mirror of
https://develop.openfoam.com/Development/openfoam.git
synced 2025-11-28 03:28:01 +00:00
Add preliminary wmake rule for using CoCo/R.
This commit is contained in:
Mark Olesen
@ -3,13 +3,8 @@ cd ${0%/*} || exit 1 # run from this directory
|
||||
|
||||
# this will have to do until we have a makefile rule
|
||||
|
||||
if type Coco > /dev/null 2>&1
|
||||
then
|
||||
Coco \
|
||||
-frames $WM_THIRD_PARTY_DIR/coco-r \
|
||||
calcEntry/calcEntry.atg
|
||||
else
|
||||
echo "Coco not installed"
|
||||
fi
|
||||
type Coco > /dev/null 2>&1 || {
|
||||
echo "Coco not installed -- cannot make this test"
|
||||
}
|
||||
|
||||
wmake
|
||||
|
||||
@ -1,8 +1,7 @@
|
||||
dictionaryTest.C
|
||||
|
||||
calcEntry/calcEntry.C
|
||||
calcEntry/calcEntryParser.ATG
|
||||
calcEntry/calcEntryInternal.C
|
||||
calcEntry/calcEntryParser.cpp
|
||||
calcEntry/calcEntryScanner.cpp
|
||||
calcEntry/calcEntry.C
|
||||
|
||||
EXE = $(FOAM_USER_APPBIN)/dictionaryTest
|
||||
|
||||
@ -0,0 +1,2 @@
|
||||
EXE_INC = -IcalcEntry -I$(OBJECTS_DIR)
|
||||
|
||||
|
||||
@ -22,7 +22,7 @@ License
|
||||
along with OpenFOAM; if not, write to the Free Software Foundation,
|
||||
Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
|
||||
|
||||
@file calcEntry.atg
|
||||
@file calcEntryParser.ATG
|
||||
|
||||
Description
|
||||
An attributed Coco/R grammar to parse simple arithmetic expressions
|
||||
@ -34,8 +34,8 @@ SourceFiles
|
||||
/*---------------------------------------------------------------------------*\
|
||||
compile with:
|
||||
Coco \
|
||||
-frames $WM_THIRD_PARTY_DIR/coco-r \
|
||||
calcEntry.atg
|
||||
[-single] -frames $WM_THIRD_PARTY_DIR/coco-r \
|
||||
calcEntryParser.ATG
|
||||
\*---------------------------------------------------------------------------*/
|
||||
|
||||
#include "dictionary.H"
|
||||
@ -61,8 +61,6 @@ private:
|
||||
//- lookup dictionary entry
|
||||
scalar getDictLookup() const
|
||||
{
|
||||
scalar dictValue = 0;
|
||||
|
||||
if (!dict_)
|
||||
{
|
||||
FatalErrorIn
|
||||
@ -83,6 +81,8 @@ private:
|
||||
word keyword(str);
|
||||
coco_string_delete(str);
|
||||
|
||||
scalar dictValue = 0;
|
||||
|
||||
entry* entryPtr = dict_->lookupEntryPtr(keyword, true, false);
|
||||
if (entryPtr && !entryPtr->isDict())
|
||||
{
|
||||
@ -237,15 +237,15 @@ Func<scalar& val>
|
||||
char* str = coco_string_create_char(t->val);
|
||||
word funcName(str);
|
||||
coco_string_delete(str);
|
||||
DynamicList<scalar> param(4); // hold parameter values
|
||||
DynamicList<scalar> stack(4);
|
||||
.)
|
||||
'('
|
||||
[ (. scalar x; .)
|
||||
Expr<x> (. param.append(x); .)
|
||||
{ ',' Expr<x> (. param.append(x); .)
|
||||
Expr<x> (. stack.append(x); .)
|
||||
{ ',' Expr<x> (. stack.append(x); .)
|
||||
}
|
||||
]
|
||||
')' (. val = dispatch(funcName, param); .)
|
||||
')' (. val = dispatch(funcName, stack); .)
|
||||
.
|
||||
|
||||
|
||||
@ -1,340 +0,0 @@
|
||||
|
||||
|
||||
#include <wchar.h>
|
||||
#include "calcEntryParser.h"
|
||||
|
||||
|
||||
namespace Foam {
|
||||
namespace functionEntries {
|
||||
namespace calcEntryInternal {
|
||||
|
||||
|
||||
// ----------------------------------------------------------------------------
|
||||
// Parser Implementation
|
||||
// ----------------------------------------------------------------------------
|
||||
|
||||
void Parser::SynErr(int n) {
|
||||
if (errDist >= minErrDist) errors->SynErr(la->line, la->col, n);
|
||||
errDist = 0;
|
||||
}
|
||||
|
||||
|
||||
void Parser::SemErr(const wchar_t* msg) {
|
||||
if (errDist >= minErrDist) errors->Error(t->line, t->col, msg);
|
||||
errDist = 0;
|
||||
}
|
||||
|
||||
|
||||
void Parser::Get() {
|
||||
for (;;) {
|
||||
t = la;
|
||||
la = scanner->Scan();
|
||||
if (la->kind <= maxT) {
|
||||
++errDist;
|
||||
break;
|
||||
}
|
||||
|
||||
if (dummyToken != t) {
|
||||
dummyToken->kind = t->kind;
|
||||
dummyToken->pos = t->pos;
|
||||
dummyToken->col = t->col;
|
||||
dummyToken->line = t->line;
|
||||
dummyToken->next = NULL;
|
||||
coco_string_delete(dummyToken->val);
|
||||
dummyToken->val = coco_string_create(t->val);
|
||||
t = dummyToken;
|
||||
}
|
||||
la = t;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
void Parser::Expect(int n) {
|
||||
if (la->kind == n) {
|
||||
Get();
|
||||
}
|
||||
else {
|
||||
SynErr(n);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
void Parser::ExpectWeak(int n, int follow) {
|
||||
if (la->kind == n) {
|
||||
Get();
|
||||
}
|
||||
else {
|
||||
SynErr(n);
|
||||
while (!StartOf(follow)) {
|
||||
Get();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
bool Parser::WeakSeparator(int n, int syFol, int repFol) {
|
||||
if (la->kind == n) {
|
||||
Get();
|
||||
return true;
|
||||
}
|
||||
else if (StartOf(repFol)) {
|
||||
return false;
|
||||
}
|
||||
else {
|
||||
SynErr(n);
|
||||
while (!(StartOf(syFol) || StartOf(repFol) || StartOf(0))) {
|
||||
Get();
|
||||
}
|
||||
return StartOf(syFol);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
void Parser::calcEntry() {
|
||||
val = 0;
|
||||
if (la->kind == 5) {
|
||||
Get();
|
||||
Expr(val);
|
||||
Expect(6);
|
||||
scanner->buffer->SetPos(t->pos + 1);
|
||||
|
||||
} else if (StartOf(1)) {
|
||||
Expr(val);
|
||||
Expect(0);
|
||||
} else SynErr(15);
|
||||
}
|
||||
|
||||
void Parser::Expr(scalar& val) {
|
||||
scalar val2 = 0;
|
||||
Term(val);
|
||||
while (la->kind == 7 || la->kind == 8) {
|
||||
if (la->kind == 7) {
|
||||
Get();
|
||||
Term(val2);
|
||||
val += val2;
|
||||
} else {
|
||||
Get();
|
||||
Term(val2);
|
||||
val -= val2;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void Parser::Term(scalar& val) {
|
||||
scalar val2 = 0;
|
||||
Factor(val);
|
||||
while (la->kind == 9 || la->kind == 10) {
|
||||
if (la->kind == 9) {
|
||||
Get();
|
||||
Factor(val2);
|
||||
val *= val2;
|
||||
} else {
|
||||
Get();
|
||||
Factor(val2);
|
||||
val /= val2;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void Parser::Factor(scalar& val) {
|
||||
bool negative = false;
|
||||
if (la->kind == 7 || la->kind == 8) {
|
||||
if (la->kind == 7) {
|
||||
Get();
|
||||
} else {
|
||||
Get();
|
||||
negative = true;
|
||||
}
|
||||
}
|
||||
if (la->kind == 1) {
|
||||
Func(val);
|
||||
} else if (la->kind == 11) {
|
||||
Get();
|
||||
Expr(val);
|
||||
Expect(12);
|
||||
} else if (la->kind == 3) {
|
||||
Get();
|
||||
val = getDictLookup();
|
||||
} else if (la->kind == 4) {
|
||||
Get();
|
||||
val = coco_string_toDouble(t->val);
|
||||
} else SynErr(16);
|
||||
if (negative) { val = -val; }
|
||||
}
|
||||
|
||||
void Parser::Func(scalar& val) {
|
||||
Expect(1);
|
||||
char* str = coco_string_create_char(t->val);
|
||||
word funcName(str);
|
||||
coco_string_delete(str);
|
||||
DynamicList<scalar> param(4); // hold parameter values
|
||||
|
||||
Expect(11);
|
||||
if (StartOf(1)) {
|
||||
scalar x;
|
||||
Expr(x);
|
||||
param.append(x);
|
||||
while (la->kind == 13) {
|
||||
Get();
|
||||
Expr(x);
|
||||
param.append(x);
|
||||
}
|
||||
}
|
||||
Expect(12);
|
||||
val = dispatch(funcName, param);
