offsets[i] = 0;
fieldSet[i] = ergba;
break;
case 's':
fields.offsets[i] = 0;
fieldSet[i] = estpq;
break;
case 'y':
fields.offsets[i] = 1;
fieldSet[i] = exyzw;
break;
case 'g':
fields.offsets[i] = 1;
fieldSet[i] = ergba;
break;
case 't':
fields.offsets[i] = 1;
fieldSet[i] = estpq;
break;
case 'z':
fields.offsets[i] = 2;
fieldSet[i] = exyzw;
break;
case 'b':
fields.offsets[i] = 2;
fieldSet[i] = ergba;
break;
case 'p':
fields.offsets[i] = 2;
fieldSet[i] = estpq;
break;
case 'w':
fields.offsets[i] = 3;
fieldSet[i] = exyzw;
break;
case 'a':
fields.offsets[i] = 3;
fieldSet[i] = ergba;
break;
case 'q':
fields.offsets[i] = 3;
fieldSet[i] = estpq;
break;
default:
error(line, "illegal vector field selection", compString.c_str(), "");
return false;
}
}
for (int i = 0; i < fields.num; ++i) {
if (fields.offsets[i] >= vecSize) {
error(line, "vector field selection out of range", compString.c_str(), "");
return false;
}
if (i > 0) {
if (fieldSet[i] != fieldSet[i-1]) {
error(line, "illegal - vector component fields not from the same set", compString.c_str(), "");
return false;
}
}
}
return true;
}
///////////////////////////////////////////////////////////////////////
//
// Errors
//
////////////////////////////////////////////////////////////////////////
//
// Used to output syntax, parsing, and semantic errors.
//
void C_DECL TParseContext::error(TSourceLoc nLine, const char *szReason, const char *szToken,
const char *szExtraInfoFormat, ...)
{
const int maxSize = GlslangMaxTokenLength + 200;
char szExtraInfo[maxSize];
va_list marker;
va_start(marker, szExtraInfoFormat);
safe_vsprintf(szExtraInfo, maxSize, szExtraInfoFormat, marker);
infoSink.info.prefix(EPrefixError);
infoSink.info.location(nLine);
infoSink.info << "'" << szToken << "' : " << szReason << " " << szExtraInfo << "\n";
va_end(marker);
++numErrors;
}
void C_DECL TParseContext::warn(TSourceLoc nLine, const char *szReason, const char *szToken,
const char *szExtraInfoFormat, ...)
{
if (messages & EShMsgSuppressWarnings)
return;
const int maxSize = GlslangMaxTokenLength + 200;
char szExtraInfo[maxSize];
va_list marker;
va_start(marker, szExtraInfoFormat);
safe_vsprintf(szExtraInfo, maxSize, szExtraInfoFormat, marker);
infoSink.info.prefix(EPrefixWarning);
infoSink.info.location(nLine);
infoSink.info << "'" << szToken << "' : " << szReason << " " << szExtraInfo << "\n";
va_end(marker);
}
TIntermTyped* TParseContext::handleVariable(int line, TSymbol* symbol, TString* string)
{
TIntermTyped* node = 0;
TAnonMember* anon = symbol ? symbol->getAsAnonMember() : 0;
if (anon) {
// it was a member of an anonymous container, have to insert its dereference
TVariable* variable = anon->getAnonContainer().getAsVariable();
TIntermTyped* container = intermediate.addSymbol(variable->getUniqueId(), variable->getName(), variable->getType(), line);
constUnion* unionArray = new constUnion[1];
unionArray->setUConst(anon->getMemberNumber());
TIntermTyped* constNode = intermediate.addConstantUnion(unionArray, TType(EbtUint, EvqConst), line);
node = intermediate.addIndex(EOpIndexDirectStruct, container, constNode, line);
node->setType(*(*variable->getType().getStruct())[anon->getMemberNumber()].type);
} else {
// The symbol table search was done in the lexical phase, but
// if this is a new symbol, it wouldn't have found it.
const TVariable* variable = symbol ? symbol->getAsVariable() : 0;
if (symbol && ! variable)
error(line, "variable name expected", string->c_str(), "");
if (! variable)
variable = new TVariable(string, TType(EbtVoid));
// don't delete $1.string, it's used by error recovery, and the pool
// pop will reclaim the memory
if (variable->getType().getQualifier().storage == EvqConst ) {
constUnion* constArray = variable->getConstUnionPointer();
TType t(variable->getType());
node = intermediate.addConstantUnion(constArray, t, line);
} else
node = intermediate.addSymbol(variable->getUniqueId(), variable->getName(), variable->getType(), line);
}
return node;
}
//
// Same error message for all places assignments don't work.
//
void TParseContext::assignError(int line, const char* op, TString left, TString right)
{
error(line, "", op, "cannot convert from '%s' to '%s'",
right.c_str(), left.c_str());
}
//
// Same error message for all places unary operations don't work.
//
void TParseContext::unaryOpError(int line, const char* op, TString operand)
{
error(line, " wrong operand type", op,
"no operation '%s' exists that takes an operand of type %s (or there is no acceptable conversion)",
op, operand.c_str());
}
//
// Same error message for all binary operations don't work.
//
void TParseContext::binaryOpError(int line, const char* op, TString left, TString right)
{
error(line, " wrong operand types:", op,
"no operation '%s' exists that takes a left-hand operand of type '%s' and "
"a right operand of type '%s' (or there is no acceptable conversion)",
op, left.c_str(), right.c_str());
}
//
// A basic type of EbtVoid is a key that the name string was seen in the source, but
// it was not found as a variable in the symbol table. If so, give the error
// message and insert a dummy variable in the symbol table to prevent future errors.
//
void TParseContext::variableCheck(TIntermTyped*& nodePtr)
{
TIntermSymbol* symbol = nodePtr->getAsSymbolNode();
if (! symbol)
return;
if (symbol->getType().getBasicType() == EbtVoid) {
error(symbol->getLine(), "undeclared identifier", symbol->getSymbol().c_str(), "");
// Add to symbol table to prevent future error messages on the same name
TVariable* fakeVariable = new TVariable(&symbol->getSymbol(), TType(EbtFloat));
symbolTable.insert(*fakeVariable);
// substitute a symbol node for this new variable
nodePtr = intermediate.addSymbol(fakeVariable->getUniqueId(),
fakeVariable->getName(),
fakeVariable->getType(), symbol->getLine());
} else {
switch (symbol->getQualifier().storage) {
case EvqPointCoord:
profileRequires(symbol->getLine(), ENoProfile, 120, 0, "gl_PointCoord");
break;
default: break; // some compilers want this
}
}
}
//
// Both test and if necessary, spit out an error, to see if the node is really
// an l-value that can be operated on this way.
//
// Returns true if the was an error.
