Lil-Ran/pycdc_fork
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 1 Activity

Use C++ streams over C style IO

Browse Source
This commit is contained in:
Perceval Wajsbürt
2023-06-02 00:36:58 +02:00
committed by Perceval Wajsbürt
parent 409f175827
commit c4c35fc531
10 changed files with 488 additions and 460 deletions
Download Patch File Download Diff File Expand all files Collapse all files

526
ASTree.cpp
View File

File diff suppressed because it is too large Load Diff

4
ASTree.h
View File

@@ -4,8 +4,8 @@
#include "ASTNode.h" #include "ASTNode.h"
PycRef<ASTNode> BuildFromCode(PycRef<PycCode> code, PycModule* mod); PycRef<ASTNode> BuildFromCode(PycRef<PycCode> code, PycModule* mod);
void print_src(PycRef<ASTNode> node, PycModule* mod); void print_src(PycRef<ASTNode> node, PycModule* mod, std::ostream& pyc_output);
void decompyle(PycRef<PycCode> code, PycModule* mod); void decompyle(PycRef<PycCode> code, PycModule* mod, std::ostream& pyc_output);
#endif #endif

148
bytecode.cpp
View File

@@ -160,10 +160,10 @@ bool Pyc::IsCompareArg(int opcode)
return (opcode == Pyc::COMPARE_OP_A); return (opcode == Pyc::COMPARE_OP_A);
} }
void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_string_quote) void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_string_quote, std::ostream& pyc_output)
{ {
if (obj == NULL) { if (obj == NULL) {
fputs("<NULL>", pyc_output); pyc_output << "<NULL>";
return; return;
} }
@@ -190,112 +190,112 @@ void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_str
case PycObject::TYPE_TUPLE: case PycObject::TYPE_TUPLE:
case PycObject::TYPE_SMALL_TUPLE: case PycObject::TYPE_SMALL_TUPLE:
{ {
fputs("(", pyc_output); pyc_output << "(";
PycTuple::value_t values = obj.cast<PycTuple>()->values(); PycTuple::value_t values = obj.cast<PycTuple>()->values();
auto it = values.cbegin(); auto it = values.cbegin();
if (it != values.cend()) { if (it != values.cend()) {
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
while (++it != values.cend()) { while (++it != values.cend()) {
fputs(", ", pyc_output); pyc_output << ", ";
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
} }
} }
if (values.size() == 1) if (values.size() == 1)
fputs(",)", pyc_output); pyc_output << ",)";
else else
fputs(")", pyc_output); pyc_output << ")";
} }
break; break;
case PycObject::TYPE_LIST: case PycObject::TYPE_LIST:
{ {
fputs("[", pyc_output); pyc_output << "[";
PycList::value_t values = obj.cast<PycList>()->values(); PycList::value_t values = obj.cast<PycList>()->values();
auto it = values.cbegin(); auto it = values.cbegin();
if (it != values.cend()) { if (it != values.cend()) {
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
while (++it != values.cend()) { while (++it != values.cend()) {
fputs(", ", pyc_output); pyc_output << ", ";
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
} }
} }
fputs("]", pyc_output); pyc_output << "]";
} }
break; break;
case PycObject::TYPE_DICT: case PycObject::TYPE_DICT:
{ {
fputs("{", pyc_output); pyc_output << "{";
PycDict::key_t keys = obj.cast<PycDict>()->keys(); PycDict::key_t keys = obj.cast<PycDict>()->keys();
PycDict::value_t values = obj.cast<PycDict>()->values(); PycDict::value_t values = obj.cast<PycDict>()->values();
auto ki = keys.cbegin(); auto ki = keys.cbegin();
auto vi = values.cbegin(); auto vi = values.cbegin();
if (ki != keys.cend()) { if (ki != keys.cend()) {
print_const(*ki, mod); print_const(*ki, mod, nullptr, pyc_output);
fputs(": ", pyc_output); pyc_output << ": ";
print_const(*vi, mod); print_const(*vi, mod, nullptr, pyc_output);
while (++ki != keys.cend()) { while (++ki != keys.cend()) {
++vi; ++vi;
fputs(", ", pyc_output); pyc_output << ", ";
print_const(*ki, mod); print_const(*ki, mod, nullptr, pyc_output);
fputs(": ", pyc_output); pyc_output << ": ";
print_const(*vi, mod); print_const(*vi, mod, nullptr, pyc_output);
} }
} }
fputs("}", pyc_output); pyc_output << "}";
} }
break; break;
case PycObject::TYPE_SET: case PycObject::TYPE_SET:
{ {
fputs("{", pyc_output); pyc_output << "{";
PycSet::value_t values = obj.cast<PycSet>()->values(); PycSet::value_t values = obj.cast<PycSet>()->values();
auto it = values.cbegin(); auto it = values.cbegin();
if (it != values.cend()) { if (it != values.cend()) {
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
while (++it != values.cend()) { while (++it != values.cend()) {
fputs(", ", pyc_output); pyc_output << ", ";
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
} }
} }
fputs("}", pyc_output); pyc_output << "}";
} }
break; break;
case PycObject::TYPE_FROZENSET: case PycObject::TYPE_FROZENSET:
{ {
fputs("frozenset({", pyc_output); pyc_output << "frozenset({";