|
||||
}
|
||||
|
||||
|
||||
|
||||
void Parser::Parse() {
|
||||
t = NULL;
|
||||
if (dummyToken) { // safety: someone might call Parse() twice
|
||||
delete dummyToken;
|
||||
}
|
||||
la = dummyToken = new Token();
|
||||
la->val = coco_string_create(L"Dummy Token");
|
||||
Get();
|
||||
calcEntry();
|
||||
// let grammar deal with end-of-file expectations
|
||||
|
||||
}
|
||||
|
||||
|
||||
Parser::Parser(Scanner* scan, Errors* err)
|
||||
:
|
||||
dummyToken(NULL),
|
||||
deleteErrorsDestruct_(!err),
|
||||
errDist(minErrDist),
|
||||
scanner(scan),
|
||||
errors(err),
|
||||
t(NULL),
|
||||
la(NULL)
|
||||
{
|
||||
if (!errors) { // add in default error handling
|
||||
errors = new Errors();
|
||||
}
|
||||
// user-defined initialization:
|
||||
dict_ = 0;
|
||||
val = 0;
|
||||
|
||||
/*---------------------------------------------------------------------------*/
|
||||
|
||||
|
||||
}
|
||||
|
||||
|
||||
bool Parser::StartOf(int s) {
|
||||
const bool T = true;
|
||||
const bool x = false;
|
||||
|
||||
static const bool set[2][16] = {
|
||||
{T,x,x,x, x,x,x,x, x,x,x,x, x,x,x,x},
|
||||
{x,T,x,T, T,x,x,T, T,x,x,T, x,x,x,x}
|
||||
};
|
||||
|
||||
|
||||
|
||||
return set[s][la->kind];
|
||||
}
|
||||
|
||||
|
||||
Parser::~Parser() {
|
||||
if (deleteErrorsDestruct_) { // delete default error handling
|
||||
delete errors;
|
||||
}
|
||||
delete dummyToken;
|
||||
// user-defined destruction:
|
||||
|
||||
}
|
||||
|
||||
|
||||
// ----------------------------------------------------------------------------
|
||||
// Errors Implementation
|
||||
// ----------------------------------------------------------------------------
|
||||
|
||||
Errors::Errors()
|
||||
:
|
||||
count(0)
|
||||
{}
|
||||
|
||||
|
||||
Errors::~Errors()
|
||||
{}
|
||||
|
||||
|
||||
void Errors::clear() {
|
||||
count = 0;
|
||||
}
|
||||
|
||||
|
||||
wchar_t* Errors::strerror(int n)
|
||||
{
|
||||
wchar_t* s;
|
||||
switch (n) {
|
||||
case 0: s = coco_string_create(L"EOF expected"); break;
|
||||
case 1: s = coco_string_create(L"ident expected"); break;
|
||||
case 2: s = coco_string_create(L"string expected"); break;
|
||||
case 3: s = coco_string_create(L"variable expected"); break;
|
||||
case 4: s = coco_string_create(L"number expected"); break;
|
||||
case 5: s = coco_string_create(L"\"{\" expected"); break;
|
||||
case 6: s = coco_string_create(L"\"}\" expected"); break;
|
||||
case 7: s = coco_string_create(L"\"+\" expected"); break;
|
||||
case 8: s = coco_string_create(L"\"-\" expected"); break;
|
||||
case 9: s = coco_string_create(L"\"*\" expected"); break;
|
||||
case 10: s = coco_string_create(L"\"/\" expected"); break;
|
||||
case 11: s = coco_string_create(L"\"(\" expected"); break;
|
||||
case 12: s = coco_string_create(L"\")\" expected"); break;
|
||||
case 13: s = coco_string_create(L"\",\" expected"); break;
|
||||
case 14: s = coco_string_create(L"??? expected"); break;
|
||||
case 15: s = coco_string_create(L"invalid calcEntry"); break;
|
||||
case 16: s = coco_string_create(L"invalid Factor"); break;
|
||||
|
||||
default:
|
||||
{
|
||||
wchar_t format[20];
|
||||
coco_swprintf(format, 20, L"error %d", n);
|
||||
s = coco_string_create(format);
|
||||
}
|
||||
break;
|
||||
}
|
||||
return s;
|
||||
}
|
||||
|
||||
|
||||
void Errors::Warning(const wchar_t* msg) {
|
||||
wprintf(L"%ls\n", msg);
|
||||
}
|
||||
|
||||
|
||||
void Errors::Warning(int line, int col, const wchar_t* msg) {
|
||||
wprintf(L"-- line %d col %d: %ls\n", line, col, msg);
|
||||
}
|
||||
|
||||
|
||||
void Errors::Error(int line, int col, const wchar_t* msg) {
|
||||
wprintf(L"-- line %d col %d: %ls\n", line, col, msg);
|
||||
count++;
|
||||
}
|
||||
|
||||
|
||||
void Errors::SynErr(int line, int col, int n) {
|
||||
wchar_t* msg = this->strerror(n);
|
||||
this->Error(line, col, msg);
|
||||
coco_string_delete(msg);
|
||||
}
|
||||
|
||||
|
||||
void Errors::Exception(const wchar_t* msg) {
|
||||
wprintf(L"%ls", msg);
|
||||
::exit(1);
|
||||
}
|
||||
|
||||
|
||||
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
|
||||
|
||||
} // namespace
|
||||
} // namespace
|
||||
} // namespace
|
||||
|
||||
|
||||
// ************************************************************************* //
|
||||
@ -1,180 +0,0 @@
|
||||
|
||||
|
||||
#ifndef COCO_calcEntryPARSER_H__
|
||||
#define COCO_calcEntryPARSER_H__
|
||||
|
||||
#include "dictionary.H"
|
||||
#include "wchar.H"
|
||||
#include "calcEntryInternal.H"
|
||||
|
||||
|
||||
#include "calcEntryScanner.h"
|
||||
|
||||
namespace Foam {
|
||||
namespace functionEntries {
|
||||
namespace calcEntryInternal {
|
||||
|
||||
|
||||
//! Parser error handing
|
||||
class Errors {
|
||||
public:
|
||||
int count; //!< The number of errors detected
|
||||
|
||||
//! Allocate and return a string describing the given error code.
|
||||
/** It is the responsibility of the caller to free this string,
|
||||
* eg, with coco_string_delete()
|
||||
*/
|
||||
static wchar_t* strerror(int n);
|
||||
|
||||
Errors(); //!< Construct null - start with no errors
|
||||
virtual ~Errors(); //!< Destructor
|
||||
virtual void clear(); //!< Clear the error count
|
||||
|
||||
//! Handle a general warning 'msg'
|
||||
virtual void Warning(const wchar_t* msg);
|
||||
//! Handle a general warning 'msg'
|
||||
virtual void Warning(int line, int col, const wchar_t* msg);
|
||||
//! Handle general error 'msg' (eg, a semantic error)
|
||||
virtual void Error(int line, int col, const wchar_t* msg);
|
||||
//! Handle syntax error 'n', uses strerror for the message, calls Error()
|
||||
virtual void SynErr(int line, int col, int n);
|
||||
//! Handle a general exception 'msg'
|
||||
virtual void Exception(const wchar_t* msg);
|
||||
|
||||
}; // Errors
|
||||
|
||||
|
||||
//! A Coco/R Parser
|
||||
class Parser {
|
||||
private:
|
||||
enum {
|
||||
_EOF=0,
|
||||
_ident=1,
|
||||
_string=2,
|
||||
_variable=3,
|
||||
_number=4,
|
||||
};
|
||||
static const int maxT = 14;
|
||||
|
||||
static const int minErrDist = 2; //!< min. distance before reporting errors
|
||||
|
||||
Token *dummyToken;
|
||||
bool deleteErrorsDestruct_; //!< delete the 'errors' member in destructor
|
||||
int errDist;
|
||||
|
||||
void SynErr(int n); //!< Handle syntax error 'n'
|
||||
void Get();
|
||||
void Expect(int n);
|
||||
bool StartOf(int s);
|
||||
void ExpectWeak(int n, int follow);
|
||||
bool WeakSeparator(int n, int syFol, int repFol);
|
||||
|
||||
public:
|
||||
Scanner *scanner;
|
||||
Errors *errors;
|
||||
|
||||
Token *t; //!< last recognized token
|
||||
Token *la; //!< lookahead token
|
||||
|
||||
private:
|
||||
//- The parent dictionary
|
||||
dictionary* dict_;
|
||||
|
||||
//- The calculation result
|
||||
scalar val;
|
||||
|
||||
//- lookup dictionary entry
|
||||
scalar getDictLookup() const
|
||||
{
|
||||
scalar dictValue = 0;
|
||||
|
||||
if (!dict_)
|
||||
{
|
||||
FatalErrorIn
|
||||
(
|
||||
"calcEntry::getDictEntry() const"
|
||||
) << "No dictionary attached!"