//
bool TParseContext::lValueErrorCheck(int line, const char* op, TIntermTyped* node)
{
TIntermSymbol* symNode = node->getAsSymbolNode();
TIntermBinary* binaryNode = node->getAsBinaryNode();
if (binaryNode) {
bool errorReturn;
switch(binaryNode->getOp()) {
case EOpIndexDirect:
case EOpIndexIndirect:
case EOpIndexDirectStruct:
return lValueErrorCheck(line, op, binaryNode->getLeft());
case EOpVectorSwizzle:
errorReturn = lValueErrorCheck(line, op, binaryNode->getLeft());
if (!errorReturn) {
int offset[4] = {0,0,0,0};
TIntermTyped* rightNode = binaryNode->getRight();
TIntermAggregate *aggrNode = rightNode->getAsAggregate();
for (TIntermSequence::iterator p = aggrNode->getSequence().begin();
p != aggrNode->getSequence().end(); p++) {
int value = (*p)->getAsTyped()->getAsConstantUnion()->getUnionArrayPointer()->getIConst();
offset[value]++;
if (offset[value] > 1) {
error(line, " l-value of swizzle cannot have duplicate components", op, "", "");
return true;
}
}
}
return errorReturn;
default:
break;
}
error(line, " l-value required", op, "", "");
return true;
}
const char* symbol = 0;
if (symNode != 0)
symbol = symNode->getSymbol().c_str();
const char* message = 0;
switch (node->getQualifier().storage) {
case EvqConst: message = "can't modify a const"; break;
case EvqConstReadOnly: message = "can't modify a const"; break;
case EvqVaryingIn: message = "can't modify shader input"; break;
case EvqUniform: message = "can't modify a uniform"; break;
case EvqInstanceId: message = "can't modify gl_InstanceID"; break;
case EvqVertexId: message = "can't modify gl_VertexID"; break;
case EvqFace: message = "can't modify gl_FrontFace"; break;
case EvqFragCoord: message = "can't modify gl_FragCoord"; break;
case EvqPointCoord: message = "can't modify gl_PointCoord"; break;
default:
//
// Type that can't be written to?
//
switch (node->getBasicType()) {
case EbtSampler:
message = "can't modify a sampler";
break;
case EbtVoid:
message = "can't modify void";
break;
default:
break;
}
}
if (message == 0 && binaryNode == 0 && symNode == 0) {
error(line, " l-value required", op, "", "");
return true;
}
//
// Everything else is okay, no error.
//
if (message == 0)
return false;
//
// If we get here, we have an error and a message.
//
if (symNode)
error(line, " l-value required", op, "\"%s\" (%s)", symbol, message);
else
error(line, " l-value required", op, "(%s)", message);
return true;
}
//
// Both test, and if necessary spit out an error, to see if the node is really
// a constant.
//
void TParseContext::constCheck(TIntermTyped* node)
{
if (node->getQualifier().storage != EvqConst)
error(node->getLine(), "constant expression required", "", "");
}
//
// Both test, and if necessary spit out an error, to see if the node is really
// an integer.
//
void TParseContext::integerCheck(TIntermTyped* node, const char* token)
{
if ((node->getBasicType() == EbtInt || node->getBasicType() == EbtUint) && node->isScalar() && ! node->isArray())
return;
error(node->getLine(), "scalar integer expression required", token, "");
}
//
// Both test, and if necessary spit out an error, to see if we are currently
// globally scoped.
//
void TParseContext::globalCheck(int line, bool global, const char* token)
{
if (! global)
error(line, "only allowed at global scope", token, "");
}
//
// For now, keep it simple: if it starts "gl_", it's reserved, independent
// of scope. Except, if the symbol table is at the built-in push-level,
// which is when we are parsing built-ins.
//
bool TParseContext::reservedErrorCheck(int line, const TString& identifier)
{
if (!symbolTable.atBuiltInLevel()) {
if (identifier.substr(0, 3) == TString("gl_")) {
error(line, "reserved built-in name", "gl_", "");
return true;
}
if (identifier.find("__") != TString::npos) {
error(line, "Two consecutive underscores are reserved for future use.", identifier.c_str(), "", "");
return true;
}
}
return false;
}
//
// Make sure there is enough data provided to the constructor to build
// something of the type of the constructor. Also returns the type of
// the constructor.
//
// Returns true if there was an error in construction.
//
bool TParseContext::constructorError(int line, TIntermNode* node, TFunction& function, TOperator op, TType& type)
{
type = function.getReturnType();
bool constructingMatrix = false;
switch(op) {
case EOpConstructMat2x2:
case EOpConstructMat2x3:
case EOpConstructMat2x4:
case EOpConstructMat3x2:
case EOpConstructMat3x3:
case EOpConstructMat3x4:
case EOpConstructMat4x2:
case EOpConstructMat4x3:
case EOpConstructMat4x4:
case EOpConstructDMat2x2:
case EOpConstructDMat2x3:
case EOpConstructDMat2x4:
case EOpConstructDMat3x2:
case EOpConstructDMat3x3:
case EOpConstructDMat3x4:
case EOpConstructDMat4x2:
case EOpConstructDMat4x3:
case EOpConstructDMat4x4:
constructingMatrix = true;
break;
default:
break;
}
//
// Note: It's okay to have too many components available, but not okay to have unused
// arguments. 'full' will go to true when enough args have been seen. If we loop
// again, there is an extra argument, so 'overfull' will become true.
//
int size = 0;
bool constType = true;
bool full = false;
bool overFull = false;
bool matrixInMatrix = false;
bool arrayArg = false;
for (int i = 0; i < function.getParamCount(); ++i) {
size += function[i].type->getObjectSize();
if (constructingMatrix && function[i].type->isMatrix())
matrixInMatrix = true;
if (full)
overFull = true;
if (op != EOpConstructStruct && ! type.isArray() && size >= type.getObjectSize())
full = true;
if (function[i].type->getQualifier().storage != EvqConst)
constType = false;
if (function[i].type->isArray())
arrayArg = true;
}
if (constType)
type.getQualifier().storage = EvqConst;
if (type.isArray()) {
if (type.getArraySize() == 0) {
// auto adapt the constructor type to the number of arguments
type.changeArraySize(function.getParamCount());
} else if (type.getArraySize() != function.getParamCount()) {
error(line, "array constructor needs one argument per array element", "constructor", "");
return true;
}
}
if (arrayArg && op != EOpConstructStruct) {
error(line, "constructing from a non-dereferenced array", "constructor", "");
return true;
}
if (matrixInMatrix && ! type.isArray()) {
profileRequires(line, ENoProfile, 120, 0, "constructing matrix from matrix");
return false;
}
if (overFull) {
error(line, "too many arguments", "constructor", "");
return true;
}
if (op == EOpConstructStruct && ! type.isArray() && type.getStruct()->size() != function.getParamCount()) {
error(line, "Number of constructor parameters does not match the number of structure fields", "constructor", "");
return true;
}
if ((op != EOpConstructStruct && size != 1 && size < type.getObjectSize()) ||
(op == EOpConstructStruct && size < type.getObjectSize())) {
error(line, "not enough data provided for construction", "constructor", "");
return true;
}
TIntermTyped* typed = node->getAsTyped();
if (typed == 0) {
error(line, "constructor argument does not have a type", "constructor", "");
return true;
}