PycSet::value_t values = obj.cast<PycSet>()->values(); PycSet::value_t values = obj.cast<PycSet>()->values();
auto it = values.cbegin(); auto it = values.cbegin();
if (it != values.cend()) { if (it != values.cend()) {
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
while (++it != values.cend()) { while (++it != values.cend()) {
fputs(", ", pyc_output); pyc_output << ", ";
print_const(*it, mod); print_const(*it, mod, nullptr, pyc_output);
} }
} }
fputs("})", pyc_output); pyc_output << "})";
} }
break; break;
case PycObject::TYPE_NONE: case PycObject::TYPE_NONE:
fputs("None", pyc_output); pyc_output << "None";
break; break;
case PycObject::TYPE_TRUE: case PycObject::TYPE_TRUE:
fputs("True", pyc_output); pyc_output << "True";
break; break;
case PycObject::TYPE_FALSE: case PycObject::TYPE_FALSE:
fputs("False", pyc_output); pyc_output << "False";
break; break;
case PycObject::TYPE_ELLIPSIS: case PycObject::TYPE_ELLIPSIS:
fputs("...", pyc_output); pyc_output << "...";
break; break;
case PycObject::TYPE_INT: case PycObject::TYPE_INT:
fprintf(pyc_output, "%d", obj.cast<PycInt>()->value()); formatted_print(pyc_output, "%d", obj.cast<PycInt>()->value());
break; break;
case PycObject::TYPE_LONG: case PycObject::TYPE_LONG:
fprintf(pyc_output, "%s", obj.cast<PycLong>()->repr().c_str()); formatted_print(pyc_output, "%s", obj.cast<PycLong>()->repr().c_str());
break; break;
case PycObject::TYPE_FLOAT: case PycObject::TYPE_FLOAT:
fprintf(pyc_output, "%s", obj.cast<PycFloat>()->value()); formatted_print(pyc_output, "%s", obj.cast<PycFloat>()->value());
break; break;
case PycObject::TYPE_COMPLEX: case PycObject::TYPE_COMPLEX:
fprintf(pyc_output, "(%s+%sj)", obj.cast<PycComplex>()->value(), formatted_print(pyc_output, "(%s+%sj)", obj.cast<PycComplex>()->value(),
obj.cast<PycComplex>()->imag()); obj.cast<PycComplex>()->imag());
break; break;
case PycObject::TYPE_BINARY_FLOAT: case PycObject::TYPE_BINARY_FLOAT:
@@ -305,31 +305,31 @@ void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_str
bool is_negative = std::signbit(value); bool is_negative = std::signbit(value);
if (std::isnan(value)) { if (std::isnan(value)) {
if (is_negative) { if (is_negative) {
fprintf(pyc_output, "float('-nan')"); pyc_output << "float('-nan')";
} else { } else {
fprintf(pyc_output, "float('nan')"); pyc_output << "float('nan')";
} }
} else if (std::isinf(value)) { } else if (std::isinf(value)) {
if (is_negative) { if (is_negative) {
fprintf(pyc_output, "float('-inf')"); pyc_output << "float('-inf')";
} else { } else {
fprintf(pyc_output, "float('inf')"); pyc_output << "float('inf')";
} }
} else { } else {
fprintf(pyc_output, "%g", value); formatted_print(pyc_output, "%g", value);
} }
} }
break; break;
case PycObject::TYPE_BINARY_COMPLEX: case PycObject::TYPE_BINARY_COMPLEX:
fprintf(pyc_output, "(%g+%gj)", obj.cast<PycCComplex>()->value(), formatted_print(pyc_output, "(%g+%gj)", obj.cast<PycCComplex>()->value(),
obj.cast<PycCComplex>()->imag()); obj.cast<PycCComplex>()->imag());
break; break;
case PycObject::TYPE_CODE: case PycObject::TYPE_CODE:
case PycObject::TYPE_CODE2: case PycObject::TYPE_CODE2:
fprintf(pyc_output, "<CODE> %s", obj.cast<PycCode>()->name()->value()); formatted_print(pyc_output, "<CODE> %s", obj.cast<PycCode>()->name()->value());
break; break;
default: default:
fprintf(pyc_output, "<TYPE: %d>\n", obj->type()); formatted_print(pyc_output, "<TYPE: %d>\n", obj->type());
} }
} }
@@ -363,7 +363,7 @@ void bc_next(PycBuffer& source, PycModule* mod, int& opcode, int& operand, int&
} }
} }
void bc_disasm(PycRef<PycCode> code, PycModule* mod, int indent, unsigned flags) void bc_disasm(PycRef<PycCode> code, PycModule* mod, int indent, unsigned flags, std::ostream &pyc_output)
{ {
static const char *cmp_strings[] = { static const char *cmp_strings[] = {
"<", "<=", "==", "!=", ">", ">=", "in", "not in", "is", "is not", "<", "<=", "==", "!=", ">", ">=", "in", "not in", "is", "is not",
@@ -388,78 +388,78 @@ void bc_disasm(PycRef<PycCode> code, PycModule* mod, int indent, unsigned flags)
continue; continue;
for (int i=0; i<indent; i++) for (int i=0; i<indent; i++)
fputs(" ", pyc_output); pyc_output << " ";
fprintf(pyc_output, "%-7d %-30s", start_pos, Pyc::OpcodeName(opcode)); formatted_print(pyc_output, "%-7d %-30s", start_pos, Pyc::OpcodeName(opcode));
if (opcode >= Pyc::PYC_HAVE_ARG) { if (opcode >= Pyc::PYC_HAVE_ARG) {
if (Pyc::IsConstArg(opcode)) { if (Pyc::IsConstArg(opcode)) {
try { try {
auto constParam = code->getConst(operand); auto constParam = code->getConst(operand);
fprintf(pyc_output, "%d: ", operand); formatted_print(pyc_output, "%d: ", operand);