|
||||
<< exit(FatalError);
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
char* str = coco_string_create_char
|
||||
(
|
||||
t->val,
|
||||
1,
|
||||
(coco_string_length(t->val) - 1)
|
||||
);
|
||||
word keyword(str);
|
||||
coco_string_delete(str);
|
||||
|
||||
entry* entryPtr = dict_->lookupEntryPtr(keyword, true, false);
|
||||
if (entryPtr && !entryPtr->isDict())
|
||||
{
|
||||
if (entryPtr->stream().size() != 1)
|
||||
{
|
||||
FatalErrorIn
|
||||
(
|
||||
"calcEntry::getDictEntry() const"
|
||||
) << "keyword " << keyword << " has "
|
||||
<< entryPtr->stream().size() << " values in dictionary "
|
||||
<< exit(FatalError);
|
||||
}
|
||||
entryPtr->stream() >> dictValue;
|
||||
}
|
||||
else
|
||||
{
|
||||
FatalErrorIn
|
||||
(
|
||||
"calcEntry::getDictEntry() const"
|
||||
) << "keyword " << keyword << " is undefined in dictionary "
|
||||
<< exit(FatalError);
|
||||
}
|
||||
|
||||
return dictValue;
|
||||
}
|
||||
|
||||
|
||||
public:
|
||||
|
||||
//- attach a dictionary
|
||||
void dict(const dictionary& dict)
|
||||
{
|
||||
dict_ = const_cast<dictionary*>(&dict);
|
||||
}
|
||||
|
||||
//- Return the calculated result
|
||||
scalar Result() const
|
||||
{
|
||||
return val;
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
//! Construct for the specified scanner
|
||||
/**
|
||||
* Use the default error handling, or optionally provide an error
|
||||
* handler, which will not be deleted upon destruction.
|
||||
*/
|
||||
Parser(Scanner* scan, Errors* err = 0);
|
||||
~Parser();
|
||||
void SemErr(const wchar_t* msg); //!< Handle semantic error
|
||||
|
||||
void calcEntry();
|
||||
void Expr(scalar& val);
|
||||
void Term(scalar& val);
|
||||
void Factor(scalar& val);
|
||||
void Func(scalar& val);
|
||||
|
||||
void Parse(); //!< Execute the parse operation
|
||||
|
||||
}; // end Parser
|
||||
|
||||
} // namespace
|
||||
} // namespace
|
||||
} // namespace
|
||||
|
||||
|
||||
#endif // COCO_calcEntryPARSER_H__
|
||||
|
||||
@ -1,932 +0,0 @@
|
||||
|
||||
|
||||
#include <sstream>
|
||||
|
||||
#include "calcEntryScanner.h"
|
||||
|
||||
// values for the file stream buffering
|
||||
#define MIN_BUFFER_LENGTH 1024 // 1KB
|
||||
#define MAX_BUFFER_LENGTH (64*MIN_BUFFER_LENGTH) // 64KB
|
||||
// value for the heap management
|
||||
#define HEAP_BLOCK_SIZE (64*1024) // 64KB
|
||||
|
||||
|
||||
namespace Foam {
|
||||
namespace functionEntries {
|
||||
namespace calcEntryInternal {
|
||||
|
||||
|
||||
// * * * * * * * * * * Wide Character String Routines * * * * * * * * * * * //
|
||||
|
||||
// string handling, wide character
|
||||
|
||||
wchar_t* coco_string_create(const wchar_t* str) {
|
||||
const int len = coco_string_length(str);
|
||||
wchar_t* dest = new wchar_t[len + 1];
|
||||
if (len) {
|
||||
wcsncpy(dest, str, len);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
wchar_t* coco_string_create(const wchar_t* str, int index, int length) {
|
||||
const int len = (str && *str) ? length : 0;
|
||||
wchar_t* dest = new wchar_t[len + 1];
|
||||
if (len) {
|
||||
wcsncpy(dest, &(str[index]), len);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
|
||||
wchar_t* coco_string_create_lower(const wchar_t* str) {
|
||||
if (!str) { return NULL; }
|
||||
return coco_string_create_lower(str, 0, wcslen(str));
|
||||
}
|
||||
|
||||
|
||||
wchar_t* coco_string_create_lower(const wchar_t* str, int index, int len) {
|
||||
if (!str) { return NULL; }
|
||||
wchar_t* dest = new wchar_t[len + 1];
|
||||
|
||||
for (int i = 0; i < len; i++) {
|
||||
const wchar_t ch = str[index + i];
|
||||
if ((L'A' <= ch) && (ch <= L'Z')) {
|
||||
dest[i] = ch - (L'A' - L'a');
|
||||
}
|
||||
else {
|
||||
dest[i] = ch;
|
||||
}
|
||||
}
|
||||
dest[len] = L'\0';
|
||||
return dest;
|
||||
}
|
||||
|
||||
|
||||
wchar_t* coco_string_create_append(const wchar_t* str1, const wchar_t* str2) {
|
||||
const int str1Len = coco_string_length(str1);
|
||||
const int str2Len = coco_string_length(str2);
|
||||
|
||||
wchar_t* dest = new wchar_t[str1Len + str2Len + 1];
|
||||
|
||||
if (str1Len) { wcscpy(dest, str1); }
|
||||
if (str2Len) { wcscpy(dest + str1Len, str2); }
|
||||
|
||||
dest[str1Len + str2Len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
wchar_t* coco_string_create_append(const wchar_t* str1, const wchar_t ch) {
|
||||
const int len = coco_string_length(str1);
|
||||
wchar_t* dest = new wchar_t[len + 2];
|
||||
wcsncpy(dest, str1, len); // or use if (len) { wcscpy(dest, str1); }
|
||||
dest[len] = ch;
|
||||
dest[len + 1] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
void coco_string_delete(wchar_t* &str) {
|
||||
delete [] str;
|
||||
str = NULL;
|
||||
}
|
||||
|
||||
int coco_string_length(const wchar_t* str) {
|
||||
return str ? wcslen(str) : 0;
|
||||
}
|
||||
|
||||
bool coco_string_endswith(const wchar_t* str, const wchar_t* endstr) {
|
||||
const int strLen = wcslen(str);
|
||||
const int endLen = wcslen(endstr);
|
||||
return (endLen <= strLen) && (wcscmp(str + strLen - endLen, endstr) == 0);
|
||||
}
|
||||
|
||||
int coco_string_indexof(const wchar_t* str, const wchar_t ch) {
|
||||
const wchar_t* fnd = wcschr(str, ch);
|
||||
return fnd ? (fnd - str) : -1;
|
||||
}
|
||||
|
||||
int coco_string_lastindexof(const wchar_t* str, const wchar_t ch) {
|
||||
const wchar_t* fnd = wcsrchr(str, ch);
|
||||
return fnd ? (fnd - str) : -1;
|
||||
}
|
||||
|
||||
void coco_string_merge(wchar_t* &dest, const wchar_t* str) {
|
||||
if (!str) { return; }
|
||||
wchar_t* newstr = coco_string_create_append(dest, str);
|
||||
delete [] dest;
|
||||
dest = newstr;
|
||||
}
|
||||
|
||||
bool coco_string_equal(const wchar_t* str1, const wchar_t* str2) {
|
||||
return wcscmp(str1, str2) == 0;
|
||||
}
|
||||
|
||||
int coco_string_compareto(const wchar_t* str1, const wchar_t* str2) {
|
||||
return wcscmp(str1, str2);
|
||||
}
|
||||
|
||||
int coco_string_hash(const wchar_t* str) {
|
||||
int h = 0;
|
||||
if (!str) { return 0; }
|
||||
while (*str != 0) {
|
||||
h = (h * 7) ^ *str;
|
||||
++str;
|
||||
}
|
||||
if (h < 0) { h = -h; }
|
||||
return h;
|
||||
}
|
||||
|
||||
|
||||
double coco_string_toDouble(const wchar_t* str)
|
||||
{
|
||||
return str ? wcstod(str, NULL) : 0;
|
||||
}
|
||||
|
||||
float coco_string_toFloat(const wchar_t* str)
|
||||
{
|
||||
return str ? wcstof(str, NULL) : 0;
|
||||
}
|
||||
|
||||
|
||||
|
||||
//
|
||||
// string handling, byte character
|
||||
//
|
||||
|
||||
wchar_t* coco_string_create(const char* str) {
|
||||
const int len = str ? strlen(str) : 0;
|
||||
wchar_t* dest = new wchar_t[len + 1];
|
||||
for (int i = 0; i < len; ++i) {
|
||||
dest[i] = wchar_t(str[i]);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
wchar_t* coco_string_create(const char* str, int index, int length) {
|
||||
const int len = str ? length : 0;
|
||||
wchar_t* dest = new wchar_t[len + 1];
|
||||
for (int i = 0; i < len; ++i) {
|
||||
dest[i] = wchar_t(str[index + i]);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
|
||||
char* coco_string_create_char(const wchar_t* str) {
|
||||
const int len = coco_string_length(str);
|
||||
char* dest = new char[len + 1];
|
||||
for (int i = 0; i < len; ++i)
|
||||
{
