cpython/Python/compile.c

/*
 * This file compiles an abstract syntax tree (AST) into Python bytecode.
 *
 * The primary entry point is PyAST_Compile(), which returns a
 * PyCodeObject.  The compiler makes several passes to build the code
 * object:
 *   1. Checks for future statements.  See future.c
 *   2. Builds a symbol table.	See symtable.c.
 *   3. Generate code for basic blocks.	 See compiler_mod() in this file.
 *   4. Assemble the basic blocks into final code.  See assemble() in
 *   this file.	 
 *
 * Note that compiler_mod() suggests module, but the module ast type
 * (mod_ty) has cases for expressions and interactive statements.
 *
 * CAUTION: The VISIT_* macros abort the current function when they
 * encounter a problem. So don't invoke them when there is memory
 * which needs to be released. Code blocks are OK, as the compiler
 * structure takes care of releasing those.
 */

#include "Python.h"

#include "Python-ast.h"
#include "node.h"
#include "pyarena.h"
#include "ast.h"
#include "code.h"
#include "compile.h"
#include "symtable.h"
#include "opcode.h"

int Py_OptimizeFlag = 0;

/*
  ISSUES:

  opcode_stack_effect() function should be reviewed since stack depth bugs
  could be really hard to find later.

  Dead code is being generated (i.e. after unconditional jumps).
    XXX(nnorwitz): not sure this is still true
*/

#define DEFAULT_BLOCK_SIZE 16
#define DEFAULT_BLOCKS 8
#define DEFAULT_CODE_SIZE 128
#define DEFAULT_LNOTAB_SIZE 16

struct instr {
	unsigned i_jabs : 1;
	unsigned i_jrel : 1;
	unsigned i_hasarg : 1;
	unsigned char i_opcode;
	int i_oparg;
	struct basicblock_ *i_target; /* target block (if jump instruction) */
	int i_lineno;
};

typedef struct basicblock_ {
    /* Each basicblock in a compilation unit is linked via b_list in the
       reverse order that the block are allocated.  b_list points to the next
       block, not to be confused with b_next, which is next by control flow. */
	struct basicblock_ *b_list;
	/* number of instructions used */
	int b_iused;
	/* length of instruction array (b_instr) */
	int b_ialloc;
	/* pointer to an array of instructions, initially NULL */
	struct instr *b_instr;
	/* If b_next is non-NULL, it is a pointer to the next
	   block reached by normal control flow. */
	struct basicblock_ *b_next;
	/* b_seen is used to perform a DFS of basicblocks. */
	unsigned b_seen : 1;
	/* b_return is true if a RETURN_VALUE opcode is inserted. */
	unsigned b_return : 1;
	/* depth of stack upon entry of block, computed by stackdepth() */
	int b_startdepth;
	/* instruction offset for block, computed by assemble_jump_offsets() */
	int b_offset;
} basicblock;

/* fblockinfo tracks the current frame block.

A frame block is used to handle loops, try/except, and try/finally.
It's called a frame block to distinguish it from a basic block in the
compiler IR.
*/

enum fblocktype { LOOP, EXCEPT, FINALLY_TRY, FINALLY_END };

struct fblockinfo {
	enum fblocktype fb_type;
	basicblock *fb_block;
};

/* The following items change on entry and exit of code blocks.
   They must be saved and restored when returning to a block.
*/
struct compiler_unit {
	PySTEntryObject *u_ste;

	PyObject *u_name;
	/* The following fields are dicts that map objects to
	   the index of them in co_XXX.	 The index is used as
	   the argument for opcodes that refer to those collections.
	*/
	PyObject *u_consts;    /* all constants */
	PyObject *u_names;     /* all names */
	PyObject *u_varnames;  /* local variables */
	PyObject *u_cellvars;  /* cell variables */
	PyObject *u_freevars;  /* free variables */

	PyObject *u_private;	/* for private name mangling */

	int u_argcount;	   /* number of arguments for block */ 
    /* Pointer to the most recently allocated block.  By following b_list
       members, you can reach all early allocated blocks. */
	basicblock *u_blocks;
	basicblock *u_curblock; /* pointer to current block */
	int u_tmpname;	   /* temporary variables for list comps */

	int u_nfblocks;
	struct fblockinfo u_fblock[CO_MAXBLOCKS];

	int u_firstlineno; /* the first lineno of the block */
	int u_lineno;	   /* the lineno for the current stmt */
	bool u_lineno_set; /* boolean to indicate whether instr
			      has been generated with current lineno */
};

/* This struct captures the global state of a compilation.  

The u pointer points to the current compilation unit, while units
for enclosing blocks are stored in c_stack.	The u and c_stack are
managed by compiler_enter_scope() and compiler_exit_scope().
*/

struct compiler {
	const char *c_filename;
	struct symtable *c_st;
	PyFutureFeatures *c_future; /* pointer to module's __future__ */
	PyCompilerFlags *c_flags;

	int c_interactive;	 /* true if in interactive mode */
	int c_nestlevel;

	struct compiler_unit *u; /* compiler state for current block */
	PyObject *c_stack;	 /* Python list holding compiler_unit ptrs */
	char *c_encoding;	 /* source encoding (a borrowed reference) */
	PyArena *c_arena;	 /* pointer to memory allocation arena */
};

struct assembler {
	PyObject *a_bytecode;  /* string containing bytecode */
	int a_offset;	       /* offset into bytecode */
	int a_nblocks;	       /* number of reachable blocks */
	basicblock **a_postorder; /* list of blocks in dfs postorder */
	PyObject *a_lnotab;    /* string containing lnotab */
	int a_lnotab_off;      /* offset into lnotab */
	int a_lineno;	       /* last lineno of emitted instruction */
	int a_lineno_off;      /* bytecode offset of last lineno */
};

static int compiler_enter_scope(struct compiler *, identifier, void *, int);
static void compiler_free(struct compiler *);
static basicblock *compiler_new_block(struct compiler *);
static int compiler_next_instr(struct compiler *, basicblock *);
static int compiler_addop(struct compiler *, int);
static int compiler_addop_o(struct compiler *, int, PyObject *, PyObject *);
static int compiler_addop_i(struct compiler *, int, int);
static int compiler_addop_j(struct compiler *, int, basicblock *, int);
static basicblock *compiler_use_new_block(struct compiler *);
static int compiler_error(struct compiler *, const char *);
static int compiler_nameop(struct compiler *, identifier, expr_context_ty);

static PyCodeObject *compiler_mod(struct compiler *, mod_ty);
static int compiler_visit_stmt(struct compiler *, stmt_ty);
static int compiler_visit_keyword(struct compiler *, keyword_ty);
static int compiler_visit_expr(struct compiler *, expr_ty);
static int compiler_augassign(struct compiler *, stmt_ty);
static int compiler_visit_slice(struct compiler *, slice_ty,
				expr_context_ty);

static int compiler_push_fblock(struct compiler *, enum fblocktype,
				basicblock *);
static void compiler_pop_fblock(struct compiler *, enum fblocktype,
				basicblock *);

static int inplace_binop(struct compiler *, operator_ty);
static int expr_constant(expr_ty e);

static int compiler_with(struct compiler *, stmt_ty);

static PyCodeObject *assemble(struct compiler *, int addNone);
static PyObject *__doc__;