print_const(constParam, mod); print_const(constParam, mod, nullptr, pyc_output);
} catch (const std::out_of_range &) { } catch (const std::out_of_range &) {
fprintf(pyc_output, "%d <INVALID>", operand); formatted_print(pyc_output, "%d <INVALID>", operand);
} }
} else if (opcode == Pyc::LOAD_GLOBAL_A) { } else if (opcode == Pyc::LOAD_GLOBAL_A) {
// Special case for Python 3.11+ // Special case for Python 3.11+
try { try {
if (operand & 1) if (operand & 1)
fprintf(pyc_output, "%d: NULL + %s", operand, code->getName(operand >> 1)->value()); formatted_print(pyc_output, "%d: NULL + %s", operand, code->getName(operand >> 1)->value());
else else
fprintf(pyc_output, "%d: %s", operand, code->getName(operand >> 1)->value()); formatted_print(pyc_output, "%d: %s", operand, code->getName(operand >> 1)->value());
} catch (const std::out_of_range &) { } catch (const std::out_of_range &) {
fprintf(pyc_output, "%d <INVALID>", operand); formatted_print(pyc_output, "%d <INVALID>", operand);
} }
} else if (Pyc::IsNameArg(opcode)) { } else if (Pyc::IsNameArg(opcode)) {
try { try {
fprintf(pyc_output, "%d: %s", operand, code->getName(operand)->value()); formatted_print(pyc_output, "%d: %s", operand, code->getName(operand)->value());
} catch (const std::out_of_range &) { } catch (const std::out_of_range &) {
fprintf(pyc_output, "%d <INVALID>", operand); formatted_print(pyc_output, "%d <INVALID>", operand);
} }
} else if (Pyc::IsVarNameArg(opcode)) { } else if (Pyc::IsVarNameArg(opcode)) {
try { try {
fprintf(pyc_output, "%d: %s", operand, code->getLocal(operand)->value()); formatted_print(pyc_output, "%d: %s", operand, code->getLocal(operand)->value());
} catch (const std::out_of_range &) { } catch (const std::out_of_range &) {
fprintf(pyc_output, "%d <INVALID>", operand); formatted_print(pyc_output, "%d <INVALID>", operand);
} }
} else if (Pyc::IsCellArg(opcode)) { } else if (Pyc::IsCellArg(opcode)) {
try { try {
fprintf(pyc_output, "%d: %s", operand, code->getCellVar(mod, operand)->value()); formatted_print(pyc_output, "%d: %s", operand, code->getCellVar(mod, operand)->value());
} catch (const std::out_of_range &) { } catch (const std::out_of_range &) {
fprintf(pyc_output, "%d <INVALID>", operand); formatted_print(pyc_output, "%d <INVALID>", operand);
} }
} else if (Pyc::IsJumpOffsetArg(opcode)) { } else if (Pyc::IsJumpOffsetArg(opcode)) {
int offs = operand; int offs = operand;
if (mod->verCompare(3, 10) >= 0) if (mod->verCompare(3, 10) >= 0)
offs *= sizeof(uint16_t); // BPO-27129 offs *= sizeof(uint16_t); // BPO-27129
fprintf(pyc_output, "%d (to %d)", operand, pos+offs); formatted_print(pyc_output, "%d (to %d)", operand, pos+offs);
} else if (Pyc::IsJumpArg(opcode)) { } else if (Pyc::IsJumpArg(opcode)) {
if (mod->verCompare(3, 10) >= 0) // BPO-27129 if (mod->verCompare(3, 10) >= 0) // BPO-27129
fprintf(pyc_output, "%d (to %d)", operand, int(operand * sizeof(uint16_t))); formatted_print(pyc_output, "%d (to %d)", operand, int(operand * sizeof(uint16_t)));
else else
fprintf(pyc_output, "%d", operand); formatted_print(pyc_output, "%d", operand);
} else if (Pyc::IsCompareArg(opcode)) { } else if (Pyc::IsCompareArg(opcode)) {
if (static_cast<size_t>(operand) < cmp_strings_len) if (static_cast<size_t>(operand) < cmp_strings_len)
fprintf(pyc_output, "%d (%s)", operand, cmp_strings[operand]); formatted_print(pyc_output, "%d (%s)", operand, cmp_strings[operand]);
else else
fprintf(pyc_output, "%d (UNKNOWN)", operand); formatted_print(pyc_output, "%d (UNKNOWN)", operand);
} else if (opcode == Pyc::BINARY_OP_A) { } else if (opcode == Pyc::BINARY_OP_A) {
if (static_cast<size_t>(operand) < binop_strings_len) if (static_cast<size_t>(operand) < binop_strings_len)
fprintf(pyc_output, "%d (%s)", operand, binop_strings[operand]); formatted_print(pyc_output, "%d (%s)", operand, binop_strings[operand]);
else else
fprintf(pyc_output, "%d (UNKNOWN)", operand); formatted_print(pyc_output, "%d (UNKNOWN)", operand);
} else if (opcode == Pyc::IS_OP_A) { } else if (opcode == Pyc::IS_OP_A) {
fprintf(pyc_output, "%d (%s)", operand, (operand == 0) ? "is" formatted_print(pyc_output, "%d (%s)", operand, (operand == 0) ? "is"
: (operand == 1) ? "is not" : (operand == 1) ? "is not"
: "UNKNOWN"); : "UNKNOWN");
} else if (opcode == Pyc::CONTAINS_OP_A) { } else if (opcode == Pyc::CONTAINS_OP_A) {
fprintf(pyc_output, "%d (%s)", operand, (operand == 0) ? "in" formatted_print(pyc_output, "%d (%s)", operand, (operand == 0) ? "in"
: (operand == 1) ? "not in" : (operand == 1) ? "not in"
: "UNKNOWN"); : "UNKNOWN");
} else { } else {
fprintf(pyc_output, "%d", operand); formatted_print(pyc_output, "%d", operand);
} }
} }
fputs("\n", pyc_output); pyc_output << "\n";
} }
} }