|
||||
dest[i] = char(str[i]);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
char* coco_string_create_char(const wchar_t* str, int index, int length) {
|
||||
const int len = (str && *str) ? length : 0;
|
||||
char* dest = new char[len + 1];
|
||||
for (int i = 0; i < len; ++i) {
|
||||
dest[i] = char(str[index + i]);
|
||||
}
|
||||
dest[len] = 0;
|
||||
return dest;
|
||||
}
|
||||
|
||||
|
||||
void coco_string_delete(char* &str) {
|
||||
delete [] str;
|
||||
str = NULL;
|
||||
}
|
||||
|
||||
|
||||
double coco_string_toDouble(const char* str)
|
||||
{
|
||||
return str ? strtod(str, NULL) : 0;
|
||||
}
|
||||
|
||||
float coco_string_toFloat(const char* str)
|
||||
{
|
||||
return str ? strtof(str, NULL) : 0;
|
||||
}
|
||||
|
||||
|
||||
// * * * * * * * * * End of Wide Character String Routines * * * * * * * * * //
|
||||
|
||||
|
||||
Token::Token()
|
||||
:
|
||||
kind(0),
|
||||
pos(0),
|
||||
col(0),
|
||||
line(0),
|
||||
val(NULL),
|
||||
next(NULL)
|
||||
{}
|
||||
|
||||
|
||||
// Note: this delete may not be correct if the token was actually
|
||||
// allocated by the internal heap mechanism
|
||||
Token::~Token() {
|
||||
coco_string_delete(val);
|
||||
}
|
||||
|
||||
|
||||
// ----------------------------------------------------------------------------
|
||||
// Buffer Implementation
|
||||
// ----------------------------------------------------------------------------
|
||||
|
||||
Buffer::Buffer(Buffer* b)
|
||||
:
|
||||
buf(b->buf),
|
||||
bufCapacity(b->bufCapacity),
|
||||
bufLen(b->bufLen),
|
||||
bufPos(b->bufPos),
|
||||
bufStart(b->bufStart),
|
||||
fileLen(b->fileLen),
|
||||
cStream(b->cStream),
|
||||
stdStream(b->stdStream),
|
||||
isUserStream_(b->isUserStream_)
|
||||
{
|
||||
// avoid accidental deletion on any of these members
|
||||
b->buf = NULL;
|
||||
b->cStream = NULL;
|
||||
b->stdStream = NULL;
|
||||
}
|
||||
|
||||
|
||||
Buffer::Buffer(FILE* istr, bool isUserStream)
|
||||
:
|
||||
buf(NULL),
|
||||
bufCapacity(0),
|
||||
bufLen(0),
|
||||
bufPos(0),
|
||||
bufStart(0),
|
||||
fileLen(0),
|
||||
cStream(istr),
|
||||
stdStream(NULL),
|
||||
isUserStream_(isUserStream)
|
||||
{
|
||||
// ensure binary read on windows
|
||||
#if _MSC_VER >= 1300
|
||||
_setmode(_fileno(cStream), _O_BINARY);
|
||||
#endif
|
||||
|
||||
if (CanSeek()) {
|
||||
fseek(cStream, 0, SEEK_END);
|
||||
fileLen = ftell(cStream);
|
||||
fseek(cStream, 0, SEEK_SET);
|
||||
bufLen = (fileLen < MAX_BUFFER_LENGTH) ? fileLen : MAX_BUFFER_LENGTH;
|
||||
bufStart = INT_MAX; // nothing in the buffer so far
|
||||
}
|
||||
|
||||
bufCapacity = (bufLen > 0) ? bufLen : MIN_BUFFER_LENGTH;
|
||||
buf = new unsigned char[bufCapacity];
|
||||
if (fileLen > 0) SetPos(0); // setup buffer to position 0 (start)
|
||||
else bufPos = 0; // index 0 is already after the file, thus Pos = 0 is invalid
|
||||
if (bufLen == fileLen && CanSeek()) Close();
|
||||
}
|
||||
|
||||
|
||||
Buffer::Buffer(std::istream* istr, bool isUserStream)
|
||||
:
|
||||
buf(NULL),
|
||||
bufCapacity(0),
|
||||
bufLen(0),
|
||||
bufPos(0),
|
||||
bufStart(0),
|
||||
fileLen(0),
|
||||
cStream(NULL),
|
||||
stdStream(istr),
|
||||
isUserStream_(isUserStream)
|
||||
{
|
||||
// ensure binary read on windows
|
||||
#if _MSC_VER >= 1300
|
||||
// TODO
|
||||
#endif
|
||||
}
|
||||
|
||||
|
||||
Buffer::Buffer(std::string& str)
|
||||
:
|
||||
buf(NULL),
|
||||
bufCapacity(0),
|
||||
bufLen(0),
|
||||
bufPos(0),
|
||||
bufStart(0),
|
||||
fileLen(0),
|
||||
cStream(NULL),
|
||||
stdStream(new std::istringstream(str)),
|
||||
isUserStream_(false)
|
||||
{}
|
||||
|
||||
|
||||
Buffer::Buffer(const unsigned char* chars, int len)
|
||||
:
|
||||
buf(new unsigned char[len]),
|
||||
bufCapacity(len),
|
||||
bufLen(len),
|
||||
bufPos(0),
|
||||
bufStart(0),
|
||||
fileLen(len),
|
||||
cStream(NULL),
|
||||
stdStream(NULL),
|
||||
isUserStream_(false)
|
||||
{
|
||||
memcpy(this->buf, chars, len*sizeof(char));
|
||||
}
|
||||
|
||||
|
||||
Buffer::Buffer(const char* chars, int len)
|
||||
:
|
||||
buf(new unsigned char[len]),
|
||||
bufCapacity(len),
|
||||
bufLen(len),
|
||||
bufPos(0),
|
||||
bufStart(0),
|
||||
fileLen(len),
|
||||
cStream(NULL),
|
||||
stdStream(NULL),
|
||||
isUserStream_(false)
|
||||
{
|
||||
memcpy(this->buf, chars, len*sizeof(char));
|
||||
}
|
||||
|
||||
|
||||
Buffer::~Buffer() {
|
||||
Close();
|
||||
if (buf) {
|
||||
delete [] buf;
|
||||
buf = NULL;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
void Buffer::Close() {
|
||||
if (!isUserStream_) {
|
||||
if (cStream) {
|
||||
fclose(cStream);
|
||||
cStream = NULL;
|
||||
}
|
||||
else if (stdStream) {
|
||||
delete stdStream;
|
||||
stdStream = 0;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
int Buffer::Read() {
|
||||
if (stdStream)
|
||||
{
|
||||
int ch = stdStream->get();
|
||||
if (stdStream->eof())
|
||||
{
|
||||
return EoF;
|
||||
}
|
||||
return ch;
|
||||
}
|
||||
|
||||
if (bufPos < bufLen) {
|
||||
return buf[bufPos++];
|
||||
} else if (GetPos() < fileLen) {
|
||||
SetPos(GetPos()); // shift buffer start to Pos
|
||||
return buf[bufPos++];
|
||||
} else if (cStream && !CanSeek() && (ReadNextStreamChunk() > 0)) {
|
||||
return buf[bufPos++];
|
||||
} else {
|
||||
return EoF;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
int UTF8Buffer::Read() {
|
||||
int ch;
|
||||
do {
|
||||
ch = Buffer::Read();
|
||||
// until we find a utf8 start (0xxxxxxx or 11xxxxxx)
|
||||
} while ((ch >= 128) && ((ch & 0xC0) != 0xC0) && (ch != EoF));
|
||||
if (ch < 128 || ch == EoF) {
|
||||
// nothing to do, first 127 chars are the same in ascii and utf8
|
||||
// 0xxxxxxx or end of file character
|
||||
} else if ((ch & 0xF0) == 0xF0) {
|
||||
// 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
|
||||
int c1 = ch & 0x07; ch = Buffer::Read();
|
||||
int c2 = ch & 0x3F; ch = Buffer::Read();
|
||||
int c3 = ch & 0x3F; ch = Buffer::Read();
|
||||
int c4 = ch & 0x3F;
|
||||
ch = (((((c1 << 6) | c2) << 6) | c3) << 6) | c4;
|
||||
} else if ((ch & 0xE0) == 0xE0) {
|
||||
// 1110xxxx 10xxxxxx 10xxxxxx
|
||||
int c1 = ch & 0x0F; ch = Buffer::Read();
|
||||
int c2 = ch & 0x3F; ch = Buffer::Read();
|
||||
int c3 = ch & 0x3F;
|
||||
ch = (((c1 << 6) | c2) << 6) | c3;
|
||||
} else if ((ch & 0xC0) == 0xC0) {
|
||||
// 110xxxxx 10xxxxxx
|
||||
int c1 = ch & 0x1F; ch = Buffer::Read();
|
||||
int c2 = ch & 0x3F;
|
||||
ch = (c1 << 6) | c2;
|
||||
}
|
||||
return ch;
|
||||
}
|
||||
|
||||
|
||||
int Buffer::Peek() {
|
||||
int curPos = GetPos();
|
||||
int ch = Read();
|
||||
SetPos(curPos);
|
||||
return ch;
|
||||
}
|
||||
|
||||
|
||||
int Buffer::GetPos() const {
|
||||
if (stdStream)
|
||||
{
|
||||
return stdStream->tellg();
|
||||
}
|
||||
|
||||
return bufPos + bufStart;
|
||||
}
|
||||
|
||||
|
||||
void Buffer::SetPos(int value) {
|
||||
if (stdStream)
|
||||
{
|
||||
stdStream->seekg(value, std::ios::beg);
|
||||
return;
|
||||
}
|
||||
|
||||
if ((value >= fileLen) && cStream && !CanSeek()) {
|
||||
// Wanted position is after buffer and the stream
|
||||
// is not seek-able e.g. network or console,
|
||||
// thus we have to read the stream manually till
|
||||
// the wanted position is in sight.