PyObject *
_Py_Mangle(PyObject *privateobj, PyObject *ident)
{
	/* Name mangling: __private becomes _classname__private.
	   This is independent from how the name is used. */
	const char *p, *name = PyString_AsString(ident);
	char *buffer;
	size_t nlen, plen;
	if (privateobj == NULL || !PyString_Check(privateobj) ||
	    name == NULL || name[0] != '_' || name[1] != '_') {
		Py_INCREF(ident);
		return ident;
	}
	p = PyString_AsString(privateobj);
	nlen = strlen(name);
	if (name[nlen-1] == '_' && name[nlen-2] == '_') {
		Py_INCREF(ident);
		return ident; /* Don't mangle __whatever__ */
	}
	/* Strip leading underscores from class name */
	while (*p == '_')
		p++;
	if (*p == '\0') {
		Py_INCREF(ident);
		return ident; /* Don't mangle if class is just underscores */
	}
	plen = strlen(p);
	ident = PyString_FromStringAndSize(NULL, 1 + nlen + plen);
	if (!ident)
		return 0;
	/* ident = "_" + p[:plen] + name # i.e. 1+plen+nlen bytes */
	buffer = PyString_AS_STRING(ident);
	buffer[0] = '_';
	strncpy(buffer+1, p, plen);
	strcpy(buffer+1+plen, name);
	return ident;
}

static int
compiler_init(struct compiler *c)
{
	memset(c, 0, sizeof(struct compiler));

	c->c_stack = PyList_New(0);
	if (!c->c_stack)
		return 0;

	return 1;
}

PyCodeObject *
PyAST_Compile(mod_ty mod, const char *filename, PyCompilerFlags *flags,
	      PyArena *arena)
{
	struct compiler c;
	PyCodeObject *co = NULL;
	PyCompilerFlags local_flags;
	int merged;

	if (!__doc__) {
		__doc__ = PyString_InternFromString("__doc__");
		if (!__doc__)
			return NULL;
	}

	if (!compiler_init(&c))
		return NULL;
	c.c_filename = filename;
	c.c_arena = arena;
	c.c_future = PyFuture_FromAST(mod, filename);
	if (c.c_future == NULL)
		goto finally;
	if (!flags) {
		local_flags.cf_flags = 0;
		flags = &local_flags;
	}
	merged = c.c_future->ff_features | flags->cf_flags;
	c.c_future->ff_features = merged;
	flags->cf_flags = merged;
	c.c_flags = flags;
	c.c_nestlevel = 0;

	c.c_st = PySymtable_Build(mod, filename, c.c_future);
	if (c.c_st == NULL) {
		if (!PyErr_Occurred())
			PyErr_SetString(PyExc_SystemError, "no symtable");
		goto finally;
	}

	/* XXX initialize to NULL for now, need to handle */
	c.c_encoding = NULL;

	co = compiler_mod(&c, mod);

 finally:
	compiler_free(&c);
	assert(co || PyErr_Occurred());
	return co;
}

PyCodeObject *
PyNode_Compile(struct _node *n, const char *filename)
{
	PyCodeObject *co = NULL;
	mod_ty mod;
	PyArena *arena = PyArena_New();
	if (!arena)
		return NULL;
	mod = PyAST_FromNode(n, NULL, filename, arena);
	if (mod)
		co = PyAST_Compile(mod, filename, NULL, arena);
	PyArena_Free(arena);
	return co;
}

static void
compiler_free(struct compiler *c)
{
	if (c->c_st)
		PySymtable_Free(c->c_st);
	if (c->c_future)
		PyObject_Free(c->c_future);
	Py_DECREF(c->c_stack);
}

static PyObject *
list2dict(PyObject *list)
{
	Py_ssize_t i, n;
	PyObject *v, *k;
	PyObject *dict = PyDict_New();
	if (!dict) return NULL;

	n = PyList_Size(list);
	for (i = 0; i < n; i++) {
		v = PyInt_FromLong(i);
		if (!v) {
			Py_DECREF(dict);
			return NULL;
		}
		k = PyList_GET_ITEM(list, i);
		k = PyTuple_Pack(2, k, k->ob_type);
		if (k == NULL || PyDict_SetItem(dict, k, v) < 0) {
			Py_XDECREF(k);
			Py_DECREF(v);
			Py_DECREF(dict);
			return NULL;
		}
		Py_DECREF(k);
		Py_DECREF(v);
	}
	return dict;
}

/* Return new dict containing names from src that match scope(s).

src is a symbol table dictionary.  If the scope of a name matches
either scope_type or flag is set, insert it into the new dict.	The
values are integers, starting at offset and increasing by one for
each key.
*/

static PyObject *
dictbytype(PyObject *src, int scope_type, int flag, int offset)
{
	Py_ssize_t pos = 0, i = offset, scope;
	PyObject *k, *v, *dest = PyDict_New();

	assert(offset >= 0);
	if (dest == NULL)
		return NULL;

	while (PyDict_Next(src, &pos, &k, &v)) {
		/* XXX this should probably be a macro in symtable.h */
		assert(PyInt_Check(v));
		scope = (PyInt_AS_LONG(v) >> SCOPE_OFF) & SCOPE_MASK;

		if (scope == scope_type || PyInt_AS_LONG(v) & flag) {
			PyObject *tuple, *item = PyInt_FromLong(i);
			if (item == NULL) {
				Py_DECREF(dest);
				return NULL;
			}
			i++;
			tuple = PyTuple_Pack(2, k, k->ob_type);
			if (!tuple || PyDict_SetItem(dest, tuple, item) < 0) {
				Py_DECREF(item);
				Py_DECREF(dest);
				Py_XDECREF(tuple);
				return NULL;
			}
			Py_DECREF(item);
			Py_DECREF(tuple);
		}
	}
	return dest;
}

/* Begin: Peephole optimizations ----------------------------------------- */

#define GETARG(arr, i) ((int)((arr[i+2]<<8) + arr[i+1]))
#define UNCONDITIONAL_JUMP(op)	(op==JUMP_ABSOLUTE || op==JUMP_FORWARD)
#define ABSOLUTE_JUMP(op) (op==JUMP_ABSOLUTE || op==CONTINUE_LOOP)
#define GETJUMPTGT(arr, i) (GETARG(arr,i) + (ABSOLUTE_JUMP(arr[i]) ? 0 : i+3))
#define SETARG(arr, i, val) arr[i+2] = val>>8; arr[i+1] = val & 255
#define CODESIZE(op)  (HAS_ARG(op) ? 3 : 1)
#define ISBASICBLOCK(blocks, start, bytes) \
	(blocks[start]==blocks[start+bytes-1])

/* Replace LOAD_CONST c1. LOAD_CONST c2 ... LOAD_CONST cn BUILD_TUPLE n
   with	   LOAD_CONST (c1, c2, ... cn).
   The consts table must still be in list form so that the
   new constant (c1, c2, ... cn) can be appended.
   Called with codestr pointing to the first LOAD_CONST.
   Bails out with no change if one or more of the LOAD_CONSTs is missing. 
   Also works for BUILD_LIST when followed by an "in" or "not in" test.
*/
static int
tuple_of_constants(unsigned char *codestr, int n, PyObject *consts)
{
	PyObject *newconst, *constant;
	Py_ssize_t i, arg, len_consts;