if (op != EOpConstructStruct && typed->getBasicType() == EbtSampler) {
error(line, "cannot convert a sampler", "constructor", "");
return true;
}
if (typed->getBasicType() == EbtVoid) {
error(line, "cannot convert a void", "constructor", "");
return true;
}
return false;
}
// This function checks to see if a void variable has been declared and raise an error message for such a case
//
// returns true in case of an error
//
bool TParseContext::voidErrorCheck(int line, const TString& identifier, const TPublicType& pubType)
{
if (pubType.basicType == EbtVoid) {
error(line, "illegal use of type 'void'", identifier.c_str(), "");
return true;
}
return false;
}
// Checks to see if the node (for the expression) contains a scalar boolean expression or not
void TParseContext::boolCheck(int line, const TIntermTyped* type)
{
if (type->getBasicType() != EbtBool || type->isArray() || type->isMatrix() || type->isVector())
error(line, "boolean expression expected", "", "");
}
// This function checks to see if the node (for the expression) contains a scalar boolean expression or not
void TParseContext::boolCheck(int line, const TPublicType& pType)
{
if (pType.basicType != EbtBool || pType.arraySizes || pType.matrixCols > 1 || (pType.vectorSize > 1))
error(line, "boolean expression expected", "", "");
}
bool TParseContext::samplerErrorCheck(int line, const TPublicType& pType, const char* reason)
{
if (pType.basicType == EbtStruct) {
if (containsSampler(*pType.userDef)) {
error(line, reason, TType::getBasicString(pType.basicType), "(structure cannot contain a sampler or image)");
return true;
}
return false;
} else if (pType.basicType == EbtSampler) {
error(line, reason, TType::getBasicString(pType.basicType), "");
return true;
}
return false;
}
void TParseContext::globalQualifierFix(int line, TQualifier& qualifier, const TPublicType& publicType)
{
if (! symbolTable.atGlobalLevel())
return;
// First, move from parameter qualifiers to shader in/out qualifiers
switch (qualifier.storage) {
case EvqIn:
profileRequires(line, ENoProfile, 130, 0, "in for stage inputs");
profileRequires(line, EEsProfile, 300, 0, "in for stage inputs");
qualifier.storage = EvqVaryingIn;
break;
case EvqOut:
profileRequires(line, ENoProfile, 130, 0, "out for stage outputs");
profileRequires(line, EEsProfile, 300, 0, "out for stage outputs");
qualifier.storage = EvqVaryingOut;
break;
case EvqVaryingIn:
case EvqVaryingOut:
break;
case EvqInOut:
qualifier.storage = EvqVaryingIn;
error(line, "cannot use 'inout' at global scope", "", "");
return;
default:
break;
}
// Do non-in/out error checks
if (qualifier.storage != EvqUniform && samplerErrorCheck(line, publicType, "samplers and images must be uniform"))
return;
if (qualifier.storage != EvqVaryingIn && qualifier.storage != EvqVaryingOut)
return;
// now, knowing it is a shader in/out, do all the in/out semantic checks
if (publicType.basicType == EbtBool) {
error(line, "cannot be bool", getStorageQualifierString(qualifier.storage), "");
return;
}
if (language == EShLangVertex && qualifier.storage == EvqVaryingIn) {
if (publicType.basicType == EbtStruct) {
error(line, "cannot be a structure or array", getStorageQualifierString(qualifier.storage), "");
return;
}
if (publicType.arraySizes) {
requireProfile(line, (EProfileMask)~EEsProfileMask, "vertex input arrays");
profileRequires(line, ENoProfile, 150, 0, "vertex input arrays");
}
}
if (language == EShLangFragment && qualifier.storage == EvqVaryingOut) {
profileRequires(line, EEsProfile, 300, 0, "fragment shader output");
if (publicType.basicType == EbtStruct) {
error(line, "cannot be a structure", getStorageQualifierString(qualifier.storage), "");
return;
}
}
if (publicType.basicType == EbtInt || publicType.basicType == EbtUint || publicType.basicType == EbtDouble) {
profileRequires(line, EEsProfile, 300, 0, "shader input/output");
if (language != EShLangVertex && qualifier.storage == EvqVaryingIn && ! qualifier.flat ||
language != EShLangFragment && qualifier.storage == EvqVaryingOut && ! qualifier.flat) {
error(line, "must be qualified as 'flat'", getStorageQualifierString(qualifier.storage), TType::getBasicString(publicType.basicType));
return;
}
}
if (language == EShLangVertex && qualifier.storage == EvqVaryingIn &&
(qualifier.isAuxillary() || qualifier.isInterpolation() || qualifier.isMemory() || qualifier.invariant)) {
error(line, "vertex input cannot be further qualified", "", "");
return;
}
}
//
// Merge characteristics of the 'src' qualifier into the 'dst'.
// If there is duplication, issue error messages, unless 'force'
// is specified, which means to just override default settings.
//
void TParseContext::mergeQualifiers(int line, TPublicType& dst, const TPublicType& src, bool force)
{
bool bad = false;
// Storage qualification
if (dst.qualifier.storage == EvqTemporary || dst.qualifier.storage == EvqGlobal)
dst.qualifier.storage = src.qualifier.storage;
else if (dst.qualifier.storage == EvqIn && src.qualifier.storage == EvqOut ||
dst.qualifier.storage == EvqOut && src.qualifier.storage == EvqIn)
dst.qualifier.storage = EvqInOut;
else if (dst.qualifier.storage == EvqIn && src.qualifier.storage == EvqConst ||
dst.qualifier.storage == EvqConst && src.qualifier.storage == EvqIn)
dst.qualifier.storage = EvqConstReadOnly;
else if (src.qualifier.storage != EvqTemporary) {
error(line, "too many storage qualifiers", getStorageQualifierString(src.qualifier.storage), "");
bad = true;
}
// Precision qualifiers
if (! force && src.qualifier.precision != EpqNone && dst.qualifier.precision != EpqNone) {
error(line, "only one precision qualifier allowed", getPrecisionQualifierString(src.qualifier.precision), "");
bad = true;
}
if (dst.qualifier.precision == EpqNone || force && src.qualifier.precision != EpqNone)
dst.qualifier.precision = src.qualifier.precision;
// Layout qualifiers
mergeLayoutQualifiers(line, dst.qualifier, src.qualifier);
// other qualifiers
#define MERGE_SINGLETON(field) bad |= dst.qualifier.field && src.qualifier.field; dst.qualifier.field |= src.qualifier.field;
MERGE_SINGLETON(invariant);
MERGE_SINGLETON(centroid);
MERGE_SINGLETON(smooth);
MERGE_SINGLETON(flat);
MERGE_SINGLETON(nopersp);
MERGE_SINGLETON(patch);
MERGE_SINGLETON(sample);
MERGE_SINGLETON(shared);
MERGE_SINGLETON(coherent);
MERGE_SINGLETON(volatil);
MERGE_SINGLETON(restrict);
MERGE_SINGLETON(readonly);
MERGE_SINGLETON(writeonly);
if (bad)
error(line, "replicated qualifiers", "", "");
}
void TParseContext::setDefaultPrecision(int line, TPublicType& publicType, TPrecisionQualifier qualifier)
{
TBasicType basicType = publicType.basicType;
if (basicType == EbtSampler) {
defaultSamplerPrecision[computeSamplerTypeIndex(publicType.sampler)] = qualifier;
return; // all is well
}
if (basicType == EbtInt || basicType == EbtFloat) {
if (publicType.isScalar()) {
defaultPrecision[basicType] = qualifier;
if (basicType == EbtInt)
defaultPrecision[EbtUint] = qualifier;
return; // all is well
}
}