|
||||
while ((value >= fileLen) && (ReadNextStreamChunk() > 0))
|
||||
{}
|
||||
}
|
||||
|
||||
if ((value < 0) || (value > fileLen)) {
|
||||
wprintf(L"--- buffer out of bounds access, position: %d\n", value);
|
||||
::exit(1);
|
||||
}
|
||||
|
||||
if ((value >= bufStart) && (value < (bufStart + bufLen))) { // already in buffer
|
||||
bufPos = value - bufStart;
|
||||
} else if (cStream) { // must be swapped in
|
||||
fseek(cStream, value, SEEK_SET);
|
||||
bufLen = fread(buf, sizeof(char), bufCapacity, cStream);
|
||||
bufStart = value; bufPos = 0;
|
||||
} else {
|
||||
bufPos = fileLen - bufStart; // make Pos return fileLen
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// Read the next chunk of bytes from the stream, increases the buffer
|
||||
// if needed and updates the fields fileLen and bufLen.
|
||||
// Returns the number of bytes read.
|
||||
int Buffer::ReadNextStreamChunk() {
|
||||
int freeLen = bufCapacity - bufLen;
|
||||
if (freeLen == 0) {
|
||||
// in the case of a growing input stream
|
||||
// we can neither seek in the stream, nor can we
|
||||
// foresee the maximum length, thus we must adapt
|
||||
// the buffer size on demand.
|
||||
bufCapacity = bufLen * 2;
|
||||
unsigned char *newBuf = new unsigned char[bufCapacity];
|
||||
memcpy(newBuf, buf, bufLen*sizeof(char));
|
||||
delete [] buf;
|
||||
buf = newBuf;
|
||||
freeLen = bufLen;
|
||||
}
|
||||
int read = fread(buf + bufLen, sizeof(char), freeLen, cStream);
|
||||
if (read > 0) {
|
||||
fileLen = bufLen = (bufLen + read);
|
||||
return read;
|
||||
}
|
||||
// end of stream reached
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
bool Buffer::CanSeek() const {
|
||||
return cStream && (ftell(cStream) != -1);
|
||||
}
|
||||
|
||||
|
||||
// ----------------------------------------------------------------------------
|
||||
// Scanner Implementation
|
||||
// ----------------------------------------------------------------------------
|
||||
|
||||
Scanner::Scanner(FILE* istr)
|
||||
:
|
||||
buffer(new Buffer(istr, true))
|
||||
{
|
||||
Init();
|
||||
}
|
||||
|
||||
|
||||
Scanner::Scanner(std::istream& istr)
|
||||
:
|
||||
buffer(new Buffer(&istr, true))
|
||||
{
|
||||
Init();
|
||||
}
|
||||
|
||||
|
||||
Scanner::Scanner(const wchar_t* fileName) {
|
||||
char *chFileName = coco_string_create_char(fileName);
|
||||
FILE* istr;
|
||||
if ((istr = fopen(chFileName, "rb")) == NULL) {
|
||||
wprintf(L"--- Cannot open file %ls\n", fileName);
|
||||
::exit(1);
|
||||
}
|
||||
coco_string_delete(chFileName);
|
||||
buffer = new Buffer(istr, false);
|
||||
Init();
|
||||
}
|
||||
|
||||
|
||||
Scanner::Scanner(const unsigned char* buf, int len)
|
||||
:
|
||||
buffer(new Buffer(buf, len))
|
||||
{
|
||||
Init();
|
||||
}
|
||||
|
||||
|
||||
Scanner::Scanner(const char* buf, int len)
|
||||
:
|
||||
buffer(new Buffer(buf, len))
|
||||
{
|
||||
Init();
|
||||
}
|
||||
|
||||
|
||||
Scanner::~Scanner() {
|
||||
char* cur = reinterpret_cast<char*>(firstHeap);
|
||||
|
||||
while (cur) {
|
||||
cur = *(reinterpret_cast<char**>(cur + HEAP_BLOCK_SIZE));
|
||||
free(firstHeap);
|
||||
firstHeap = cur;
|
||||
}
|
||||
delete [] tval;
|
||||
delete buffer;
|
||||
}
|
||||
|
||||
|
||||
void Scanner::Init() {
|
||||
for (int i = 65; i <= 90; ++i) start.set(i, 1);
|
||||
for (int i = 95; i <= 95; ++i) start.set(i, 1);
|
||||
for (int i = 97; i <= 122; ++i) start.set(i, 1);
|
||||
start.set(45, 21);
|
||||
for (int i = 48; i <= 57; ++i) start.set(i, 9);
|
||||
start.set(34, 2);
|
||||
start.set(36, 5);
|
||||
start.set(46, 7);
|
||||
start.set(123, 14);
|
||||
start.set(125, 15);
|
||||
start.set(43, 22);
|
||||
start.set(42, 16);
|
||||
start.set(47, 17);
|
||||
start.set(40, 18);
|
||||
start.set(41, 19);
|
||||
start.set(44, 20);
|
||||
start.set(Buffer::EoF, -1);
|
||||
|
||||
|
||||
|
||||
tvalLength = 128;
|
||||
tval = new wchar_t[tvalLength]; // text of current token
|
||||
tlen = 0;
|
||||
tval[tlen] = 0;
|
||||
|
||||
// HEAP_BLOCK_SIZE byte heap + pointer to next heap block
|
||||
heap = malloc(HEAP_BLOCK_SIZE + sizeof(void*));
|
||||
firstHeap = heap;
|
||||
heapEnd =
|
||||
reinterpret_cast<void**>
|
||||
(reinterpret_cast<char*>(heap) + HEAP_BLOCK_SIZE);
|
||||
*heapEnd = 0;
|
||||
heapTop = heap;
|
||||
if (sizeof(Token) > HEAP_BLOCK_SIZE) {
|
||||
wprintf(L"--- Too small HEAP_BLOCK_SIZE\n");
|
||||
::exit(1);
|
||||
}
|
||||
|
||||
pos = -1; line = 1; col = 0;
|
||||
oldEols = 0;
|
||||
NextCh();
|
||||
if (ch == 0xEF) { // check optional byte order mark for UTF-8
|
||||
NextCh(); int ch1 = ch;
|
||||
NextCh(); int ch2 = ch;
|
||||
if (ch1 != 0xBB || ch2 != 0xBF) {
|
||||
wprintf(L"Illegal byte order mark at start of file");
|
||||
::exit(1);
|
||||
}
|
||||
Buffer *oldBuf = buffer;
|
||||
buffer = new UTF8Buffer(buffer); col = 0;
|
||||
delete oldBuf; oldBuf = NULL;
|
||||
NextCh();
|
||||
}
|
||||
|
||||
|
||||
pt = tokens = CreateToken(); // first token is a dummy
|
||||
}
|
||||
|
||||
|
||||
void Scanner::NextCh() {
|
||||
if (oldEols > 0) {
|
||||
ch = EOL;
|
||||
oldEols--;
|
||||
}
|
||||
else {
|
||||
pos = buffer->GetPos();
|
||||
ch = buffer->Read(); col++;
|
||||
// replace isolated '\r' by '\n' in order to make
|
||||
// eol handling uniform across Windows, Unix and Mac
|
||||
if (ch == L'\r' && buffer->Peek() != L'\n') ch = EOL;
|
||||
if (ch == EOL) { line++; col = 0; }
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
|
||||
void Scanner::AddCh() {
|
||||
if (tlen >= tvalLength) {
|
||||
tvalLength *= 2;
|
||||
wchar_t *newBuf = new wchar_t[tvalLength];
|
||||
memcpy(newBuf, tval, tlen*sizeof(wchar_t));
|
||||
delete [] tval;
|
||||
tval = newBuf;
|
||||
}
|
||||
if (ch != Buffer::EoF) {
|
||||
tval[tlen++] = ch;
|
||||
NextCh();
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
|
||||
bool Scanner::Comment0() {
|
||||
int level = 1, pos0 = pos, line0 = line, col0 = col;
|
||||
NextCh();
|
||||
if (ch == L'/') {
|
||||
NextCh();
|
||||
for(;;) {
|
||||
if (ch == 10) {
|
||||
level--;
|
||||
if (level == 0) { oldEols = line - line0; NextCh(); return true; }
|
||||
NextCh();
|
||||
} else if (ch == buffer->EoF) return false;
|
||||
else NextCh();
|
||||
}
|
||||
} else {
|
||||
buffer->SetPos(pos0); NextCh(); line = line0; col = col0;
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
bool Scanner::Comment1() {
|
||||
int level = 1, pos0 = pos, line0 = line, col0 = col;
|
||||
NextCh();
|
||||
if (ch == L'*') {
|
||||
NextCh();
|
||||
for(;;) {
|
||||