4
bytecode.h
View File

@@ -35,6 +35,6 @@ bool IsCompareArg(int opcode);
} }
void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_string_quote = nullptr); void print_const(PycRef<PycObject> obj, PycModule* mod, const char* parent_f_string_quote, std::ostream& pyc_output);
void bc_next(PycBuffer& source, PycModule* mod, int& opcode, int& operand, int& pos); void bc_next(PycBuffer& source, PycModule* mod, int& opcode, int& operand, int& pos);
void bc_disasm(PycRef<PycCode> code, PycModule* mod, int indent, unsigned flags); void bc_disasm(PycRef<PycCode> code, PycModule* mod, int indent, unsigned flags, std::ostream& pyc_output);

22
data.cpp
View File

@@ -1,7 +1,7 @@
#include "data.h" #include "data.h"
#include <cstring> #include <cstring>
#include <ostream>
FILE* pyc_output = stdout; #include <vector>
/* PycData */ /* PycData */
int PycData::get16() int PycData::get16()
@@ -82,3 +82,21 @@ int PycBuffer::getBuffer(int bytes, void* buffer)
memcpy(buffer, (m_buffer + m_pos), bytes); memcpy(buffer, (m_buffer + m_pos), bytes);
return bytes; return bytes;
} }
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wvarargs"
int formatted_print(std::ostream& stream, const std::string& format, ...) {
va_list args;
va_start(args, format);
size_t len = std::vsnprintf(NULL, 0, format.c_str(), args);
va_end(args);
std::vector<char> vec(len + 1);
va_start(args, format);
std::vsnprintf(&vec[0], len + 1, format.c_str(), args);
va_end(args);
stream << &vec[0];
return 0;
}
#pragma clang diagnostic pop

3
data.h
View File

@@ -2,6 +2,7 @@
#define _PYC_FILE_H #define _PYC_FILE_H
#include <cstdio> #include <cstdio>
#include <ostream>
#ifdef WIN32 #ifdef WIN32
typedef __int64 Pyc_INT64; typedef __int64 Pyc_INT64;
@@ -56,6 +57,6 @@ private:
int m_size, m_pos; int m_size, m_pos;
}; };
extern FILE* pyc_output; int formatted_print(std::ostream& stream, const std::string& format, ...);
#endif #endif

41
pyc_string.cpp
View File

@@ -2,6 +2,7 @@
#include "pyc_module.h" #include "pyc_module.h"
#include "data.h" #include "data.h"
#include <stdexcept> #include <stdexcept>
#include <ostream>
static bool check_ascii(const std::string& data) static bool check_ascii(const std::string& data)
{ {
@@ -56,15 +57,15 @@ bool PycString::isEqual(PycRef<PycObject> obj) const
return isEqual(strObj->m_value); return isEqual(strObj->m_value);
} }
void OutputString(PycRef<PycString> str, char prefix, bool triple, FILE* F, const char* parent_f_string_quote) void OutputString(PycRef<PycString> str, char prefix, bool triple, std::ostream &pyc_output, const char* parent_f_string_quote)
{ {
if (prefix != 0) if (prefix != 0)
fputc(prefix, F); pyc_output << prefix;
const char* ch = str->value(); const char* ch = str->value();
int len = str->length(); int len = str->length();
if (ch == 0) { if (ch == 0) {
fputs("''", F); pyc_output << "''";
return; return;
} }
@@ -89,51 +90,51 @@ void OutputString(PycRef<PycString> str, char prefix, bool triple, FILE* F, cons
// Output the string // Output the string
if (!parent_f_string_quote) { if (!parent_f_string_quote) {
if (triple) if (triple)
fputs(useQuotes ? "\"\"\"" : "'''", F); pyc_output << (useQuotes ? "\"\"\"" : "'''");
else else
fputc(useQuotes ? '"' : '\'', F); pyc_output << (useQuotes ? '"' : '\'');
} }
while (len--) { while (len--) {
if ((unsigned char)(*ch) < 0x20 || *ch == 0x7F) { if ((unsigned char)(*ch) < 0x20 || *ch == 0x7F) {
if (*ch == '\r') { if (*ch == '\r') {
fputs("\\r", F); pyc_output << "\\r";
} else if (*ch == '\n') { } else if (*ch == '\n') {
if (triple) if (triple)
fputc('\n', F); pyc_output << '\n';
else else
fputs("\\n", F); pyc_output << "\\n";
} else if (*ch == '\t') { } else if (*ch == '\t') {
fputs("\\t", F); pyc_output << "\\t";
} else { } else {
fprintf(F, "\\x%02x", (*ch & 0xFF)); formatted_print(pyc_output, "\\x%02x", (*ch & 0xFF));
} }
} else if ((unsigned char)(*ch) >= 0x80) { } else if ((unsigned char)(*ch) >= 0x80) {
if (str->type() == PycObject::TYPE_UNICODE) { if (str->type() == PycObject::TYPE_UNICODE) {
// Unicode stored as UTF-8... Let the stream interpret it // Unicode stored as UTF-8... Let the stream interpret it
fputc(*ch, F); pyc_output << *ch;
} else { } else {
fprintf(F, "\\x%x", (*ch & 0xFF)); formatted_print(pyc_output, "\\x%x", (*ch & 0xFF));
} }
} else { } else {
if (!useQuotes && *ch == '\'') if (!useQuotes && *ch == '\'')
fputs("\\'", F); pyc_output << "\\'";
else if (useQuotes && *ch == '"') else if (useQuotes && *ch == '"')
fputs("\\\"", F); pyc_output << "\\\"";
else if (*ch == '\\') else if (*ch == '\\')
fputs("\\\\", F); pyc_output << "\\\\";
else if (parent_f_string_quote && *ch == '{') else if (parent_f_string_quote && *ch == '{')
fputs("{{", F); pyc_output << "{{";
else if (parent_f_string_quote && *ch == '}') else if (parent_f_string_quote && *ch == '}')
fputs("}}", F); pyc_output << "}}";
else else
fputc(*ch, F); pyc_output << *ch;
} }
ch++; ch++;
} }
if (!parent_f_string_quote) { if (!parent_f_string_quote) {
if (triple) if (triple)
fputs(useQuotes ? "\"\"\"" : "'''", F); pyc_output << (useQuotes ? "\"\"\"" : "'''");
else else
fputc(useQuotes ? '"' : '\'', F); pyc_output << (useQuotes ? '"' : '\'');
} }
} }