error(line, "cannot apply precision statement to this type; use 'float', 'int' or a sampler type", TType::getBasicString(basicType), "");
}
// used to flatten the sampler type space into a single dimension
// correlates with the declaration of defaultSamplerPrecision[]
int TParseContext::computeSamplerTypeIndex(TSampler& sampler)
{
int arrayIndex = sampler.arrayed ? 1 : 0;
int shadowIndex = sampler.shadow ? 1 : 0;
return EsdNumDims * (EbtNumTypes * (2 * arrayIndex + shadowIndex) + sampler.type) + sampler.dim;
}
TPrecisionQualifier TParseContext::getDefaultPrecision(TPublicType& publicType)
{
if (publicType.basicType == EbtSampler)
return defaultSamplerPrecision[computeSamplerTypeIndex(publicType.sampler)];
else
return defaultPrecision[publicType.basicType];
}
void TParseContext::precisionQualifierCheck(int line, TPublicType& publicType)
{
if (profile != EEsProfile)
return;
if (publicType.basicType == EbtFloat || publicType.basicType == EbtUint || publicType.basicType == EbtInt || publicType.basicType == EbtSampler) {
if (publicType.qualifier.precision == EpqNone)
error(line, "type requires declaration of default precision qualifier", TType::getBasicString(publicType.basicType), "");
}
}
void TParseContext::parameterSamplerCheck(int line, TStorageQualifier qualifier, const TType& type)
{
if ((qualifier == EvqOut || qualifier == EvqInOut) && type.getBasicType() != EbtStruct && type.getBasicType() == EbtSampler)
error(line, "samplers cannot be output parameters", type.getCompleteTypeString().c_str(), "");
}
bool TParseContext::containsSampler(const TType& type)
{
if (type.getBasicType() == EbtSampler)
return true;
if (type.getBasicType() == EbtStruct) {
TTypeList& structure = *type.getStruct();
for (unsigned int i = 0; i < structure.size(); ++i) {
if (containsSampler(*structure[i].type))
return true;
}
}
return false;
}
bool TParseContext::insertBuiltInArrayAtGlobalLevel()
{
TString *name = NewPoolTString("gl_TexCoord");
TSymbol* symbol = symbolTable.find(*name);
if (! symbol) {
// assume it was not added due to version/profile
return false;
}
TVariable* variable = symbol->getAsVariable();
if (! variable) {
error(0, "INTERNAL ERROR, variable expected", name->c_str(), "");
return true;
}
TVariable* newVariable = new TVariable(name, variable->getType());
if (! symbolTable.insert(*newVariable)) {
delete newVariable;
error(0, "INTERNAL ERROR inserting new symbol", name->c_str(), "");
return true;
}
return false;
}
//
// Do size checking for an array type's size.
//
void TParseContext::arraySizeCheck(int line, TIntermTyped* expr, int& size)
{
TIntermConstantUnion* constant = expr->getAsConstantUnion();
if (constant == 0 || (constant->getBasicType() != EbtInt && constant->getBasicType() != EbtUint)) {
error(line, "array size must be a constant integer expression", "", "");
size = 1;
return;
}
size = constant->getUnionArrayPointer()->getIConst();
if (size <= 0) {
error(line, "array size must be a positive integer", "", "");
size = 1;
return;
}
}
//
// See if this qualifier can be an array.
//
// Returns true if there is an error.
//
bool TParseContext::arrayQualifierError(int line, const TPublicType& type)
{
if (type.qualifier.storage == EvqConst)
profileRequires(line, ENoProfile, 120, "GL_3DL_array_objects", "const array");
if (type.qualifier.storage == EvqVaryingIn && language == EShLangVertex) {
requireProfile(line, (EProfileMask)~EEsProfileMask, "vertex input arrays");
profileRequires(line, ENoProfile, 150, 0, "vertex input arrays");
}
return false;
}
//
// Require array to have size
//
void TParseContext::arraySizeRequiredCheck(int line, int& size)
{
if (size == 0) {
error(line, "array size required", "", "");
size = 1;
}
}
//
// Do all the semantic checking for declaring an array, with and
// without a size, and make the right changes to the symbol table.
//
// size == 0 means no specified size.
//
void TParseContext::arrayCheck(int line, TString& identifier, const TPublicType& type, TVariable*& variable)
{
//
// Don't check for reserved word use until after we know it's not in the symbol table,
// because reserved arrays can be redeclared.
//
bool sameScope = false;
TSymbol* symbol = symbolTable.find(identifier, 0, &sameScope);
if (symbol == 0 || !sameScope) {
if (reservedErrorCheck(line, identifier))
return;
variable = new TVariable(&identifier, TType(type));
if (! symbolTable.insert(*variable)) {
delete variable;
error(line, "INTERNAL ERROR inserting new symbol", identifier.c_str(), "");
return;
}
} else {
variable = symbol->getAsVariable();
if (! variable) {
error(line, "array variable name expected", identifier.c_str(), "");
return;
}
if (! variable->getType().isArray()) {
error(line, "redeclaring non-array as array", identifier.c_str(), "");
return;
}
if (variable->getType().getArraySize() > 0) {
error(line, "redeclaration of array with size", identifier.c_str(), "");
return;
}
if (! variable->getType().sameElementType(TType(type))) {
error(line, "redeclaration of array with a different type", identifier.c_str(), "");
return;
}
TType* t = variable->getArrayInformationType();
while (t != 0) {
if (t->getMaxArraySize() > type.arraySizes->front()) {
error(line, "higher index value already used for the array", identifier.c_str(), "");
return;
}
t->setArraySizes(type.arraySizes);
t = t->getArrayInformationType();
}
variable->getType().setArraySizes(type.arraySizes);
}
voidErrorCheck(line, identifier, type);
}
bool TParseContext::arraySetMaxSize(TIntermSymbol *node, TType* type, int size, bool updateFlag, TSourceLoc line)
{
TSymbol* symbol = symbolTable.find(node->getSymbol());
if (symbol == 0) {
error(line, " undeclared identifier", node->getSymbol().c_str(), "");
return true;
}
TVariable* variable = symbol->getAsVariable();
if (! variable) {
error(0, "array variable name expected", node->getSymbol().c_str(), "");
return true;
}
type->setArrayInformationType(variable->getArrayInformationType());
variable->updateArrayInformationType(type);
// special casing to test index value of gl_TexCoord. If the accessed index is >= gl_MaxTextureCoords
// its an error
if (node->getSymbol() == "gl_TexCoord") {
TSymbol* texCoord = symbolTable.find("gl_MaxTextureCoords");
if (! texCoord || ! texCoord->getAsVariable()) {
infoSink.info.message(EPrefixInternalError, "gl_MaxTextureCoords not defined", line);
return true;
}
int texCoordValue = texCoord->getAsVariable()->getConstUnionPointer()[0].getIConst();
if (texCoordValue <= size) {
error(line, "", "[", "gl_TexCoord can only have a max array size of up to gl_MaxTextureCoords", "");
return true;
}
}
// we dont want to update the maxArraySize when this flag is not set, we just want to include this
// node type in the chain of node types so that its updated when a higher maxArraySize comes in.
if (!updateFlag)
return false;
size++;
variable->getType().setMaxArraySize(size);
type->setMaxArraySize(size);
TType* tt = type;
while(tt->getArrayInformationType() != 0) {
tt = tt->getArrayInformationType();
tt->setMaxArraySize(size);
}
return false;
}
//
// Enforce non-initializer type/qualifier rules.