if (ch == L'*') {
|
||||
NextCh();
|
||||
if (ch == L'/') {
|
||||
level--;
|
||||
if (level == 0) { oldEols = line - line0; NextCh(); return true; }
|
||||
NextCh();
|
||||
}
|
||||
} else if (ch == L'/') {
|
||||
NextCh();
|
||||
if (ch == L'*') {
|
||||
level++; NextCh();
|
||||
}
|
||||
} else if (ch == buffer->EoF) return false;
|
||||
else NextCh();
|
||||
}
|
||||
} else {
|
||||
buffer->SetPos(pos0); NextCh(); line = line0; col = col0;
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
|
||||
void Scanner::CreateHeapBlock() {
|
||||
char* cur = reinterpret_cast<char*>(firstHeap);
|
||||
|
||||
// release unused blocks
|
||||
while
|
||||
(
|
||||
(reinterpret_cast<char*>(tokens) < cur)
|
||||
|| (reinterpret_cast<char*>(tokens) > (cur + HEAP_BLOCK_SIZE))
|
||||
) {
|
||||
cur = *(reinterpret_cast<char**>(cur + HEAP_BLOCK_SIZE));
|
||||
free(firstHeap);
|
||||
firstHeap = cur;
|
||||
}
|
||||
|
||||
// HEAP_BLOCK_SIZE byte heap + pointer to next heap block
|
||||
void* newHeap = malloc(HEAP_BLOCK_SIZE + sizeof(void*));
|
||||
*heapEnd = newHeap;
|
||||
heapEnd =
|
||||
reinterpret_cast<void**>
|
||||
(reinterpret_cast<char*>(newHeap) + HEAP_BLOCK_SIZE);
|
||||
*heapEnd = 0;
|
||||
heap = newHeap;
|
||||
heapTop = heap;
|
||||
}
|
||||
|
||||
|
||||
Token* Scanner::CreateToken() {
|
||||
const int reqMem = sizeof(Token);
|
||||
if
|
||||
(
|
||||
(reinterpret_cast<char*>(heapTop) + reqMem)
|
||||
>= reinterpret_cast<char*>(heapEnd)
|
||||
) {
|
||||
CreateHeapBlock();
|
||||
}
|
||||
// token 'occupies' heap starting at heapTop
|
||||
Token* tok = reinterpret_cast<Token*>(heapTop);
|
||||
// increment past this part of the heap, which is now used
|
||||
heapTop =
|
||||
reinterpret_cast<void*>
|
||||
(reinterpret_cast<char*>(heapTop) + reqMem);
|
||||
tok->val = NULL;
|
||||
tok->next = NULL;
|
||||
return tok;
|
||||
}
|
||||
|
||||
|
||||
void Scanner::AppendVal(Token* tok) {
|
||||
const int reqMem = (tlen + 1) * sizeof(wchar_t);
|
||||
if
|
||||
(
|
||||
(reinterpret_cast<char*>(heapTop) + reqMem)
|
||||
>= reinterpret_cast<char*>(heapEnd)
|
||||
) {
|
||||
if (reqMem > HEAP_BLOCK_SIZE) {
|
||||
wprintf(L"--- Too long token value\n");
|
||||
::exit(1);
|
||||
}
|
||||
CreateHeapBlock();
|
||||
}
|
||||
|
||||
// add text value from heap
|
||||
tok->val = reinterpret_cast<wchar_t*>(heapTop);
|
||||
|
||||
// increment past this part of the heap, which is now used
|
||||
heapTop =
|
||||
reinterpret_cast<void*>
|
||||
(reinterpret_cast<char*>(heapTop) + reqMem);
|
||||
|
||||
// copy the currently parsed tval into the token
|
||||
wcsncpy(tok->val, tval, tlen);
|
||||
tok->val[tlen] = L'\0';
|
||||
}
|
||||
|
||||
|
||||
Token* Scanner::NextToken() {
|
||||
while (ch == ' ' ||
|
||||
(ch >= 9 && ch <= 10) || ch == 13
|
||||
) NextCh();
|
||||
if ((ch == L'/' && Comment0()) || (ch == L'/' && Comment1())) return NextToken();
|
||||
t = CreateToken();
|
||||
t->pos = pos; t->col = col; t->line = line;
|
||||
int state = start.state(ch);
|
||||
tlen = 0; AddCh();
|
||||
|
||||
switch (state) {
|
||||
case -1: { t->kind = eofSym; break; } // NextCh already done
|
||||
case 0: { t->kind = noSym; break; } // NextCh already done
|
||||
case 1:
|
||||
case_1:
|
||||
if ((ch >= L'0' && ch <= L'9') || (ch >= L'A' && ch <= L'Z') || ch == L'_' || (ch >= L'a' && ch <= L'z')) {AddCh(); goto case_1;}
|
||||
else {t->kind = 1; break;}
|
||||
case 2:
|
||||
case_2:
|
||||
if (ch <= 9 || (ch >= 11 && ch <= 12) || (ch >= 14 && ch <= L'!') || (ch >= L'#' && ch <= L'[') || (ch >= L']' && ch <= 65535)) {AddCh(); goto case_2;}
|
||||
else if (ch == L'"') {AddCh(); goto case_4;}
|
||||
else if (ch == 92) {AddCh(); goto case_3;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 3:
|
||||
case_3:
|
||||
if ((ch >= L' ' && ch <= L'~')) {AddCh(); goto case_2;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 4:
|
||||
case_4:
|
||||
{t->kind = 2; break;}
|
||||
case 5:
|
||||
if ((ch >= L'A' && ch <= L'Z') || ch == L'_' || (ch >= L'a' && ch <= L'z')) {AddCh(); goto case_6;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 6:
|
||||
case_6:
|
||||
if ((ch >= L'0' && ch <= L'9') || (ch >= L'A' && ch <= L'Z') || ch == L'_' || (ch >= L'a' && ch <= L'z')) {AddCh(); goto case_6;}
|
||||
else {t->kind = 3; break;}
|
||||
case 7:
|
||||
case_7:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_8;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 8:
|
||||
case_8:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_8;}
|
||||
else {t->kind = 4; break;}
|
||||
case 9:
|
||||
case_9:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_9;}
|
||||
else if (ch == L'E' || ch == L'e') {AddCh(); goto case_10;}
|
||||
else if (ch == L'.') {AddCh(); goto case_13;}
|
||||
else {t->kind = 4; break;}
|
||||
case 10:
|
||||
case_10:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_12;}
|
||||
else if (ch == L'+' || ch == L'-') {AddCh(); goto case_11;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 11:
|
||||
case_11:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_12;}
|
||||
else {t->kind = noSym; break;}
|
||||
case 12:
|
||||
case_12:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_12;}
|
||||
else {t->kind = 4; break;}
|
||||
case 13:
|
||||
case_13:
|
||||
if ((ch >= L'0' && ch <= L'9')) {AddCh(); goto case_13;}
|
||||
else if (ch == L'E' || ch == L'e') {AddCh(); goto case_10;}
|
||||
else {t->kind = 4; break;}
|
||||
case 14:
|
||||
{t->kind = 5; break;}
|
||||
case 15:
|
||||
{t->kind = 6; break;}
|
||||
case 16:
|
||||
{t->kind = 9; break;}
|
||||
case 17:
|
||||
{t->kind = 10; break;}
|
||||
case 18:
|
||||
{t->kind = 11; break;}
|
||||
case 19:
|
||||
{t->kind = 12; break;}
|
||||
case 20:
|
||||
{t->kind = 13; break;}
|
||||
case 21:
|
||||
if (ch == L'.') {AddCh(); goto case_7;}
|
||||
else {t->kind = 8; break;}
|
||||
case 22:
|
||||
if (ch == L'.') {AddCh(); goto case_7;}
|
||||
else {t->kind = 7; break;}
|
||||
|
||||
}
|
||||
AppendVal(t);
|
||||
return t;
|
||||
}
|
||||
|
||||
|
||||
// get the next token (possibly a token already seen during peeking)
|
||||
Token* Scanner::Scan() {
|
||||
if (tokens->next == NULL) {
|
||||
return pt = tokens = NextToken();
|
||||
} else {
|
||||
pt = tokens = tokens->next;
|
||||
return tokens;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
// peek for the next token, ignore pragmas
|
||||
Token* Scanner::Peek() {
|
||||
do {
|
||||
if (pt->next == NULL) {
|
||||
pt->next = NextToken();
|
||||
}
|
||||
pt = pt->next;
|
||||