4
pyc_string.h
View File

@@ -31,7 +31,7 @@ private:
std::string m_value; std::string m_value;
}; };
void OutputString(PycRef<PycString> str, char prefix = 0, bool triple = false, void OutputString(PycRef<PycString> str, char prefix, bool triple,
FILE* F = pyc_output, const char* parent_f_string_quote = nullptr); std::ostream& stream, const char* parent_f_string_quote = nullptr);
#endif #endif

180
pycdas.cpp
View File

@@ -2,9 +2,11 @@
#include <cstring> #include <cstring>
#include <cstdarg> #include <cstdarg>
#include <string> #include <string>
#include <iostream>
#include "pyc_module.h" #include "pyc_module.h"
#include "pyc_numeric.h" #include "pyc_numeric.h"
#include "bytecode.h" #include "bytecode.h"
#include <fstream>
#ifdef WIN32 #ifdef WIN32
# define PATHSEP '\\' # define PATHSEP '\\'
@@ -25,57 +27,57 @@ static const char* flag_names[] = {
"<0x10000000>", "<0x20000000>", "<0x40000000>", "<0x80000000>" "<0x10000000>", "<0x20000000>", "<0x40000000>", "<0x80000000>"
}; };
static void print_coflags(unsigned long flags) static void print_coflags(unsigned long flags, std::ostream& pyc_output)
{ {
if (flags == 0) { if (flags == 0) {
fputs("\n", pyc_output); pyc_output << "\n";
return; return;
} }
fputs(" (", pyc_output); pyc_output << " (";
unsigned long f = 1; unsigned long f = 1;
int k = 0; int k = 0;
while (k < 32) { while (k < 32) {
if ((flags & f) != 0) { if ((flags & f) != 0) {
flags &= ~f; flags &= ~f;
if (flags == 0) if (flags == 0)
fputs(flag_names[k], pyc_output); pyc_output << flag_names[k];
else else
fprintf(pyc_output, "%s | ", flag_names[k]); pyc_output << flag_names[k] << " | ";
} }
++k; ++k;
f <<= 1; f <<= 1;
} }
fputs(")\n", pyc_output); pyc_output << ")\n";
} }
static void iputs(int indent, const char* text) static void iputs(std::ostream& pyc_output, int indent, const char* text)
{ {
for (int i=0; i<indent; i++) for (int i=0; i<indent; i++)
fputs(" ", pyc_output); pyc_output << " ";
fputs(text, pyc_output); pyc_output << text;
} }
static void ivprintf(int indent, const char* fmt, va_list varargs) static void ivprintf(std::ostream& pyc_output, int indent, const char* fmt, va_list varargs)
{ {
for (int i=0; i<indent; i++) for (int i=0; i<indent; i++)
fputs(" ", pyc_output); pyc_output << " ";
vfprintf(pyc_output, fmt, varargs); formatted_print(pyc_output, fmt, varargs);
} }
static void iprintf(int indent, const char* fmt, ...) static void iprintf(std::ostream& pyc_output, int indent, const char* fmt, ...)
{ {
va_list varargs; va_list varargs;
va_start(varargs, fmt); va_start(varargs, fmt);
ivprintf(indent, fmt, varargs); ivprintf(pyc_output, indent, fmt, varargs);
va_end(varargs); va_end(varargs);
} }
void output_object(PycRef<PycObject> obj, PycModule* mod, int indent, void output_object(PycRef<PycObject> obj, PycModule* mod, int indent,
unsigned flags) unsigned flags, std::ostream& pyc_output)
{ {
if (obj == NULL) { if (obj == NULL) {
iputs(indent, "<NULL>"); iputs(pyc_output, indent, "<NULL>");
return; return;
} }
@@ -84,175 +86,175 @@ void output_object(PycRef<PycObject> obj, PycModule* mod, int indent,
case PycObject::TYPE_CODE2: case PycObject::TYPE_CODE2:
{ {
PycRef<PycCode> codeObj = obj.cast<PycCode>(); PycRef<PycCode> codeObj = obj.cast<PycCode>();
iputs(indent, "[Code]\n"); iputs(pyc_output, indent, "[Code]\n");
iprintf(indent + 1, "File Name: %s\n", codeObj->fileName()->value()); iprintf(pyc_output, indent + 1, "File Name: %s\n", codeObj->fileName()->value());
iprintf(indent + 1, "Object Name: %s\n", codeObj->name()->value()); iprintf(pyc_output, indent + 1, "Object Name: %s\n", codeObj->name()->value());
if (mod->verCompare(3, 11) >= 0) if (mod->verCompare(3, 11) >= 0)
iprintf(indent + 1, "Qualified Name: %s\n", codeObj->qualName()->value()); iprintf(pyc_output, indent + 1, "Qualified Name: %s\n", codeObj->qualName()->value());
iprintf(indent + 1, "Arg Count: %d\n", codeObj->argCount()); iprintf(pyc_output, indent + 1, "Arg Count: %d\n", codeObj->argCount());
if (mod->verCompare(3, 8) >= 0) if (mod->verCompare(3, 8) >= 0)
iprintf(indent + 1, "Pos Only Arg Count: %d\n", codeObj->posOnlyArgCount()); iprintf(pyc_output, indent + 1, "Pos Only Arg Count: %d\n", codeObj->posOnlyArgCount());
if (mod->majorVer() >= 3) if (mod->majorVer() >= 3)
iprintf(indent + 1, "KW Only Arg Count: %d\n", codeObj->kwOnlyArgCount()); iprintf(pyc_output, indent + 1, "KW Only Arg Count: %d\n", codeObj->kwOnlyArgCount());
if (mod->verCompare(3, 11) < 0) if (mod->verCompare(3, 11) < 0)
iprintf(indent + 1, "Locals: %d\n", codeObj->numLocals()); iprintf(pyc_output, indent + 1, "Locals: %d\n", codeObj->numLocals());
if (mod->verCompare(1, 5) >= 0) if (mod->verCompare(1, 5) >= 0)
iprintf(indent + 1, "Stack Size: %d\n", codeObj->stackSize()); iprintf(pyc_output, indent + 1, "Stack Size: %d\n", codeObj->stackSize());
if (mod->verCompare(1, 3) >= 0) { if (mod->verCompare(1, 3) >= 0) {
iprintf(indent + 1, "Flags: 0x%08X", codeObj->flags()); iprintf(pyc_output, indent + 1, "Flags: 0x%08X", codeObj->flags());
print_coflags(codeObj->flags()); print_coflags(codeObj->flags(), pyc_output);
} }
iputs(indent + 1, "[Names]\n"); iputs(pyc_output, indent + 1, "[Names]\n");
for (int i=0; i<codeObj->names()->size(); i++) for (int i=0; i<codeObj->names()->size(); i++)
output_object(codeObj->names()->get(i), mod, indent + 2, flags); output_object(codeObj->names()->get(i), mod, indent + 2, flags, pyc_output);
if (mod->verCompare(1, 3) >= 0 && mod->verCompare(3, 11) < 0) { if (mod->verCompare(1, 3) >= 0 && mod->verCompare(3, 11) < 0) {