//
void TParseContext::nonInitConstCheck(int line, TString& identifier, TPublicType& type)
{
//
// Make the qualifier make sense.
//
if (type.qualifier.storage == EvqConst) {
type.qualifier.storage = EvqTemporary;
error(line, "variables with qualifier 'const' must be initialized", identifier.c_str(), "");
}
}
//
// Do semantic checking for a variable declaration that has no initializer,
// and update the symbol table.
//
void TParseContext::nonInitCheck(int line, TString& identifier, TPublicType& type)
{
reservedErrorCheck(line, identifier);
TVariable* variable = new TVariable(&identifier, TType(type));
if (! symbolTable.insert(*variable)) {
error(line, "redefinition", variable->getName().c_str(), "");
delete variable;
} else
voidErrorCheck(line, identifier, type);
}
void TParseContext::paramCheck(int line, TStorageQualifier qualifier, TType* type)
{
switch (qualifier) {
case EvqConst:
case EvqConstReadOnly:
type->getQualifier().storage = EvqConstReadOnly;
break;
case EvqIn:
case EvqOut:
case EvqInOut:
type->getQualifier().storage = qualifier;
break;
case EvqTemporary:
type->getQualifier().storage = EvqIn;
break;
default:
type->getQualifier().storage = EvqIn;
error(line, "qualifier not allowed on function parameter", getStorageQualifierString(qualifier), "");
break;
}
}
void TParseContext::nestedBlockCheck(int line)
{
if (structNestingLevel > 0)
error(line, "cannot nest a block definition inside a structure or block", "", "");
++structNestingLevel;
}
void TParseContext::nestedStructCheck(int line)
{
if (structNestingLevel > 0)
error(line, "cannot nest a structure definition inside a structure or block", "", "");
++structNestingLevel;
}
//
// Layout qualifier stuff.
//
// Put the id's layout qualification into the public type.
void TParseContext::setLayoutQualifier(int line, TPublicType& publicType, TString& id)
{
std::transform(id.begin(), id.end(), id.begin(), ::tolower);
if (id == TQualifier::getLayoutMatrixString(ElmColumnMajor))
publicType.qualifier.layoutMatrix = ElmColumnMajor;
else if (id == TQualifier::getLayoutMatrixString(ElmRowMajor))
publicType.qualifier.layoutMatrix = ElmRowMajor;
else if (id == TQualifier::getLayoutPackingString(ElpPacked))
publicType.qualifier.layoutPacking = ElpPacked;
else if (id == TQualifier::getLayoutPackingString(ElpShared))
publicType.qualifier.layoutPacking = ElpShared;
else if (id == TQualifier::getLayoutPackingString(ElpStd140))
publicType.qualifier.layoutPacking = ElpStd140;
else if (id == TQualifier::getLayoutPackingString(ElpStd430))
publicType.qualifier.layoutPacking = ElpStd430;
else if (id == "location")
error(line, "requires an integer assignment (e.g., location = 4)", "location", "");
else if (id == "binding")
error(line, "requires an integer assignment (e.g., binding = 4)", "binding", "");
else
error(line, "unrecognized layout identifier", id.c_str(), "");
}
// Put the id's layout qualifier value into the public type.
void TParseContext::setLayoutQualifier(int line, TPublicType& publicType, TString& id, int value)
{
std::transform(id.begin(), id.end(), id.begin(), ::tolower);
if (id == "location") {
if ((unsigned int)value >= TQualifier::layoutLocationEnd)
error(line, "value is too large", id.c_str(), "");
else
publicType.qualifier.layoutSlotLocation = value;
} else if (id == "binding")
error(line, "not supported", "binding", "");
else
error(line, "there is no such layout identifier taking an assigned value", id.c_str(), "");
// TODO: semantics: error check: make sure locations are non-overlapping across the whole stage
// TODO: semantics: error check: if more than one fragment output, all must have a location
// TODO: semantics: error check: output arrays can only be indexed with a constant (es 300)
}
// Merge any layout qualifier information from src into dst, leaving everything else in dst alone
void TParseContext::mergeLayoutQualifiers(int line, TQualifier& dst, const TQualifier& src)
{
if (src.layoutMatrix != ElmNone)
dst.layoutMatrix = src.layoutMatrix;
if (src.layoutPacking != ElpNone)
dst.layoutPacking = src.layoutPacking;
if (src.hasLocation())
dst.layoutSlotLocation = src.layoutSlotLocation;
}
/////////////////////////////////////////////////////////////////////////////////
//
// Non-Errors.
//
/////////////////////////////////////////////////////////////////////////////////
//
// Look up a function name in the symbol table, and make sure it is a function.
//
// Return the function symbol if found, otherwise 0.
//
const TFunction* TParseContext::findFunction(int line, TFunction* call, bool *builtIn)
{
TSymbol* symbol = symbolTable.find(call->getMangledName(), builtIn);
if (symbol == 0) {
error(line, "no matching overloaded function found", call->getName().c_str(), "");
return 0;
}
const TFunction* function = symbol->getAsFunction();
if (! function) {
error(line, "function name expected", call->getName().c_str(), "");
return 0;
}
return function;
}
//
// Initializers show up in several places in the grammar. Have one set of
// code to handle them here.
//
bool TParseContext::executeInitializerError(TSourceLoc line, TString& identifier, TPublicType& pType,
TIntermTyped* initializer, TIntermNode*& intermNode, TVariable* variable)
{
TType type(pType);
if (variable == 0) {
if (reservedErrorCheck(line, identifier))
return true;
if (voidErrorCheck(line, identifier, pType))
return true;
//
// add variable to symbol table
//
variable = new TVariable(&identifier, type);
if (! symbolTable.insert(*variable)) {
error(line, "redefinition", variable->getName().c_str(), "");
return true;
// don't delete variable, it's used by error recovery, and the pool
// pop will take care of the memory
}
}
//
// identifier must be of type constant, a global, or a temporary
//
TStorageQualifier qualifier = variable->getType().getQualifier().storage;
if ((qualifier != EvqTemporary) && (qualifier != EvqGlobal) && (qualifier != EvqConst)) {
error(line, " cannot initialize this type of qualifier ", variable->getType().getStorageQualifierString(), "");
return true;
}
// Fix arrayness if variable is unsized, getting size for initializer
if (initializer->getType().isArray() && initializer->getType().getArraySize() > 0 &&
type.isArray() && type.getArraySize() == 0)
type.changeArraySize(initializer->getType().getArraySize());
//
// test for and propagate constant
//
if (qualifier == EvqConst) {
if (qualifier != initializer->getType().getQualifier().storage) {
error(line, " assigning non-constant to", "=", "'%s'", variable->getType().getCompleteString().c_str());
variable->getType().getQualifier().storage = EvqTemporary;
return true;
}
if (type != initializer->getType()) {
error(line, " non-matching types for const initializer ",
variable->getType().getStorageQualifierString(), "");
variable->getType().getQualifier().storage = EvqTemporary;
return true;
}
if (initializer->getAsConstantUnion()) {
constUnion* unionArray = variable->getConstUnionPointer();
if (type.getObjectSize() == 1 && type.getBasicType() != EbtStruct) {
*unionArray = (initializer->getAsConstantUnion()->getUnionArrayPointer())[0];
} else {
variable->shareConstPointer(initializer->getAsConstantUnion()->getUnionArrayPointer());
}
} else if (initializer->getAsSymbolNode()) {
TSymbol* symbol = symbolTable.find(initializer->getAsSymbolNode()->getSymbol());
if (TVariable* tVar = symbol->getAsVariable()) {
constUnion* constArray = tVar->getConstUnionPointer();
variable->shareConstPointer(constArray);
} else {
error(line, "expected variable", initializer->getAsSymbolNode()->getSymbol().c_str(), "");
return true;
}
} else {
error(line, " cannot assign to", "=", "'%s'", variable->getType().getCompleteString().c_str());
variable->getType().getQualifier().storage = EvqTemporary;
return true;
}
}
if (qualifier != EvqConst) {
TIntermSymbol* intermSymbol = intermediate.addSymbol(variable->getUniqueId(), variable->getName(), variable->getType(), line);
intermNode = intermediate.addAssign(EOpAssign, intermSymbol, initializer, line);
if (intermNode == 0) {
assignError(line, "=", intermSymbol->getCompleteString(), initializer->getCompleteString());
return true;
}
} else
intermNode = 0;
return false;
}
// This function is used to test for the correctness of the parameters passed to various constructor functions
// and also convert them to the right datatype if it is allowed and required.