} while (pt->kind > maxT); // skip pragmas
|
||||
|
||||
return pt;
|
||||
}
|
||||
|
||||
|
||||
// make sure that peeking starts at the current scan position
|
||||
void Scanner::ResetPeek() {
|
||||
pt = tokens;
|
||||
}
|
||||
|
||||
|
||||
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
|
||||
|
||||
} // namespace
|
||||
} // namespace
|
||||
} // namespace
|
||||
|
||||
|
||||
// ************************************************************************* //
|
||||
@ -1,396 +0,0 @@
|
||||
|
||||
|
||||
#ifndef COCO_calcEntrySCANNER_H__
|
||||
#define COCO_calcEntrySCANNER_H__
|
||||
|
||||
#include <climits>
|
||||
#include <cstdio>
|
||||
#include <cstdlib>
|
||||
#include <cstring>
|
||||
#include <cwchar>
|
||||
#include <string>
|
||||
#include <fstream>
|
||||
#include <iostream>
|
||||
|
||||
// io.h and fcntl are used to ensure binary read from streams on windows
|
||||
#if _MSC_VER >= 1300
|
||||
#include <io.h>
|
||||
#include <fcntl.h>
|
||||
#endif
|
||||
|
||||
#if _MSC_VER >= 1400
|
||||
#define coco_swprintf swprintf_s
|
||||
#elif _MSC_VER >= 1300
|
||||
#define coco_swprintf _snwprintf
|
||||
#else
|
||||
// assume every other compiler knows swprintf
|
||||
#define coco_swprintf swprintf
|
||||
#endif
|
||||
|
||||
|
||||
#define COCO_WCHAR_MAX 65535
|
||||
|
||||
|
||||
namespace Foam {
|
||||
namespace functionEntries {
|
||||
namespace calcEntryInternal {
|
||||
|
||||
|
||||
|
||||
// * * * * * * * * * * Wide Character String Routines * * * * * * * * * * * //
|
||||
|
||||
//
|
||||
// string handling, wide character
|
||||
// ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
//! Create by copying str
|
||||
wchar_t* coco_string_create(const wchar_t* str);
|
||||
|
||||
//! Create a substring of str starting at index and length characters long
|
||||
wchar_t* coco_string_create(const wchar_t* str, int index, int length);
|
||||
|
||||
//! Create a lowercase string from str
|
||||
wchar_t* coco_string_create_lower(const wchar_t* str);
|
||||
|
||||
//! Create a lowercase substring from str starting at index and length characters long
|
||||
wchar_t* coco_string_create_lower(const wchar_t* str, int index, int length);
|
||||
|
||||
//! Create a string by concatenating str1 and str2
|
||||
wchar_t* coco_string_create_append(const wchar_t* str1, const wchar_t* str2);
|
||||
|
||||
//! Create a string by concatenating a character to the end of str
|
||||
wchar_t* coco_string_create_append(const wchar_t* str, const wchar_t ch);
|
||||
|
||||
//! Free storage and nullify the argument
|
||||
void coco_string_delete(wchar_t* &str);
|
||||
|
||||
//! The length of the str, or 0 if the str is NULL
|
||||
int coco_string_length(const wchar_t* str);
|
||||
|
||||
//! Return true if the str ends with the endstr
|
||||
bool coco_string_endswith(const wchar_t* str, const wchar_t* endstr);
|
||||
|
||||
//! Return the index of the first occurrence of ch.
|
||||
// Return -1 if nothing is found.
|
||||
int coco_string_indexof(const wchar_t* str, const wchar_t ch);
|
||||
|
||||
//! Return the index of the last occurrence of ch.
|
||||
// Return -1 if nothing is found.
|
||||
int coco_string_lastindexof(const wchar_t* str, const wchar_t ch);
|
||||
|
||||
//! Append str to dest
|
||||
void coco_string_merge(wchar_t* &dest, const wchar_t* str);
|
||||
|
||||
//! Compare strings, return true if they are equal
|
||||
bool coco_string_equal(const wchar_t* str1, const wchar_t* str2);
|
||||
|
||||
//! Compare strings, return 0 if they are equal
|
||||
int coco_string_compareto(const wchar_t* str1, const wchar_t* str2);
|
||||
|
||||
//! Simple string hashing function
|
||||
int coco_string_hash(const wchar_t* str);
|
||||
|
||||
//
|
||||
// String conversions
|
||||
// ~~~~~~~~~~~~~~~~~~
|
||||
|
||||
//! Convert wide string to double
|
||||
double coco_string_toDouble(const wchar_t* str);
|
||||
|
||||
//! Convert wide string to float
|
||||
float coco_string_toFloat(const wchar_t* str);
|
||||
|
||||
//
|
||||
// String handling, byte character
|
||||
// ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
//! Create by copying byte str
|
||||
wchar_t* coco_string_create(const char* str);
|
||||
|
||||
//! Create a substring of byte str starting at index and length characters long
|
||||
wchar_t* coco_string_create(const char* str, int index, int length);
|
||||
|
||||
//! Create a byte string by copying str
|
||||
char* coco_string_create_char(const wchar_t* str);
|
||||
|
||||
//! Create a byte substring of str starting at index and length characters long
|
||||
char* coco_string_create_char(const wchar_t* str, int index, int length);
|
||||
|
||||
//! Free storage and nullify the argument
|
||||
void coco_string_delete(char* &str);
|
||||
|
||||
|
||||
//
|
||||
// String conversions, byte character
|
||||
// ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
//! Convert byte string to double
|
||||
double coco_string_toDouble(const char* str);
|
||||
|
||||
//! Convert byte string to float
|
||||
float coco_string_toFloat(const char* str);
|
||||
|
||||
// * * * * * * * * * End of Wide Character String Routines * * * * * * * * * //
|
||||
|
||||
|
||||
//! Scanner Token
|
||||
class Token
|
||||
{
|
||||
public:
|
||||
int kind; //!< token kind
|
||||
int pos; //!< token position in the source text (starting at 0)
|
||||
int col; //!< token column (starting at 1)
|
||||
int line; //!< token line (starting at 1)
|
||||
wchar_t* val; //!< token value
|
||||
Token *next; //!< Peek tokens are kept in linked list
|
||||
|
||||
Token(); //!< Construct null
|
||||
~Token(); //!< Destructor - cleanup allocated val??
|
||||
};
|
||||
|
||||
|
||||
//! Scanner Buffer
|
||||
//
|
||||
//! This Buffer supports the following cases:
|
||||
//! -# seekable stream (file)
|
||||
//! -# whole stream in buffer
|
||||
//! -# part of stream in buffer
|
||||
//! -# non seekable stream (network, console)
|
||||
class Buffer {
|
||||
private:
|
||||
unsigned char *buf; //!< input buffer
|
||||
int bufCapacity; //!< capacity of buf
|
||||
int bufLen; //!< length of buffer
|
||||
int bufPos; //!< current position in buffer
|
||||
int bufStart; //!< position of first byte in buffer relative to input stream
|
||||
int fileLen; //!< length of input stream (may change if the stream is no file)
|
||||
FILE* cStream; //!< input stdio stream (normally seekable)
|
||||
std::istream* stdStream; //!< STL std stream (seekable)
|
||||
bool isUserStream_; //!< was the stream opened by the user?