if (mod->verCompare(3, 11) >= 0) if (mod->verCompare(3, 11) >= 0)
iputs(indent + 1, "[Locals+Names]\n"); iputs(pyc_output, indent + 1, "[Locals+Names]\n");
else else
iputs(indent + 1, "[Var Names]\n"); iputs(pyc_output, indent + 1, "[Var Names]\n");
for (int i=0; i<codeObj->localNames()->size(); i++) for (int i=0; i<codeObj->localNames()->size(); i++)
output_object(codeObj->localNames()->get(i), mod, indent + 2, flags); output_object(codeObj->localNames()->get(i), mod, indent + 2, flags, pyc_output);
} }
if (mod->verCompare(3, 11) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) { if (mod->verCompare(3, 11) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) {
iputs(indent + 1, "[Locals+Kinds]\n"); iputs(pyc_output, indent + 1, "[Locals+Kinds]\n");
output_object(codeObj->localKinds().cast<PycObject>(), mod, indent + 2, flags); output_object(codeObj->localKinds().cast<PycObject>(), mod, indent + 2, flags, pyc_output);
} }
if (mod->verCompare(2, 1) >= 0 && mod->verCompare(3, 11) < 0) { if (mod->verCompare(2, 1) >= 0 && mod->verCompare(3, 11) < 0) {
iputs(indent + 1, "[Free Vars]\n"); iputs(pyc_output, indent + 1, "[Free Vars]\n");
for (int i=0; i<codeObj->freeVars()->size(); i++) for (int i=0; i<codeObj->freeVars()->size(); i++)
output_object(codeObj->freeVars()->get(i), mod, indent + 2, flags); output_object(codeObj->freeVars()->get(i), mod, indent + 2, flags, pyc_output);
iputs(indent + 1, "[Cell Vars]\n"); iputs(pyc_output, indent + 1, "[Cell Vars]\n");
for (int i=0; i<codeObj->cellVars()->size(); i++) for (int i=0; i<codeObj->cellVars()->size(); i++)
output_object(codeObj->cellVars()->get(i), mod, indent + 2, flags); output_object(codeObj->cellVars()->get(i), mod, indent + 2, flags, pyc_output);
} }
iputs(indent + 1, "[Constants]\n"); iputs(pyc_output, indent + 1, "[Constants]\n");
for (int i=0; i<codeObj->consts()->size(); i++) for (int i=0; i<codeObj->consts()->size(); i++)
output_object(codeObj->consts()->get(i), mod, indent + 2, flags); output_object(codeObj->consts()->get(i), mod, indent + 2, flags, pyc_output);
iputs(indent + 1, "[Disassembly]\n"); iputs(pyc_output, indent + 1, "[Disassembly]\n");
bc_disasm(codeObj, mod, indent + 2, flags); bc_disasm(codeObj, mod, indent + 2, flags, pyc_output);
if (mod->verCompare(1, 5) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) { if (mod->verCompare(1, 5) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) {
iputs(indent + 1, "[Line Number Table]\n"); iputs(pyc_output, indent + 1, "[Line Number Table]\n");
output_object(codeObj->lnTable().cast<PycObject>(), mod, indent + 2, flags); output_object(codeObj->lnTable().cast<PycObject>(), mod, indent + 2, flags, pyc_output);
} }
if (mod->verCompare(3, 11) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) { if (mod->verCompare(3, 11) >= 0 && (flags & Pyc::DISASM_PYCODE_VERBOSE) != 0) {
iputs(indent + 1, "[Exception Table]\n"); iputs(pyc_output, indent + 1, "[Exception Table]\n");
output_object(codeObj->exceptTable().cast<PycObject>(), mod, indent + 2, flags); output_object(codeObj->exceptTable().cast<PycObject>(), mod, indent + 2, flags, pyc_output);
} }
} }
break; break;
case PycObject::TYPE_STRING: case PycObject::TYPE_STRING:
iputs(indent, ""); iputs(pyc_output, indent, "");
OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 'b' : 0); OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 'b' : 0, false, pyc_output);
fputs("\n", pyc_output); pyc_output << "\n";
break; break;
case PycObject::TYPE_UNICODE: case PycObject::TYPE_UNICODE:
iputs(indent, ""); iputs(pyc_output, indent, "");
OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 0 : 'u'); OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 0 : 'u', false, pyc_output);
fputs("\n", pyc_output); pyc_output << "\n";
break; break;
case PycObject::TYPE_INTERNED: case PycObject::TYPE_INTERNED:
case PycObject::TYPE_ASCII: case PycObject::TYPE_ASCII:
case PycObject::TYPE_ASCII_INTERNED: case PycObject::TYPE_ASCII_INTERNED:
case PycObject::TYPE_SHORT_ASCII: case PycObject::TYPE_SHORT_ASCII:
case PycObject::TYPE_SHORT_ASCII_INTERNED: case PycObject::TYPE_SHORT_ASCII_INTERNED:
iputs(indent, ""); iputs(pyc_output, indent, "");
if (mod->majorVer() >= 3) if (mod->majorVer() >= 3)
OutputString(obj.cast<PycString>(), 0); OutputString(obj.cast<PycString>(), 0, false, pyc_output);
else else
OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 'b' : 0); OutputString(obj.cast<PycString>(), mod->strIsUnicode() ? 'b' : 0, false, pyc_output);
fputs("\n", pyc_output); pyc_output << "\n";
break; break;
case PycObject::TYPE_TUPLE: case PycObject::TYPE_TUPLE:
case PycObject::TYPE_SMALL_TUPLE: case PycObject::TYPE_SMALL_TUPLE:
{ {
iputs(indent, "(\n"); iputs(pyc_output, indent, "(\n");
for (const auto& val : obj.cast<PycTuple>()->values()) for (const auto& val : obj.cast<PycTuple>()->values())
output_object(val, mod, indent + 1, flags); output_object(val, mod, indent + 1, flags, pyc_output);
iputs(indent, ")\n"); iputs(pyc_output, indent, ")\n");
} }
break; break;
case PycObject::TYPE_LIST: case PycObject::TYPE_LIST:
{ {
iputs(indent, "[\n"); iputs(pyc_output, indent, "[\n");
for (const auto& val : obj.cast<PycList>()->values()) for (const auto& val : obj.cast<PycList>()->values())
output_object(val, mod, indent + 1, flags); output_object(val, mod, indent + 1, flags, pyc_output);
iputs(indent, "]\n"); iputs(pyc_output, indent, "]\n");
} }
break; break;
case PycObject::TYPE_DICT: case PycObject::TYPE_DICT:
{ {
iputs(indent, "{\n"); iputs(pyc_output, indent, "{\n");