//
// Returns 0 for an error or the constructed node (aggregate or typed) for no error.
//
TIntermTyped* TParseContext::addConstructor(TIntermNode* node, const TType& type, TOperator op, TFunction* fnCall, TSourceLoc line)
{
if (node == 0)
return 0;
TIntermAggregate* aggrNode = node->getAsAggregate();
TTypeList::iterator memberTypes;
if (op == EOpConstructStruct)
memberTypes = type.getStruct()->begin();
TType elementType(type);
if (type.isArray())
elementType.dereference();
bool singleArg;
if (aggrNode) {
if (aggrNode->getOp() != EOpNull || aggrNode->getSequence().size() == 1)
singleArg = true;
else
singleArg = false;
} else
singleArg = true;
TIntermTyped *newNode;
if (singleArg) {
// If structure constructor or array constructor is being called
// for only one parameter inside the structure, we need to call constructStruct function once.
if (type.isArray())
newNode = constructStruct(node, elementType, 1, node->getLine());
else if (op == EOpConstructStruct)
newNode = constructStruct(node, *(*memberTypes).type, 1, node->getLine());
else
newNode = constructBuiltIn(type, op, node, node->getLine(), false);
if (newNode && (type.isArray() || op == EOpConstructStruct))
newNode = intermediate.setAggregateOperator(newNode, EOpConstructStruct, type, line);
return newNode;
}
//
// Handle list of arguments.
//
TIntermSequence &sequenceVector = aggrNode->getSequence() ; // Stores the information about the parameter to the constructor
// if the structure constructor contains more than one parameter, then construct
// each parameter
int paramCount = 0; // keeps a track of the constructor parameter number being checked
// for each parameter to the constructor call, check to see if the right type is passed or convert them
// to the right type if possible (and allowed).
// for structure constructors, just check if the right type is passed, no conversion is allowed.
for (TIntermSequence::iterator p = sequenceVector.begin();
p != sequenceVector.end(); p++, paramCount++) {
if (type.isArray())
newNode = constructStruct(*p, elementType, paramCount+1, node->getLine());
else if (op == EOpConstructStruct)
newNode = constructStruct(*p, *(memberTypes[paramCount]).type, paramCount+1, node->getLine());
else
newNode = constructBuiltIn(type, op, *p, node->getLine(), true);
if (newNode) {
p = sequenceVector.erase(p);
p = sequenceVector.insert(p, newNode);
}
}
TIntermTyped* constructor = intermediate.setAggregateOperator(aggrNode, op, type, line);
return constructor;
}
// Function for constructor implementation. Calls addUnaryMath with appropriate EOp value
// for the parameter to the constructor (passed to this function). Essentially, it converts
// the parameter types correctly. If a constructor expects an int (like ivec2) and is passed a
// float, then float is converted to int.
//
// Returns 0 for an error or the constructed node.
//
TIntermTyped* TParseContext::constructBuiltIn(const TType& type, TOperator op, TIntermNode* node, TSourceLoc line, bool subset)
{
TIntermTyped* newNode;
TOperator basicOp;
//
// First, convert types as needed.
//
switch (op) {
case EOpConstructVec2:
case EOpConstructVec3:
case EOpConstructVec4:
case EOpConstructMat2x2:
case EOpConstructMat2x3:
case EOpConstructMat2x4:
case EOpConstructMat3x2:
case EOpConstructMat3x3:
case EOpConstructMat3x4:
case EOpConstructMat4x2:
case EOpConstructMat4x3:
case EOpConstructMat4x4:
case EOpConstructFloat:
basicOp = EOpConstructFloat;
break;
case EOpConstructDVec2:
case EOpConstructDVec3:
case EOpConstructDVec4:
case EOpConstructDMat2x2:
case EOpConstructDMat2x3:
case EOpConstructDMat2x4:
case EOpConstructDMat3x2:
case EOpConstructDMat3x3:
case EOpConstructDMat3x4:
case EOpConstructDMat4x2:
case EOpConstructDMat4x3:
case EOpConstructDMat4x4:
case EOpConstructDouble:
basicOp = EOpConstructDouble;
break;
case EOpConstructIVec2:
case EOpConstructIVec3:
case EOpConstructIVec4:
case EOpConstructInt:
basicOp = EOpConstructInt;
break;
case EOpConstructUVec2:
case EOpConstructUVec3:
case EOpConstructUVec4:
case EOpConstructUint:
basicOp = EOpConstructUint;
break;
case EOpConstructBVec2:
case EOpConstructBVec3:
case EOpConstructBVec4:
case EOpConstructBool:
basicOp = EOpConstructBool;
break;
default:
error(line, "unsupported construction", "", "");
return 0;
}
newNode = intermediate.addUnaryMath(basicOp, node, node->getLine());
if (newNode == 0) {
error(line, "can't convert", "constructor", "");
return 0;
}
//
// Now, if there still isn't an operation to do the construction, and we need one, add one.
//
// Otherwise, skip out early.
if (subset || newNode != node && newNode->getType() == type)
return newNode;
// setAggregateOperator will insert a new node for the constructor, as needed.
return intermediate.setAggregateOperator(newNode, op, type, line);
}
// This function tests for the type of the parameters to the structures constructors. Raises
// an error message if the expected type does not match the parameter passed to the constructor.
//
// Returns 0 for an error or the input node itself if the expected and the given parameter types match.
//
TIntermTyped* TParseContext::constructStruct(TIntermNode* node, const TType& type, int paramCount, TSourceLoc line)
{
TIntermTyped* converted = intermediate.addConversion(EOpConstructStruct, type, node->getAsTyped());
if (! converted || converted->getType() != type) {
error(line, "", "constructor", "cannot convert parameter %d from '%s' to '%s'", paramCount,
node->getAsTyped()->getType().getCompleteTypeString().c_str(), type.getCompleteTypeString().c_str());
return 0;
}
return converted;
}
//
// Do everything needed to add an interface block.