	/* Pre-conditions */
	assert(PyList_CheckExact(consts));
	assert(codestr[n*3] == BUILD_TUPLE || codestr[n*3] == BUILD_LIST);
	assert(GETARG(codestr, (n*3)) == n);
	for (i=0 ; i<n ; i++)
		assert(codestr[i*3] == LOAD_CONST);

	/* Buildup new tuple of constants */
	newconst = PyTuple_New(n);
	if (newconst == NULL)
		return 0;
	len_consts = PyList_GET_SIZE(consts);
	for (i=0 ; i<n ; i++) {
		arg = GETARG(codestr, (i*3));
		assert(arg < len_consts);
		constant = PyList_GET_ITEM(consts, arg);
		Py_INCREF(constant);
		PyTuple_SET_ITEM(newconst, i, constant);
	}

	/* Append folded constant onto consts */
	if (PyList_Append(consts, newconst)) {
		Py_DECREF(newconst);
		return 0;
	}
	Py_DECREF(newconst);

	/* Write NOPs over old LOAD_CONSTS and
	   add a new LOAD_CONST newconst on top of the BUILD_TUPLE n */
	memset(codestr, NOP, n*3);
	codestr[n*3] = LOAD_CONST;
	SETARG(codestr, (n*3), len_consts);
	return 1;
}

/* Replace LOAD_CONST c1. LOAD_CONST c2 BINOP
   with	   LOAD_CONST binop(c1,c2)
   The consts table must still be in list form so that the
   new constant can be appended.
   Called with codestr pointing to the first LOAD_CONST. 
   Abandons the transformation if the folding fails (i.e.  1+'a').  
   If the new constant is a sequence, only folds when the size
   is below a threshold value.	That keeps pyc files from
   becoming large in the presence of code like:	 (None,)*1000.
*/
static int
fold_binops_on_constants(unsigned char *codestr, PyObject *consts)
{
	PyObject *newconst, *v, *w;
	Py_ssize_t len_consts, size;
	int opcode;

	/* Pre-conditions */
	assert(PyList_CheckExact(consts));
	assert(codestr[0] == LOAD_CONST);
	assert(codestr[3] == LOAD_CONST);

	/* Create new constant */
	v = PyList_GET_ITEM(consts, GETARG(codestr, 0));
	w = PyList_GET_ITEM(consts, GETARG(codestr, 3));
	opcode = codestr[6];
	switch (opcode) {
		case BINARY_POWER:
			newconst = PyNumber_Power(v, w, Py_None);
			break;
		case BINARY_MULTIPLY:
			newconst = PyNumber_Multiply(v, w);
			break;
		case BINARY_DIVIDE:
			/* Cannot fold this operation statically since
                           the result can depend on the run-time presence
                           of the -Qnew flag */
			return 0;
		case BINARY_TRUE_DIVIDE:
			newconst = PyNumber_TrueDivide(v, w);
			break;
		case BINARY_FLOOR_DIVIDE:
			newconst = PyNumber_FloorDivide(v, w);
			break;
		case BINARY_MODULO:
			newconst = PyNumber_Remainder(v, w);
			break;
		case BINARY_ADD:
			newconst = PyNumber_Add(v, w);
			break;
		case BINARY_SUBTRACT:
			newconst = PyNumber_Subtract(v, w);
			break;
		case BINARY_SUBSCR:
			newconst = PyObject_GetItem(v, w);
			break;
		case BINARY_LSHIFT:
			newconst = PyNumber_Lshift(v, w);
			break;
		case BINARY_RSHIFT:
			newconst = PyNumber_Rshift(v, w);
			break;
		case BINARY_AND:
			newconst = PyNumber_And(v, w);
			break;
		case BINARY_XOR:
			newconst = PyNumber_Xor(v, w);
			break;
		case BINARY_OR:
			newconst = PyNumber_Or(v, w);
			break;
		default:
			/* Called with an unknown opcode */
			PyErr_Format(PyExc_SystemError,
			     "unexpected binary operation %d on a constant",
				     opcode);
			return 0;
	}
	if (newconst == NULL) {
		PyErr_Clear();
		return 0;
	}
	size = PyObject_Size(newconst);
	if (size == -1)
		PyErr_Clear();
	else if (size > 20) {
		Py_DECREF(newconst);
		return 0;
	}

	/* Append folded constant into consts table */
	len_consts = PyList_GET_SIZE(consts);
	if (PyList_Append(consts, newconst)) {
		Py_DECREF(newconst);
		return 0;
	}
	Py_DECREF(newconst);

	/* Write NOP NOP NOP NOP LOAD_CONST newconst */
	memset(codestr, NOP, 4);
	codestr[4] = LOAD_CONST;
	SETARG(codestr, 4, len_consts);
	return 1;
}

static int
fold_unaryops_on_constants(unsigned char *codestr, PyObject *consts)
{
	PyObject *newconst=NULL, *v;
	Py_ssize_t len_consts;
	int opcode;

	/* Pre-conditions */
	assert(PyList_CheckExact(consts));
	assert(codestr[0] == LOAD_CONST);

	/* Create new constant */
	v = PyList_GET_ITEM(consts, GETARG(codestr, 0));
	opcode = codestr[3];
	switch (opcode) {
		case UNARY_NEGATIVE:
			/* Preserve the sign of -0.0 */
			if (PyObject_IsTrue(v) == 1)
				newconst = PyNumber_Negative(v);
			break;
		case UNARY_CONVERT:
			newconst = PyObject_Repr(v);
			break;
		case UNARY_INVERT:
			newconst = PyNumber_Invert(v);
			break;
		default:
			/* Called with an unknown opcode */
			PyErr_Format(PyExc_SystemError,
			     "unexpected unary operation %d on a constant",
				     opcode);
			return 0;
	}
	if (newconst == NULL) {
		PyErr_Clear();
		return 0;
	}

	/* Append folded constant into consts table */
	len_consts = PyList_GET_SIZE(consts);
	if (PyList_Append(consts, newconst)) {
		Py_DECREF(newconst);
		return 0;
	}
	Py_DECREF(newconst);

	/* Write NOP LOAD_CONST newconst */
	codestr[0] = NOP;
	codestr[1] = LOAD_CONST;
	SETARG(codestr, 1, len_consts);
	return 1;
}

static unsigned int *
markblocks(unsigned char *code, int len)
{
	unsigned int *blocks = (unsigned int *)PyMem_Malloc(len*sizeof(int));
	int i,j, opcode, blockcnt = 0;

	if (blocks == NULL) {
		PyErr_NoMemory();
		return NULL;
	}
	memset(blocks, 0, len*sizeof(int));

	/* Mark labels in the first pass */
	for (i=0 ; i<len ; i+=CODESIZE(opcode)) {
		opcode = code[i];
		switch (opcode) {
			case FOR_ITER:
			case JUMP_FORWARD:
			case JUMP_IF_FALSE:
			case JUMP_IF_TRUE:
			case JUMP_ABSOLUTE:
			case CONTINUE_LOOP:
			case SETUP_LOOP:
			case SETUP_EXCEPT:
			case SETUP_FINALLY:
				j = GETJUMPTGT(code, i);
				blocks[j] = 1;
				break;
		}
	}
	/* Build block numbers in the second pass */
	for (i=0 ; i<len ; i++) {
		blockcnt += blocks[i];	/* increment blockcnt over labels */
		blocks[i] = blockcnt;
	}
	return blocks;
}

/* Perform basic peephole optimizations to components of a code object.
   The consts object should still be in list form to allow new constants 
   to be appended.