PycDict::key_t keys = obj.cast<PycDict>()->keys(); PycDict::key_t keys = obj.cast<PycDict>()->keys();
PycDict::value_t values = obj.cast<PycDict>()->values(); PycDict::value_t values = obj.cast<PycDict>()->values();
PycDict::key_t::const_iterator ki = keys.begin(); PycDict::key_t::const_iterator ki = keys.begin();
PycDict::value_t::const_iterator vi = values.begin(); PycDict::value_t::const_iterator vi = values.begin();
while (ki != keys.end()) { while (ki != keys.end()) {
output_object(*ki, mod, indent + 1, flags); output_object(*ki, mod, indent + 1, flags, pyc_output);
output_object(*vi, mod, indent + 2, flags); output_object(*vi, mod, indent + 2, flags, pyc_output);
++ki, ++vi; ++ki, ++vi;
} }
iputs(indent, "}\n"); iputs(pyc_output, indent, "}\n");
} }
break; break;
case PycObject::TYPE_SET: case PycObject::TYPE_SET:
{ {
iputs(indent, "{\n"); iputs(pyc_output, indent, "{\n");
for (const auto& val : obj.cast<PycSet>()->values()) for (const auto& val : obj.cast<PycSet>()->values())
output_object(val, mod, indent + 1, flags); output_object(val, mod, indent + 1, flags, pyc_output);
iputs(indent, "}\n"); iputs(pyc_output, indent, "}\n");
} }
break; break;
case PycObject::TYPE_FROZENSET: case PycObject::TYPE_FROZENSET:
{ {
iputs(indent, "frozenset({\n"); iputs(pyc_output, indent, "frozenset({\n");
for (const auto& val : obj.cast<PycSet>()->values()) for (const auto& val : obj.cast<PycSet>()->values())
output_object(val, mod, indent + 1, flags); output_object(val, mod, indent + 1, flags, pyc_output);
iputs(indent, "})\n"); iputs(pyc_output, indent, "})\n");
} }
break; break;
case PycObject::TYPE_NONE: case PycObject::TYPE_NONE:
iputs(indent, "None\n"); iputs(pyc_output, indent, "None\n");
break; break;
case PycObject::TYPE_FALSE: case PycObject::TYPE_FALSE:
iputs(indent, "False\n"); iputs(pyc_output, indent, "False\n");
break; break;
case PycObject::TYPE_TRUE: case PycObject::TYPE_TRUE:
iputs(indent, "True\n"); iputs(pyc_output, indent, "True\n");
break; break;
case PycObject::TYPE_ELLIPSIS: case PycObject::TYPE_ELLIPSIS:
iputs(indent, "...\n"); iputs(pyc_output, indent, "...\n");
break; break;
case PycObject::TYPE_INT: case PycObject::TYPE_INT:
iprintf(indent, "%d\n", obj.cast<PycInt>()->value()); iprintf(pyc_output, indent, "%d\n", obj.cast<PycInt>()->value());
break; break;
case PycObject::TYPE_LONG: case PycObject::TYPE_LONG:
iprintf(indent, "%s\n", obj.cast<PycLong>()->repr().c_str()); iprintf(pyc_output, indent, "%s\n", obj.cast<PycLong>()->repr().c_str());
break; break;
case PycObject::TYPE_FLOAT: case PycObject::TYPE_FLOAT:
iprintf(indent, "%s\n", obj.cast<PycFloat>()->value()); iprintf(pyc_output, indent, "%s\n", obj.cast<PycFloat>()->value());
break; break;
case PycObject::TYPE_COMPLEX: case PycObject::TYPE_COMPLEX:
iprintf(indent, "(%s+%sj)\n", obj.cast<PycComplex>()->value(), iprintf(pyc_output, indent, "(%s+%sj)\n", obj.cast<PycComplex>()->value(),
obj.cast<PycComplex>()->imag()); obj.cast<PycComplex>()->imag());
break; break;
case PycObject::TYPE_BINARY_FLOAT: case PycObject::TYPE_BINARY_FLOAT:
iprintf(indent, "%g\n", obj.cast<PycCFloat>()->value()); iprintf(pyc_output, indent, "%g\n", obj.cast<PycCFloat>()->value());
break; break;
case PycObject::TYPE_BINARY_COMPLEX: case PycObject::TYPE_BINARY_COMPLEX:
iprintf(indent, "(%g+%gj)\n", obj.cast<PycCComplex>()->value(), iprintf(pyc_output, indent, "(%g+%gj)\n", obj.cast<PycCComplex>()->value(),
obj.cast<PycCComplex>()->imag()); obj.cast<PycCComplex>()->imag());
break; break;
default: default:
iprintf(indent, "<TYPE: %d>\n", obj->type()); iprintf(pyc_output, indent, "<TYPE: %d>\n", obj->type());
} }
} }
@@ -262,18 +264,20 @@ int main(int argc, char* argv[])
bool marshalled = false; bool marshalled = false;
const char* version = nullptr; const char* version = nullptr;
unsigned disasm_flags = 0; unsigned disasm_flags = 0;
std::ostream &pyc_output = std::cout;
for (int arg = 1; arg < argc; ++arg) { for (int arg = 1; arg < argc; ++arg) {
if (strcmp(argv[arg], "-o") == 0) { if (strcmp(argv[arg], "-o") == 0) {
if (arg + 1 < argc) { if (arg + 1 < argc) {
const char* filename = argv[++arg]; const char* filename = argv[++arg];
FILE* outfile = fopen(filename, "w");
if (!outfile) { auto* outfile = new std::filebuf;
if(! outfile->open(filename, std::ios::out)) {
fprintf(stderr, "Error opening file '%s' for writing\n", fprintf(stderr, "Error opening file '%s' for writing\n",
argv[arg]); argv[arg]);
return 1; return 1;
} }
pyc_output = outfile; pyc_output.rdbuf(outfile);
} else { } else {
fputs("Option '-o' requires a filename\n", stderr); fputs("Option '-o' requires a filename\n", stderr);
return 1; return 1;
@@ -339,10 +343,10 @@ int main(int argc, char* argv[])
} }
const char* dispname = strrchr(infile, PATHSEP); const char* dispname = strrchr(infile, PATHSEP);
dispname = (dispname == NULL) ? infile : dispname + 1; dispname = (dispname == NULL) ? infile : dispname + 1;
fprintf(pyc_output, "%s (Python %d.%d%s)\n", dispname, mod.majorVer(), mod.minorVer(), formatted_print(pyc_output, "%s (Python %d.%d%s)\n", dispname, mod.majorVer(), mod.minorVer(),
(mod.majorVer() < 3 && mod.isUnicode()) ? " -U" : ""); (mod.majorVer() < 3 && mod.isUnicode()) ? " -U" : "");
try { try {
output_object(mod.code().try_cast<PycObject>(), &mod, 0, disasm_flags); output_object(mod.code().try_cast<PycObject>(), &mod, 0, disasm_flags, pyc_output);
} catch (std::exception& ex) { } catch (std::exception& ex) {
fprintf(stderr, "Error disassembling %s: %s\n", infile, ex.what()); fprintf(stderr, "Error disassembling %s: %s\n", infile, ex.what());
return 1; return 1;