//
void TParseContext::addBlock(int line, TTypeList& typeList, const TString* instanceName, TArraySizes arraySizes)
{
// First, error checks
if (reservedErrorCheck(line, *blockName))
return;
if (instanceName && reservedErrorCheck(line, *instanceName))
return;
if (profile == EEsProfile && arraySizes)
arraySizeRequiredCheck(line, arraySizes->front());
if (publicBlockType.basicType != EbtVoid) {
error(line, "interface blocks cannot be declared with a type", blockName->c_str(), "");
return;
}
if (publicBlockType.qualifier.storage == EvqUniform) {
requireProfile(line, (EProfileMask)(~ENoProfileMask), "uniform block");
profileRequires(line, EEsProfile, 300, 0, "uniform block");
} else {
error(line, "only uniform interface blocks are supported", blockName->c_str(), "");
return;
}
// check for qualifiers and types that don't belong within a block
for (unsigned int member = 0; member < typeList.size(); ++member) {
TQualifier memberQualifier = typeList[member].type->getQualifier();
if (memberQualifier.storage != EvqTemporary && memberQualifier.storage != EvqGlobal && memberQualifier.storage != publicBlockType.qualifier.storage)
error(line, "member storage qualifier cannot contradict block storage qualifier", typeList[member].type->getFieldName().c_str(), "");
if (publicBlockType.qualifier.storage == EvqUniform && memberQualifier.isInterpolation() || memberQualifier.isAuxillary())
error(line, "member of uniform block cannot have an auxillary or interpolation qualifier", typeList[member].type->getFieldName().c_str(), "");
TBasicType basicType = typeList[member].type->getBasicType();
if (basicType == EbtSampler)
error(line, "member of block cannot be a sampler type", typeList[member].type->getFieldName().c_str(), "");
}
// Make default block qualification, and adjust the member qualifications
TQualifier defaultQualification = defaultGlobalQualification;
mergeLayoutQualifiers(line, defaultQualification, publicBlockType.qualifier);
for (unsigned int member = 0; member < typeList.size(); ++member) {
TQualifier memberQualification = defaultQualification;
mergeLayoutQualifiers(line, memberQualification, typeList[member].type->getQualifier());
typeList[member].type->getQualifier() = memberQualification;
}
// Build and add the interface block as a new type named blockName
TType blockType(&typeList, *blockName, publicBlockType.qualifier.storage);
if (arraySizes)
blockType.setArraySizes(arraySizes);
blockType.getQualifier().layoutPacking = defaultQualification.layoutPacking;
TVariable* userTypeDef = new TVariable(blockName, blockType, true);
if (! symbolTable.insert(*userTypeDef)) {
error(line, "redefinition", blockName->c_str(), "block name");
return;
}
// Add the variable, as anonymous or named instanceName
// make an anonymous variable if no name was provided
if (! instanceName)
instanceName = new TString("");
TVariable* variable = new TVariable(instanceName, blockType);
if (! symbolTable.insert(*variable)) {
if (*instanceName == "")
error(line, "nameless block contains a member that already has a name at global scope", blockName->c_str(), "");
else
error(line, "block instance name redefinition", variable->getName().c_str(), "");
return;
}
}
//
// Take the sequence of statements that has been built up since the last case/default,
// put it on the list of top-level nodes for the current (inner-most) switch statement,
// and follow that by the case/default we are on now. (See switch topology comment on
// TIntermSwitch.)
//
void TParseContext::wrapupSwitchSubsequence(TIntermAggregate* statements, TIntermNode* branchNode)
{
TIntermSequence* switchSequence = switchSequenceStack.back();
if (statements) {
if (switchSequence->size() == 0)
error(statements->getLine(), "cannot have statements before first case/default label", "switch", "");
statements->setOperator(EOpSequence);
switchSequence->push_back(statements);
}
if (branchNode)
switchSequence->push_back(branchNode);
// TODO: semantics: verify no duplicated case values
}
//
// Turn the top-level node sequence built up of wrapupSwitchSubsequence9)
// into a switch node.
//
TIntermNode* TParseContext::addSwitch(int line, TIntermTyped* expression, TIntermAggregate* lastStatements)
{
profileRequires(line, EEsProfile, 300, 0, "switch statements");
profileRequires(line, ENoProfile, 130, 0, "switch statements");
wrapupSwitchSubsequence(lastStatements, 0);
if (expression == 0 ||
expression->getBasicType() != EbtInt && expression->getBasicType() != EbtUint ||
expression->getType().isArray() || expression->getType().isMatrix() || expression->getType().isVector())
error(line, "condition must be a scalar integer expression", "switch", "");
// If there is nothing to do, drop the switch but still execute the expression
TIntermSequence* switchSequence = switchSequenceStack.back();
if (switchSequence->size() == 0)
return expression;
if (lastStatements == 0) {
error(line, "last case/default label must be followed by statements", "switch", "");
return expression;
}
TIntermAggregate* body = new TIntermAggregate(EOpSequence);
body->getSequence() = *switchSequenceStack.back();
body->setLine(line);
TIntermSwitch* switchNode = new TIntermSwitch(expression, body);
switchNode->setLine(line);
return switchNode;
}
void TParseContext::updateDefaults(int line, const TPublicType& publicType, const TString* id)
{
bool cantHaveId = false;
TQualifier qualifier = publicType.qualifier;
if (qualifier.storage == EvqUniform) {
if (qualifier.layoutMatrix != ElmNone) {
cantHaveId = true;
defaultGlobalQualification.layoutMatrix = qualifier.layoutMatrix;
}
if (qualifier.layoutPacking != ElpNone) {
cantHaveId = true;
defaultGlobalQualification.layoutPacking = qualifier.layoutPacking;
}
} else if (qualifier.storage == EvqVaryingIn) {
if (qualifier.hasLayout() && language != EShLangVertex) {
error(line, "can only use location layout qualifier on a vertex input or fragment output", id->c_str(), "");
}
} else if (qualifier.storage == EvqVaryingOut) {
if (qualifier.hasLayout() && language != EShLangFragment) {
error(line, "can only use location layout qualifier on a vertex input or fragment output", id->c_str(), "");
}
}
if (cantHaveId && id)
error(line, "cannot set global layout qualifiers on uniform variable, use just 'uniform' or a block", id->c_str(), "");
}
//
// This function returns the tree representation for the vector field(s) being accessed from contant vector.
// If only one component of vector is accessed (v.x or v[0] where v is a contant vector), then a contant node is
// returned, else an aggregate node is returned (for v.xy). The input to this function could either be the symbol
// node or it could be the intermediate tree representation of accessing fields in a constant structure or column of
// a constant matrix.