   To keep the optimizer simple, it bails out (does nothing) for code
   containing extended arguments or that has a length over 32,700.  That 
   allows us to avoid overflow and sign issues.	 Likewise, it bails when
   the lineno table has complex encoding for gaps >= 255.

   Optimizations are restricted to simple transformations occuring within a
   single basic block.	All transformations keep the code size the same or 
   smaller.  For those that reduce size, the gaps are initially filled with 
   NOPs.  Later those NOPs are removed and the jump addresses retargeted in 
   a single pass.  Line numbering is adjusted accordingly. */

static PyObject *
optimize_code(PyObject *code, PyObject* consts, PyObject *names,
              PyObject *lineno_obj)
{
	Py_ssize_t i, j, codelen;
	int nops, h, adj;
	int tgt, tgttgt, opcode;
	unsigned char *codestr = NULL;
	unsigned char *lineno;
	int *addrmap = NULL;
	int new_line, cum_orig_line, last_line, tabsiz;
	int cumlc=0, lastlc=0;	/* Count runs of consecutive LOAD_CONSTs */
	unsigned int *blocks = NULL;
	char *name;

	/* Bail out if an exception is set */
	if (PyErr_Occurred())
		goto exitUnchanged;

	/* Bypass optimization when the lineno table is too complex */
	assert(PyString_Check(lineno_obj));
	lineno = (unsigned char*)PyString_AS_STRING(lineno_obj);
	tabsiz = PyString_GET_SIZE(lineno_obj);
	if (memchr(lineno, 255, tabsiz) != NULL)
		goto exitUnchanged;

	/* Avoid situations where jump retargeting could overflow */
	assert(PyString_Check(code));
	codelen = PyString_Size(code);
	if (codelen > 32700)
		goto exitUnchanged;

	/* Make a modifiable copy of the code string */
	codestr = (unsigned char *)PyMem_Malloc(codelen);
	if (codestr == NULL)
		goto exitUnchanged;
	codestr = (unsigned char *)memcpy(codestr, 
                                        PyString_AS_STRING(code), codelen);

	/* Verify that RETURN_VALUE terminates the codestring.	This allows
	   the various transformation patterns to look ahead several
	   instructions without additional checks to make sure they are not
	   looking beyond the end of the code string.
	*/
	if (codestr[codelen-1] != RETURN_VALUE)
		goto exitUnchanged;

	/* Mapping to new jump targets after NOPs are removed */
	addrmap = (int *)PyMem_Malloc(codelen * sizeof(int));
	if (addrmap == NULL)
		goto exitUnchanged;

	blocks = markblocks(codestr, codelen);
	if (blocks == NULL)
		goto exitUnchanged;
	assert(PyList_Check(consts));

	for (i=0 ; i<codelen ; i += CODESIZE(codestr[i])) {
		opcode = codestr[i];

		lastlc = cumlc;
		cumlc = 0;

		switch (opcode) {

			/* Replace UNARY_NOT JUMP_IF_FALSE POP_TOP with 
			   with	   JUMP_IF_TRUE POP_TOP */
			case UNARY_NOT:
				if (codestr[i+1] != JUMP_IF_FALSE  ||
				    codestr[i+4] != POP_TOP  ||
				    !ISBASICBLOCK(blocks,i,5))
					continue;
				tgt = GETJUMPTGT(codestr, (i+1));
				if (codestr[tgt] != POP_TOP)
					continue;
				j = GETARG(codestr, i+1) + 1;
				codestr[i] = JUMP_IF_TRUE;
				SETARG(codestr, i, j);
				codestr[i+3] = POP_TOP;
				codestr[i+4] = NOP;
				break;

				/* not a is b -->  a is not b
				   not a in b -->  a not in b
				   not a is not b -->  a is b
				   not a not in b -->  a in b
				*/
			case COMPARE_OP:
				j = GETARG(codestr, i);
				if (j < 6  ||  j > 9  ||
				    codestr[i+3] != UNARY_NOT  || 
				    !ISBASICBLOCK(blocks,i,4))
					continue;
				SETARG(codestr, i, (j^1));
				codestr[i+3] = NOP;
				break;

				/* Replace LOAD_GLOBAL/LOAD_NAME None
                                   with LOAD_CONST None */
			case LOAD_NAME:
			case LOAD_GLOBAL:
				j = GETARG(codestr, i);
				name = PyString_AsString(PyTuple_GET_ITEM(names, j));
				if (name == NULL  ||  strcmp(name, "None") != 0)
					continue;
				for (j=0 ; j < PyList_GET_SIZE(consts) ; j++) {
					if (PyList_GET_ITEM(consts, j) == Py_None) {
						codestr[i] = LOAD_CONST;
						SETARG(codestr, i, j);
						cumlc = lastlc + 1;
						break;
					}
				}
				break;

				/* Skip over LOAD_CONST trueconst
                                   JUMP_IF_FALSE xx  POP_TOP */
			case LOAD_CONST:
				cumlc = lastlc + 1;
				j = GETARG(codestr, i);
				if (codestr[i+3] != JUMP_IF_FALSE  ||
				    codestr[i+6] != POP_TOP  ||
				    !ISBASICBLOCK(blocks,i,7)  ||
				    !PyObject_IsTrue(PyList_GET_ITEM(consts, j)))
					continue;
				memset(codestr+i, NOP, 7);
				cumlc = 0;
				break;

				/* Try to fold tuples of constants (includes a case for lists
				   which are only used for "in" and "not in" tests).
				   Skip over BUILD_SEQN 1 UNPACK_SEQN 1.
				   Replace BUILD_SEQN 2 UNPACK_SEQN 2 with ROT2.
				   Replace BUILD_SEQN 3 UNPACK_SEQN 3 with ROT3 ROT2. */
			case BUILD_TUPLE:
			case BUILD_LIST:
				j = GETARG(codestr, i);
				h = i - 3 * j;
				if (h >= 0  &&
				    j <= lastlc	 &&
				    ((opcode == BUILD_TUPLE && 
				      ISBASICBLOCK(blocks, h, 3*(j+1))) ||
				     (opcode == BUILD_LIST && 
				      codestr[i+3]==COMPARE_OP && 
				      ISBASICBLOCK(blocks, h, 3*(j+2)) &&
				      (GETARG(codestr,i+3)==6 ||
				       GETARG(codestr,i+3)==7))) &&
				    tuple_of_constants(&codestr[h], j, consts)) {
					assert(codestr[i] == LOAD_CONST);
					cumlc = 1;
					break;
				}
				if (codestr[i+3] != UNPACK_SEQUENCE  ||
				    !ISBASICBLOCK(blocks,i,6) ||
				    j != GETARG(codestr, i+3))
					continue;
				if (j == 1) {
					memset(codestr+i, NOP, 6);
				} else if (j == 2) {
					codestr[i] = ROT_TWO;
					memset(codestr+i+1, NOP, 5);
				} else if (j == 3) {
					codestr[i] = ROT_THREE;
					codestr[i+1] = ROT_TWO;
					memset(codestr+i+2, NOP, 4);
				}
				break;