|
||||
|
||||
int ReadNextStreamChunk();
|
||||
bool CanSeek() const; //!< true if stream can be seeked otherwise false
|
||||
|
||||
protected:
|
||||
Buffer(Buffer*); //!< for the UTF8Buffer
|
||||
|
||||
public:
|
||||
static const int EoF = COCO_WCHAR_MAX + 1;
|
||||
|
||||
//! Attach buffer to a stdio stream.
|
||||
//! User streams are not closed in the destructor
|
||||
Buffer(FILE*, bool isUserStream = true);
|
||||
|
||||
//! Attach buffer to an STL std stream
|
||||
//! User streams are not closed in the destructor
|
||||
explicit Buffer(std::istream*, bool isUserStream = true);
|
||||
|
||||
//! Copy buffer contents from constant string
|
||||
//! Handled internally as an istringstream
|
||||
explicit Buffer(std::string&);
|
||||
|
||||
//! Copy buffer contents from constant character string
|
||||
Buffer(const unsigned char* chars, int len);
|
||||
//! Copy buffer contents from constant character string
|
||||
Buffer(const char* chars, int len);
|
||||
|
||||
//! Close stream (but not user streams) and free buf (if any)
|
||||
virtual ~Buffer();
|
||||
|
||||
virtual void Close(); //!< Close stream (but not user streams)
|
||||
virtual int Read(); //!< Get character from stream or buffer
|
||||
virtual int Peek(); //!< Peek character from stream or buffer
|
||||
|
||||
virtual int GetPos() const;
|
||||
virtual void SetPos(int value);
|
||||
};
|
||||
|
||||
|
||||
//! A Scanner buffer that handles UTF-8 characters
|
||||
class UTF8Buffer : public Buffer {
|
||||
public:
|
||||
UTF8Buffer(Buffer* b) : Buffer(b) {}
|
||||
virtual int Read();
|
||||
};
|
||||
|
||||
|
||||
//------------------------------------------------------------------------------
|
||||
// StartStates
|
||||
//------------------------------------------------------------------------------
|
||||
//! maps characters (integers) to start states of tokens
|
||||
class StartStates {
|
||||
class Elem {
|
||||
public:
|
||||
int key, val;
|
||||
Elem *next;
|
||||
Elem(int k, int v) :
|
||||
key(k), val(v), next(0)
|
||||
{}
|
||||
};
|
||||
|
||||
Elem **tab;
|
||||
|
||||
public:
|
||||
StartStates() :
|
||||
tab(new Elem*[128])
|
||||
{
|
||||
memset(tab, 0, 128*sizeof(Elem*));
|
||||
}
|
||||
|
||||
virtual ~StartStates() {
|
||||
for (int i = 0; i < 128; ++i) {
|
||||
Elem *e = tab[i];
|
||||
while (e) {
|
||||
Elem *next = e->next;
|
||||
delete e;
|
||||
e = next;
|
||||
}
|
||||
}
|
||||
delete [] tab;
|
||||
}
|
||||
|
||||
void set(int key, int val) {
|
||||
Elem *e = new Elem(key, val);
|
||||
const int k = unsigned(key) % 128;
|
||||
e->next = tab[k];
|
||||
tab[k] = e;
|
||||
}
|
||||
|
||||
int state(int key) {
|
||||
Elem *e = tab[unsigned(key) % 128];
|
||||
while (e && e->key != key) e = e->next;
|
||||
return e ? e->val : 0;
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
//------------------------------------------------------------------------------
|
||||
// KeywordMap
|
||||
//------------------------------------------------------------------------------
|
||||
//! maps strings to integers (identifiers to keyword kinds)
|
||||
class KeywordMap {
|
||||
class Elem {
|
||||
public:
|
||||
wchar_t *key;
|
||||
int val;
|
||||
Elem *next;
|
||||
Elem(const wchar_t *k, int v) :
|
||||
key(coco_string_create(k)), val(v), next(0)
|
||||
{}
|
||||
virtual ~Elem() {
|
||||
coco_string_delete(key);
|
||||
}
|
||||
};
|
||||
|
||||
Elem **tab;
|
||||
|
||||
public:
|
||||
KeywordMap() :
|
||||
tab(new Elem*[128])
|
||||
{
|
||||
memset(tab, 0, 128*sizeof(Elem*));
|
||||
}
|
||||
|
||||
virtual ~KeywordMap() {
|
||||
for (int i = 0; i < 128; ++i) {
|
||||
Elem *e = tab[i];
|
||||
while (e) {
|
||||
Elem *next = e->next;
|
||||
delete e;
|
||||
e = next;
|
||||
}
|
||||
}
|
||||
delete [] tab;
|
||||
}
|
||||
|
||||
void set(const wchar_t *key, int val) {
|
||||
Elem *e = new Elem(key, val);
|
||||
const int k = coco_string_hash(key) % 128;
|
||||
e->next = tab[k];
|
||||
tab[k] = e;
|
||||
}
|
||||
|
||||
int get(const wchar_t *key, int defaultVal) {
|
||||
Elem *e = tab[coco_string_hash(key) % 128];
|
||||
while (e && !coco_string_equal(e->key, key)) e = e->next;
|
||||
return e ? e->val : defaultVal;
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
//! A Coco/R Scanner
|
||||
class Scanner {
|
||||
private:
|
||||
static const int maxT = 14;
|
||||
static const int noSym = 14;
|
||||
|
||||
static const int eofSym = 0; //!< end-of-file token id
|
||||
static const char EOL = '\n'; //!< end-of-line character
|
||||
|
||||
void *firstHeap; //!< the start of the heap management
|
||||
void *heap; //!< the currently active block
|
||||
void *heapTop; //!< the top of the heap
|
||||
void **heapEnd; //!< the end of the last heap block
|
||||
|
||||
StartStates start; //!< A map of start states for particular characters
|
||||
KeywordMap keywords; //!< A hash of keyword literals to token kind
|
||||
|
||||
Token *t; //!< current token
|
||||
wchar_t *tval; //!< text of current token
|
||||
int tvalLength; //!< maximum capacity (length) for tval
|
||||
int tlen; //!< length of tval
|
||||
|
||||
Token *tokens; //!< list of tokens already peeked (first token is a dummy)
|
||||
Token *pt; //!< current peek token
|
||||
|
||||
int ch; //!< current input character
|
||||
|
||||
int pos; //!< byte position of current character
|
||||
int line; //!< line number of current character
|
||||
int col; //!< column number of current character
|
||||
int oldEols; //!< the number of EOLs that appeared in a comment
|
||||
|
||||
void CreateHeapBlock(); //!< add a heap block, freeing unused ones
|
||||
Token* CreateToken(); //!< fit token on the heap
|
||||
void AppendVal(Token* tok); //!< adjust tok->val to point to the heap and copy tval into it
|
||||
|
||||
void Init(); //!< complete the initialization for the constructors
|
||||
void NextCh(); //!< get the next input character into ch
|
||||
void AddCh(); //!< append the character ch to tval
|
||||
bool Comment0();
|
||||
bool Comment1();
|
||||
|
||||
Token* NextToken(); //!< get the next token
|
||||
|
||||
public:
|
||||
//! The scanner buffer
|
||||
Buffer *buffer;
|
||||
|
||||
//! Using an existing open file handle for the scanner
|
||||
Scanner(FILE*);
|
||||
|
||||
//! Using an existing open STL std stream
|
||||
explicit Scanner(std::istream&);
|
||||
|
||||
//! Open a file for reading and attach scanner
|
||||
explicit Scanner(const wchar_t* fileName);
|
||||
|
||||
//! Attach scanner to an existing character buffer
|
||||
Scanner(const unsigned char* chars, int len);
|
||||
//! Attach scanner to an existing character buffer
|
||||
Scanner(const char* chars, int len);
|
||||
|
||||
~Scanner(); //!< free heap and allocated memory
|
||||
Token* Scan(); //!< get the next token (possibly a token already seen during peeking)
|
||||
Token* Peek(); //!< peek for the next token, ignore pragmas
|
||||
void ResetPeek(); //!< ensure that peeking starts at the current scan position
|
||||
|
||||
}; // end Scanner
|
||||
|
||||
} // namespace
|
||||
} // namespace
|
||||
} // namespace
|
||||
|
||||
|
||||
#endif // COCO_calcEntrySCANNER_H__
|
||||
|
||||
Reference in New Issue
Block a user