//
TIntermTyped* TParseContext::addConstVectorNode(TVectorFields& fields, TIntermTyped* node, TSourceLoc line)
{
TIntermTyped* typedNode;
TIntermConstantUnion* tempConstantNode = node->getAsConstantUnion();
constUnion *unionArray;
if (tempConstantNode) {
unionArray = tempConstantNode->getUnionArrayPointer();
if (!unionArray) { // this error message should never be raised
infoSink.info.message(EPrefixInternalError, "constUnion not initialized in addConstVectorNode function", line);
return node;
}
} else { // The node has to be either a symbol node or an aggregate node or a tempConstant node, else, its an error
error(line, "Cannot offset into the vector", "Error", "");
return 0;
}
constUnion* constArray = new constUnion[fields.num];
for (int i = 0; i < fields.num; i++) {
if (fields.offsets[i] >= node->getType().getObjectSize()) {
error(line, "", "[", "vector index out of range '%d'", fields.offsets[i]);
fields.offsets[i] = 0;
}
constArray[i] = unionArray[fields.offsets[i]];
}
typedNode = intermediate.addConstantUnion(constArray, node->getType(), line);
return typedNode;
}
//
// This function returns the column being accessed from a constant matrix. The values are retrieved from
// the symbol table and parse-tree is built for a vector (each column of a matrix is a vector). The input
// to the function could either be a symbol node (m[0] where m is a constant matrix)that represents a
// constant matrix or it could be the tree representation of the constant matrix (s.m1[0] where s is a constant structure)
//
TIntermTyped* TParseContext::addConstMatrixNode(int index, TIntermTyped* node, TSourceLoc line)
{
TIntermTyped* typedNode;
TIntermConstantUnion* tempConstantNode = node->getAsConstantUnion();
if (index >= node->getType().getMatrixCols()) {
error(line, "", "[", "matrix field selection out of range '%d'", index);
index = 0;
}
if (tempConstantNode) {
constUnion* unionArray = tempConstantNode->getUnionArrayPointer();
int size = tempConstantNode->getType().getMatrixRows();
// Note: the type is corrected (dereferenced) by the caller
typedNode = intermediate.addConstantUnion(&unionArray[size*index], tempConstantNode->getType(), line);
} else {
error(line, "Cannot offset into the matrix", "Error", "");
return 0;
}
return typedNode;
}
//
// This function returns an element of an array accessed from a constant array. The values are retrieved from
// the symbol table and parse-tree is built for the type of the element. The input
// to the function could either be a symbol node (a[0] where a is a constant array)that represents a
// constant array or it could be the tree representation of the constant array (s.a1[0] where s is a constant structure)
//
TIntermTyped* TParseContext::addConstArrayNode(int index, TIntermTyped* node, TSourceLoc line)
{
TIntermTyped* typedNode;
TIntermConstantUnion* tempConstantNode = node->getAsConstantUnion();
int arraySize = node->getType().getArraySize();
TType arrayElementType(node->getType());
arrayElementType.dereference();
if (index >= node->getType().getArraySize() || index < 0) {
error(line, "", "[", "array index '%d' out of range", index);
index = 0;
}
int arrayElementSize = arrayElementType.getObjectSize();
if (tempConstantNode) {
constUnion* unionArray = tempConstantNode->getUnionArrayPointer();
typedNode = intermediate.addConstantUnion(&unionArray[arrayElementSize * index], tempConstantNode->getType(), line);
} else {
error(line, "Cannot offset into the array", "Error", "");
return 0;
}
return typedNode;
}
//
// This function returns the value of a particular field inside a constant structure from the symbol table.
// If there is an embedded/nested struct, it appropriately calls addConstStructNested or addConstStructFromAggr
// function and returns the parse-tree with the values of the embedded/nested struct.
//
TIntermTyped* TParseContext::addConstStruct(TString& identifier, TIntermTyped* node, TSourceLoc line)
{
TTypeList* fields = node->getType().getStruct();
TIntermTyped *typedNode;
int instanceSize = 0;
unsigned int index = 0;
TIntermConstantUnion *tempConstantNode = node->getAsConstantUnion();
for ( index = 0; index < fields->size(); ++index) {
if ((*fields)[index].type->getFieldName() == identifier) {
break;
} else {
instanceSize += (*fields)[index].type->getObjectSize();
}
}
if (tempConstantNode) {
constUnion* constArray = tempConstantNode->getUnionArrayPointer();
typedNode = intermediate.addConstantUnion(constArray+instanceSize, tempConstantNode->getType(), line); // type will be changed in the calling function
} else {
error(line, "Cannot offset into the structure", "Error", "");
return 0;
}
return typedNode;
}
//
// Initialize all supported extensions to disable
//
void TParseContext::initializeExtensionBehavior()
{
//
// example code: extensionBehavior["test"] = EBhDisable; // where "test" is the name of
// supported extension
//
extensionBehavior["GL_ARB_texture_rectangle"] = EBhDisable;
extensionBehavior["GL_3DL_array_objects"] = EBhDisable;
}
OS_TLSIndex GlobalParseContextIndex = OS_INVALID_TLS_INDEX;
bool InitializeParseContextIndex()
{
if (GlobalParseContextIndex != OS_INVALID_TLS_INDEX) {
assert(0 && "InitializeParseContextIndex(): Parse Context already initialised");
return false;
}
//
// Allocate a TLS index.
//
GlobalParseContextIndex = OS_AllocTLSIndex();
if (GlobalParseContextIndex == OS_INVALID_TLS_INDEX) {
assert(0 && "InitializeParseContextIndex(): Parse Context already initialised");
return false;
}
return true;
}
bool InitializeGlobalParseContext()
{
if (GlobalParseContextIndex == OS_INVALID_TLS_INDEX) {
assert(0 && "InitializeGlobalParseContext(): Parse Context index not initialized");
return false;
}
TThreadParseContext *lpParseContext = static_cast<TThreadParseContext *>(OS_GetTLSValue(GlobalParseContextIndex));
if (lpParseContext != 0) {
assert(0 && "InitializeParseContextIndex(): Parse Context already initialized");
return false;
}
TThreadParseContext *lpThreadData = new TThreadParseContext();
if (lpThreadData == 0) {
assert(0 && "InitializeGlobalParseContext(): Unable to create thread parse context");
return false;
}
lpThreadData->lpGlobalParseContext = 0;
OS_SetTLSValue(GlobalParseContextIndex, lpThreadData);
return true;
}
TParseContextPointer& GetGlobalParseContext()
{
//
// Minimal error checking for speed
//
TThreadParseContext *lpParseContext = static_cast<TThreadParseContext *>(OS_GetTLSValue(GlobalParseContextIndex));
return lpParseContext->lpGlobalParseContext;
}
bool FreeParseContext()
{
if (GlobalParseContextIndex == OS_INVALID_TLS_INDEX) {
assert(0 && "FreeParseContext(): Parse Context index not initialized");
return false;
}
TThreadParseContext *lpParseContext = static_cast<TThreadParseContext *>(OS_GetTLSValue(GlobalParseContextIndex));
if (lpParseContext)
delete lpParseContext;
return true;
}
bool FreeParseContextIndex()
{
OS_TLSIndex tlsiIndex = GlobalParseContextIndex;
if (GlobalParseContextIndex == OS_INVALID_TLS_INDEX) {
assert(0 && "FreeParseContextIndex(): Parse Context index not initialized");
return false;
}
GlobalParseContextIndex = OS_INVALID_TLS_INDEX;
return OS_FreeTLSIndex(tlsiIndex);
}