				/* Fold binary ops on constants.
				   LOAD_CONST c1 LOAD_CONST c2 BINOP -->  LOAD_CONST binop(c1,c2) */
			case BINARY_POWER:
			case BINARY_MULTIPLY:
			case BINARY_TRUE_DIVIDE:
			case BINARY_FLOOR_DIVIDE:
			case BINARY_MODULO:
			case BINARY_ADD:
			case BINARY_SUBTRACT:
			case BINARY_SUBSCR:
			case BINARY_LSHIFT:
			case BINARY_RSHIFT:
			case BINARY_AND:
			case BINARY_XOR:
			case BINARY_OR:
				if (lastlc >= 2	 &&
				    ISBASICBLOCK(blocks, i-6, 7)  &&
				    fold_binops_on_constants(&codestr[i-6], consts)) {
					i -= 2;
					assert(codestr[i] == LOAD_CONST);
					cumlc = 1;
				}
				break;

				/* Fold unary ops on constants.
				   LOAD_CONST c1  UNARY_OP -->	LOAD_CONST unary_op(c) */
			case UNARY_NEGATIVE:
			case UNARY_CONVERT:
			case UNARY_INVERT:
				if (lastlc >= 1	 &&
				    ISBASICBLOCK(blocks, i-3, 4)  &&
				    fold_unaryops_on_constants(&codestr[i-3], consts))	{
					i -= 2;
					assert(codestr[i] == LOAD_CONST);
					cumlc = 1;
				}
				break;

				/* Simplify conditional jump to conditional jump where the
				   result of the first test implies the success of a similar
				   test or the failure of the opposite test.
				   Arises in code like:
				   "if a and b:"
				   "if a or b:"
				   "a and b or c"
				   "(a and b) and c"
				   x:JUMP_IF_FALSE y   y:JUMP_IF_FALSE z  -->  x:JUMP_IF_FALSE z
				   x:JUMP_IF_FALSE y   y:JUMP_IF_TRUE z	 -->  x:JUMP_IF_FALSE y+3
				   where y+3 is the instruction following the second test.
				*/
			case JUMP_IF_FALSE:
			case JUMP_IF_TRUE:
				tgt = GETJUMPTGT(codestr, i);
				j = codestr[tgt];
				if (j == JUMP_IF_FALSE	||  j == JUMP_IF_TRUE) {
					if (j == opcode) {
						tgttgt = GETJUMPTGT(codestr, tgt) - i - 3;
						SETARG(codestr, i, tgttgt);
					} else {
						tgt -= i;
						SETARG(codestr, i, tgt);
					}
					break;
				}
				/* Intentional fallthrough */  

				/* Replace jumps to unconditional jumps */
			case FOR_ITER:
			case JUMP_FORWARD:
			case JUMP_ABSOLUTE:
			case CONTINUE_LOOP:
			case SETUP_LOOP:
			case SETUP_EXCEPT:
			case SETUP_FINALLY:
				tgt = GETJUMPTGT(codestr, i);
				if (!UNCONDITIONAL_JUMP(codestr[tgt]))
					continue;
				tgttgt = GETJUMPTGT(codestr, tgt);
				if (opcode == JUMP_FORWARD) /* JMP_ABS can go backwards */
					opcode = JUMP_ABSOLUTE;
				if (!ABSOLUTE_JUMP(opcode))
					tgttgt -= i + 3;     /* Calc relative jump addr */
				if (tgttgt < 0)		  /* No backward relative jumps */
					continue;
				codestr[i] = opcode;
				SETARG(codestr, i, tgttgt);
				break;

			case EXTENDED_ARG:
				goto exitUnchanged;

				/* Replace RETURN LOAD_CONST None RETURN with just RETURN */
			case RETURN_VALUE:
				if (i+4 >= codelen  ||
				    codestr[i+4] != RETURN_VALUE	 ||
				    !ISBASICBLOCK(blocks,i,5))
					continue;
				memset(codestr+i+1, NOP, 4);
				break;
		}
	}

	/* Fixup linenotab */
	for (i=0, nops=0 ; i<codelen ; i += CODESIZE(codestr[i])) {
		addrmap[i] = i - nops;
		if (codestr[i] == NOP)
			nops++;
	}
	cum_orig_line = 0;
	last_line = 0;
	for (i=0 ; i < tabsiz ; i+=2) {
		cum_orig_line += lineno[i];
		new_line = addrmap[cum_orig_line];
		assert (new_line - last_line < 255);
		lineno[i] =((unsigned char)(new_line - last_line));
		last_line = new_line;
	}

	/* Remove NOPs and fixup jump targets */
	for (i=0, h=0 ; i<codelen ; ) {
		opcode = codestr[i];
		switch (opcode) {
			case NOP:
				i++;
				continue;

			case JUMP_ABSOLUTE:
			case CONTINUE_LOOP:
				j = addrmap[GETARG(codestr, i)];
				SETARG(codestr, i, j);
				break;

			case FOR_ITER:
			case JUMP_FORWARD:
			case JUMP_IF_FALSE:
			case JUMP_IF_TRUE:
			case SETUP_LOOP:
			case SETUP_EXCEPT:
			case SETUP_FINALLY:
				j = addrmap[GETARG(codestr, i) + i + 3] - addrmap[i] - 3;
				SETARG(codestr, i, j);
				break;
		}
		adj = CODESIZE(opcode);
		while (adj--)
			codestr[h++] = codestr[i++];
	}
	assert(h + nops == codelen);

	code = PyString_FromStringAndSize((char *)codestr, h);
	PyMem_Free(addrmap);
	PyMem_Free(codestr);
	PyMem_Free(blocks);
	return code;

 exitUnchanged:
	if (blocks != NULL)
		PyMem_Free(blocks);
	if (addrmap != NULL)
		PyMem_Free(addrmap);
	if (codestr != NULL)
		PyMem_Free(codestr);
	Py_INCREF(code);
	return code;
}

/* End: Peephole optimizations ----------------------------------------- */

/*

Leave this debugging code for just a little longer.

static void
compiler_display_symbols(PyObject *name, PyObject *symbols)
{
PyObject *key, *value;
int flags;
Py_ssize_t pos = 0;

fprintf(stderr, "block %s\n", PyString_AS_STRING(name));
while (PyDict_Next(symbols, &pos, &key, &value)) {
flags = PyInt_AsLong(value);
fprintf(stderr, "var %s:", PyString_AS_STRING(key));
if (flags & DEF_GLOBAL)
fprintf(stderr, " declared_global");
if (flags & DEF_LOCAL)
fprintf(stderr, " local");
if (flags & DEF_PARAM)
fprintf(stderr, " param");
if (flags & DEF_STAR)
fprintf(stderr, " stararg");
if (flags & DEF_DOUBLESTAR)
fprintf(stderr, " starstar");
if (flags & DEF_INTUPLE)
fprintf(stderr, " tuple");
if (flags & DEF_FREE)
fprintf(stderr, " free");
if (flags & DEF_FREE_GLOBAL)
fprintf(stderr, " global");
if (flags & DEF_FREE_CLASS)
fprintf(stderr, " free/class");
if (flags & DEF_IMPORT)
fprintf(stderr, " import");
fprintf(stderr, "\n");
}
	fprintf(stderr, "\n");
}
*/