16
pycdc.cpp
View File

@@ -1,4 +1,6 @@
#include <cstring> #include <cstring>
#include <fstream>
#include <iostream>
#include "ASTree.h" #include "ASTree.h"
#ifdef WIN32 #ifdef WIN32
@@ -12,17 +14,19 @@ int main(int argc, char* argv[])
const char* infile = nullptr; const char* infile = nullptr;
bool marshalled = false; bool marshalled = false;
const char* version = nullptr; const char* version = nullptr;
std::ostream &pyc_output = std::cout;
for (int arg = 1; arg < argc; ++arg) { for (int arg = 1; arg < argc; ++arg) {
if (strcmp(argv[arg], "-o") == 0) { if (strcmp(argv[arg], "-o") == 0) {
if (arg + 1 < argc) { if (arg + 1 < argc) {
const char* filename = argv[++arg]; const char* filename = argv[++arg];
FILE* outfile = fopen(filename, "w"); auto* outfile = new std::filebuf;
if (!outfile) { if(! outfile->open(filename, std::ios::out)) {
fprintf(stderr, "Error opening file '%s' for writing\n", fprintf(stderr, "Error opening file '%s' for writing\n",
argv[arg]); argv[arg]);
return 1; return 1;
} }
pyc_output = outfile; pyc_output.rdbuf(outfile);
} else { } else {
fputs("Option '-o' requires a filename\n", stderr); fputs("Option '-o' requires a filename\n", stderr);
return 1; return 1;
@@ -84,11 +88,11 @@ int main(int argc, char* argv[])
} }
const char* dispname = strrchr(infile, PATHSEP); const char* dispname = strrchr(infile, PATHSEP);
dispname = (dispname == NULL) ? infile : dispname + 1; dispname = (dispname == NULL) ? infile : dispname + 1;
fputs("# Source Generated with Decompyle++\n", pyc_output); std::cout << "# Source Generated with Decompyle++\n";
fprintf(pyc_output, "# File: %s (Python %d.%d%s)\n\n", dispname, mod.majorVer(), mod.minorVer(), formatted_print(pyc_output, "# File: %s (Python %d.%d%s)\n\n", dispname, mod.majorVer(), mod.minorVer(),
(mod.majorVer() < 3 && mod.isUnicode()) ? " Unicode" : ""); (mod.majorVer() < 3 && mod.isUnicode()) ? " Unicode" : "");
try { try {
decompyle(mod.code(), &mod); decompyle(mod.code(), &mod, pyc_output);
} catch (std::exception& ex) { } catch (std::exception& ex) {
fprintf(stderr, "Error decompyling %s: %s\n", infile, ex.what()); fprintf(stderr, "Error decompyling %s: %s\n", infile, ex.what());
return 1; return 1;