static void
compiler_unit_check(struct compiler_unit *u)
{
	basicblock *block;
	for (block = u->u_blocks; block != NULL; block = block->b_list) {
		assert(block != (void *)0xcbcbcbcb);
		assert(block != (void *)0xfbfbfbfb);
		assert(block != (void *)0xdbdbdbdb);
		if (block->b_instr != NULL) {
			assert(block->b_ialloc > 0);
			assert(block->b_iused > 0);
			assert(block->b_ialloc >= block->b_iused);
		}
		else {
			assert (block->b_iused == 0);
			assert (block->b_ialloc == 0);
		}
	}
}

static void
compiler_unit_free(struct compiler_unit *u)
{
	basicblock *b, *next;

	compiler_unit_check(u);
	b = u->u_blocks;
	while (b != NULL) {
		if (b->b_instr)
			PyObject_Free((void *)b->b_instr);
		next = b->b_list;
		PyObject_Free((void *)b);
		b = next;
	}
	Py_CLEAR(u->u_ste);
	Py_CLEAR(u->u_name);
	Py_CLEAR(u->u_consts);
	Py_CLEAR(u->u_names);
	Py_CLEAR(u->u_varnames);
	Py_CLEAR(u->u_freevars);
	Py_CLEAR(u->u_cellvars);
	Py_CLEAR(u->u_private);
	PyObject_Free(u);
}

static int
compiler_enter_scope(struct compiler *c, identifier name, void *key,
		     int lineno)
{
	struct compiler_unit *u;

	u = (struct compiler_unit *)PyObject_Malloc(sizeof(
                                                struct compiler_unit));
	if (!u) {
		PyErr_NoMemory();
		return 0;
	}
	memset(u, 0, sizeof(struct compiler_unit));
	u->u_argcount = 0;
	u->u_ste = PySymtable_Lookup(c->c_st, key);
	if (!u->u_ste) {
		compiler_unit_free(u);
		return 0;
	}
	Py_INCREF(name);
	u->u_name = name;
	u->u_varnames = list2dict(u->u_ste->ste_varnames);
	u->u_cellvars = dictbytype(u->u_ste->ste_symbols, CELL, 0, 0);
	if (!u->u_varnames || !u->u_cellvars) {
		compiler_unit_free(u);
		return 0;
	}

	u->u_freevars = dictbytype(u->u_ste->ste_symbols, FREE, DEF_FREE_CLASS,
				   PyDict_Size(u->u_cellvars));
	if (!u->u_freevars) {
		compiler_unit_free(u);
		return 0;
	}

	u->u_blocks = NULL;
	u->u_tmpname = 0;
	u->u_nfblocks = 0;
	u->u_firstlineno = lineno;
	u->u_lineno = 0;
	u->u_lineno_set = false;
	u->u_consts = PyDict_New();
	if (!u->u_consts) {
		compiler_unit_free(u);
		return 0;
	}
	u->u_names = PyDict_New();
	if (!u->u_names) {
		compiler_unit_free(u);
		return 0;
	}

	u->u_private = NULL;

	/* Push the old compiler_unit on the stack. */
	if (c->u) {
		PyObject *wrapper = PyCObject_FromVoidPtr(c->u, NULL);
		if (!wrapper || PyList_Append(c->c_stack, wrapper) < 0) {
			Py_XDECREF(wrapper);
			compiler_unit_free(u);
			return 0;
		}
		Py_DECREF(wrapper);
		u->u_private = c->u->u_private;
		Py_XINCREF(u->u_private);
	}
	c->u = u;

	c->c_nestlevel++;
	if (compiler_use_new_block(c) == NULL)
		return 0;

	return 1;
}

static void
compiler_exit_scope(struct compiler *c)
{
	int n;
	PyObject *wrapper;

	c->c_nestlevel--;
	compiler_unit_free(c->u);
	/* Restore c->u to the parent unit. */
	n = PyList_GET_SIZE(c->c_stack) - 1;
	if (n >= 0) {
		wrapper = PyList_GET_ITEM(c->c_stack, n);
		c->u = (struct compiler_unit *)PyCObject_AsVoidPtr(wrapper);
		/* we are deleting from a list so this really shouldn't fail */
		if (PySequence_DelItem(c->c_stack, n) < 0)
			Py_FatalError("compiler_exit_scope()");
		compiler_unit_check(c->u);
	}
	else
		c->u = NULL;

}

/* Allocate a new "anonymous" local variable.
   Used by list comprehensions and with statements.
*/

static PyObject *
compiler_new_tmpname(struct compiler *c)
{
	char tmpname[256];
	PyOS_snprintf(tmpname, sizeof(tmpname), "_[%d]", ++c->u->u_tmpname);
	return PyString_FromString(tmpname);
}

/* Allocate a new block and return a pointer to it.
   Returns NULL on error.
*/

static basicblock *
compiler_new_block(struct compiler *c)
{
	basicblock *b;
	struct compiler_unit *u;

	u = c->u;
	b = (basicblock *)PyObject_Malloc(sizeof(basicblock));
	if (b == NULL) {
		PyErr_NoMemory();
		return NULL;
	}
	memset((void *)b, 0, sizeof(basicblock));
    /* Extend the singly linked list of blocks with new block. */
	b->b_list = u->u_blocks;
	u->u_blocks = b;
	return b;
}

static basicblock *
compiler_use_new_block(struct compiler *c)
{
	basicblock *block = compiler_new_block(c);
	if (block == NULL)
		return NULL;
	c->u->u_curblock = block;
	return block;
}

static basicblock *
compiler_next_block(struct compiler *c)
{
	basicblock *block = compiler_new_block(c);
	if (block == NULL)
		return NULL;
	c->u->u_curblock->b_next = block;
	c->u->u_curblock = block;
	return block;
}

static basicblock *
compiler_use_next_block(struct compiler *c, basicblock *block)
{
	assert(block != NULL);
	c->u->u_curblock->b_next = block;
	c->u->u_curblock = block;
	return block;
}

/* Returns the offset of the next instruction in the current block's
   b_instr array.  Resizes the b_instr as necessary.
   Returns -1 on failure.
 */

static int
compiler_next_instr(struct compiler *c, basicblock *b)
{
	assert(b != NULL);
	if (b->b_instr == NULL) {
		b->b_instr = (struct instr *)PyObject_Malloc(
                                 sizeof(struct instr) * DEFAULT_BLOCK_SIZE);
		if (b->b_instr == NULL) {
			PyErr_NoMemory();
			return -1;
		}
		b->b_ialloc = DEFAULT_BLOCK_SIZE;
		memset((char *)b->b_instr, 0,
		       sizeof(struct instr) * DEFAULT_BLOCK_SIZE);
	}
	else if (b->b_iused == b->b_ialloc) {
		struct instr *tmp;
		size_t oldsize, newsize;
		oldsize = b->b_ialloc * sizeof(struct instr);
		newsize = oldsize << 1;
		if (newsize == 0) {
			PyErr_NoMemory();
			return -1;
		}
		b->b_ialloc <<= 1;
		tmp = (struct instr *)PyObject_Realloc(
                                                (void *)b->b_instr, newsize);
		if (tmp == NULL) {
			PyErr_NoMemory();
			return -1;
		}
		b->b_instr = tmp;
		memset((char *)b->b_instr + oldsize, 0, newsize - oldsize);
	}
	return b->b_iused++;
}

/* Set the i_lineno member of the instruction at offse off if the
   line number for the current expression/statement (?) has not
   already been set.  If it has been set, the call has no effect.

   Every time a new node is b
   */

static void
compiler_set_lineno(struct compiler *c, int off)
{
	basicblock *b;
	if (c->u->u_lineno_set)
		return;
	c->u->u_lineno_set = true;
	b = c->u->u_curblock;
	b->b_instr[off].i_lineno = c->u->u_lineno;
}

static int
opcode_stack_effect(int opcode, int oparg)
{
	switch (opcode) {
		case POP_TOP:
			return -1;
		case ROT_TWO:
		case ROT_THREE:
			return 0;
		case DUP_TOP:
			return 1;
		case ROT_FOUR:
			return 0;

		case UNARY_POSITIVE:
		case UNARY_NEGATIVE:
		case UNARY_NOT:
		case UNARY_CONVERT:
		case UNARY_INVERT:
			return 0;

		case LIST_APPEND:
			return -2;

		case BINARY_POWER:
		case BINARY_MULTIPLY:
		case BINARY_DIVIDE:
		case BINARY_MODULO:
		case BINARY_ADD:
		case BINARY_SUBTRACT:
		case BINARY_SUBSCR:
		case BINARY_FLOOR_DIVIDE:
		case BINARY_TRUE_DIVIDE:
			return -1;
		case INPLACE_FLOOR_DIVIDE:
		case INPLACE_TRUE_DIVIDE:
			return -1;

		case SLICE+0:
			return 1;
		case SLICE+1:
			return 0;
		case SLICE+2:
			return 0;
		case SLICE+3:
			return -1;

		case STORE_SLICE+0:
			return -2;
		case STORE_SLICE+1:
			return -3;
		case STORE_SLICE+2:
			return -3;
		case STORE_SLICE+3:
			return -4;

		case DELETE_SLICE+0:
			return -1;
		case DELETE_SLICE+1:
			return -2;
		case DELETE_SLICE+2:
			return -2;
		case DELETE_SLICE+3:
			return -3;

		case INPLACE_ADD:
		case INPLACE_SUBTRACT:
		case INPLACE_MULTIPLY:
		case INPLACE_DIVIDE:
		case INPLACE_MODULO:
			return -1;
		case STORE_SUBSCR:
			return -3;
		case DELETE_SUBSCR:
			return -2;

		case BINARY_LSHIFT:
		case BINARY_RSHIFT:
		case BINARY_AND:
		case BINARY_XOR:
		case BINARY_OR:
			return -1;
		case INPLACE_POWER:
			return -1;
		case GET_ITER:
			return 0;

		case PRINT_EXPR:
			return -1;
		case PRINT_ITEM:
			return -1;
		case PRINT_NEWLINE:
			return 0;
		case PRINT_ITEM_TO:
			return -2;
		case PRINT_NEWLINE_TO:
			return -1;
		case INPLACE_LSHIFT:
		case INPLACE_RSHIFT:
		case INPLACE_AND:
		case INPLACE_XOR:
		case INPLACE_OR:
			return -1;
		case BREAK_LOOP:
			return 0;
		case WITH_CLEANUP:
			return -1; /* XXX Sometimes more */
		case LOAD_LOCALS:
			return 1;
		case RETURN_VALUE:
			return -1;
		case IMPORT_STAR:
			return -1;
		case EXEC_STMT:
			return -3;
		case YIELD_VALUE:
			return 0;

		case POP_BLOCK:
			return 0;
		case END_FINALLY:
			return -1; /* or -2 or -3 if exception occurred */
		case BUILD_CLASS:
			return -2;

		case STORE_NAME:
			return -1;
		case DELETE_NAME:
			return 0;
		case UNPACK_SEQUENCE:
			return oparg-1;
		case FOR_ITER:
			return 1;

		case STORE_ATTR:
			return -2;
		case DELETE_ATTR:
			return -1;
		case STORE_GLOBAL:
			return -1;
		case DELETE_GLOBAL:
			return 0;
		case DUP_TOPX:
			return oparg;
		case LOAD_CONST:
			return 1;
		case LOAD_NAME:
			return 1;
		case BUILD_TUPLE:
		case BUILD_LIST:
			return 1-oparg;
		case BUILD_MAP:
			return 1;
		case LOAD_ATTR:
			return 0;
		case COMPARE_OP:
			return -1;
		case IMPORT_NAME:
			return 0;
		case IMPORT_FROM:
			return 1;

		case JUMP_FORWARD:
		case JUMP_IF_FALSE:
		case JUMP_IF_TRUE:
		case JUMP_ABSOLUTE:
			return 0;

		case LOAD_GLOBAL:
			return 1;

		case CONTINUE_LOOP:
			return 0;
		case SETUP_LOOP:
			return 0;
		case SETUP_EXCEPT:
		case SETUP_FINALLY:
			return 3; /* actually pushed by an exception */

		case LOAD_FAST:
			return 1;
		case STORE_FAST:
			return -1;
		case DELETE_FAST:
			return 0;

		case RAISE_VARARGS:
			return -oparg;
#define NARGS(o) (((o) % 256) + 2*((o) / 256))
		case CALL_FUNCTION:
			return -NARGS(oparg);
		case CALL_FUNCTION_VAR:
		case CALL_FUNCTION_KW:
			return -NARGS(oparg)-1;
		case CALL_FUNCTION_VAR_KW:
			return -NARGS(oparg)-2;
#undef NARGS
		case MAKE_FUNCTION:
			return -oparg;
		case BUILD_SLICE:
			if (oparg == 3)
				return -2;
			else
				return -1;

		case MAKE_CLOSURE:
			return -oparg;
		case LOAD_CLOSURE:
			return 1;
		case LOAD_DEREF:
			return 1;
		case STORE_DEREF:
			return -1;
		default:
			fprintf(stderr, "opcode = %d\n", opcode);
			Py_FatalError("opcode_stack_effect()");

	}
	return 0; /* not reachable */
}

/* Add an opcode with no argument.
   Returns 0 on failure, 1 on success.
*/

static int
compiler_addop(struct compiler *c, int opcode)
{
	basicblock *b;
	struct instr *i;
	int off;
	off = compiler_next_instr(c, c->u->u_curblock);
	if (off < 0)
		return 0;
	b = c->u->u_curblock;
	i = &b->b_instr[off];
	i->i_opcode = opcode;
	i->i_hasarg = 0;
	if (opcode == RETURN_VALUE)
		b->b_return = 1;
	compiler_set_lineno(c, off);
	return 1;
}

static int
compiler_add_o(struct compiler *c, PyObject *dict, PyObject *o)
{
	PyObject *t, *v;
	Py_ssize_t arg;

	/* necessary to make sure types aren't coerced (e.g., int and long) */
	t = PyTuple_Pack(2, o, o->ob_type);
	if (t == NULL)
	    return -1;

	v = PyDict_GetItem(dict, t);
	if (!v) {
		arg = PyDict_Size(dict);
		v = PyInt_FromLong(arg);
		if (!v) {
			Py_DECREF(t);
			return -1;
		}
		if (PyDict_SetItem(dict, t, v) < 0) {
			Py_DECREF(t);
			Py_DECREF(v);
			return -1;
		}
		Py_DECREF(v);
	}
	else
		arg = PyInt_AsLong(v);
	Py_DECREF(t);
	return arg;
}

static int
compiler_addop_o(struct compiler *c, int opcode, PyObject *dict,
		     PyObject *o)
{
    int arg = compiler_add_o(c, dict, o);
    if (arg < 0)
	return 0;
    return compiler_addop_i(c, opcode, arg);
}

static int
compiler_addop_name(struct compiler *c, int opcode, PyObject *dict,
		    PyObject *o)
{
    int arg;
    PyObject *mangled = _Py_Mangle(c->u->u_private, o);
    if (!mangled)
	return 0;
    arg = compiler_add_o(c, dict, mangled);
    Py_DECREF(mangled);
    if (arg < 0)
	return 0;
    return compiler_addop_i(c, opcode, arg);
}

/* Add an opcode with an integer argument.
   Returns 0 on failure, 1 on success.
*/

static int
compiler_addop_i(struct compiler *c, int opcode, int oparg)
{
	struct instr *i;
	int off;
	off = compiler_next_instr(c, c->u->u_curblock);
	if (off < 0)
		return 0;
	i = &c->u->u_curblock->b_instr[off];
	i->i_opcode = opcode;
	i->i_oparg = oparg;
	i->i_hasarg = 1;
	compiler_set_lineno(c, off);
	return 1;
}

static int
compiler_addop_j(struct compiler *c, int opcode, basicblock *b, int absolute)
{
	struct instr *i;
	int off;

	assert(b != NULL);
	off = compiler_next_instr(c, c->u->u_curblock);
	if (off < 0)
		return 0;
	i = &c->u->u_curblock->b_instr[off];
	i->i_opcode = opcode;
	i->i_target = b;
	i->i_hasarg = 1;
	if (absolute)
		i->i_jabs = 1;
	else
		i->i_jrel = 1;
	compiler_set_lineno(c, off);
	return 1;
}

/* The distinction between NEW_BLOCK and NEXT_BLOCK is subtle.	(I'd
   like to find better names.)	NEW_BLOCK() creates a new block and sets
   it as the current block.  NEXT_BLOCK() also creates an implicit jump
   from the current block to the new block.
*/

/* XXX The returns inside these macros make it impossible to decref
   objects created in the local function.
*/


#define NEW_BLOCK(C) { \
	if (compiler_use_new_block((C)) == NULL) \
		return 0; \
}

#define NEXT_BLOCK(C) { \
	if (compiler_next_block((C)) == NULL) \
		return 0; \
}

#define ADDOP(C, OP) { \
	if (!compiler_addop((C), (OP))) \
		return 0; \
}

#define ADDOP_IN_SCOPE(C, OP) { \
	if (!compiler_addop((C), (OP))) { \
		compiler_exit_scope(c); \
		return 0; \
	} \
}

#define ADDOP_O(C, OP, O, TYPE) { \
	if (!compiler_addop_o((C), (OP), (C)->u->u_ ## TYPE, (O))) \
		return 0; \
}

#define ADDOP_NAME(C, OP, O, TYPE) { \
	if (!compiler_addop_name((C), (OP), (C)->u->u_ ## TYPE, (O))) \
		return 0; \
}

#define ADDOP_I(C, OP, O) { \
	if (!compiler_addop_i((C), (OP), (O))) \
		return 0; \
}

#define ADDOP_JABS(C, OP, O) { \
	if (!compiler_addop_j((C), (OP), (O), 1)) \
		return 0; \
}

#define ADDOP_JREL(C, OP, O) { \
	if (!compiler_addop_j((C), (OP), (O), 0)) \
		return 0; \
}

/* VISIT and VISIT_SEQ takes an ASDL type as their second argument.  They use
   the ASDL name to synthesize the name of the C type and the visit function.
*/

#define VISIT(C, TYPE, V) {\
	if (!compiler_visit_ ## TYPE((C), (V))) \
		return 0; \
}

#define VISIT_IN_SCOPE(C, TYPE, V) {\
	if (!compiler_visit_ ## TYPE((C), (V))) { \
		compiler_exit_scope(c); \
		return 0; \
	} \
}

#define VISIT_SLICE(C, V, CTX) {\
	if (!compiler_visit_slice((C), (V), (CTX))) \
		return 0; \
}

#define VISIT_SEQ(C, TYPE, SEQ) { \
	int _i; \
	asdl_seq *seq = (SEQ); /* avoid variable capture */ \
	for (_i = 0; _i < asdl_seq_LEN(seq); _i++) { \
		TYPE ## _ty elt = (TYPE ## _ty)asdl_seq_GET(seq, _i); \
		if (!compiler_visit_ ## TYPE((C), elt)) \
			return 0; \
	} \
}

#define VISIT_SEQ_IN_SCOPE(C, TYPE, SEQ) { \
	int _i; \
	asdl_seq *seq = (SEQ); /* avoid variable capture */ \
	for (_i = 0; _i < asdl_seq_LEN(seq); _i++) { \
		TYPE ## _ty elt = (TYPE ## _ty)asdl_seq_GET(seq, _i); \
		if (!compiler_visit_ ## TYPE((C), elt)) { \
			compiler_exit_scope(c); \
			return 0; \
		} \
	} \
}

static int
compiler_isdocstring(stmt_ty s)
{
    if (s->kind != Expr_kind)
	return 0;
    return s->v.Expr.value->kind == Str_kind;
}

/* Compile a sequence of statements, checking for a docstring. */

static int
compiler_body(struct compiler *c, asdl_seq *stmts)
{
	int i = 0;
	stmt_ty st;

	if (!asdl_seq_LEN(stmts))
		return 1;
	st = (stmt_ty)asdl_seq_GET(stmts, 0);
	if (compiler_isdocstring(st)) {
		i = 1;
		VISIT(c, expr, st->v.Expr.value);
		if (!compiler_nameop(c, __doc__, Store))
			return 0;
	}
	for (; i < asdl_seq_LEN(stmts); i++)
	    VISIT(c, stmt, (stmt_ty)asdl_seq_GET(stmts, i));
	return 1;
}

static PyCodeObject *
compiler_mod(struct compiler *c, mod_ty mod)
{
	PyCodeObject *co;
	int addNone = 1;
	static PyObject *module;
	if (!module) {
		module = PyString_FromString("<module>");
		if (!module)
			return NULL;
	}
	/* Use 0 for firstlineno initially, will fixup in assemble(). */
	if (!compiler_enter_scope(c, module, mod, 0))
		return NULL;
	switch (mod->kind) {
	case Module_kind: 
		if (!compiler_body(c, mod->v.Module.body)) {
			compiler_exit_scope(c);
			return 0;
		}
		break;
	case Interactive_kind:
		c->c_interactive = 1;
		VISIT_SEQ_IN_SCOPE(c, stmt, 
                                        mod->v.Interactive.body);
		break;
	case Expression_kind:
		VISIT_IN_SCOPE(c, expr, mod->v.Expression.body);
		addNone = 0;
		break;
	case Suite_kind:
		PyErr_SetString(PyExc_SystemError,
				"suite should not be possible");
		return 0;
	default:
		PyErr_Format(PyExc_SystemError,
			     "module kind %d should not be possible",
			     mod->kind);
		return 0;
	}
	co = assemble(c, addNone);
	compiler_exit_scope(c);
	return co;
}

/* The test for LOCAL must come before the test for FREE in order to
   handle classes where name is both local and free.  The local var is
   a method and the free var is a free var referenced within a method.
*/

static int
get_ref_type(struct compiler *c, PyObject *name)
{
	int scope = PyST_GetScope(c->u->u_ste, name);
	if (scope == 0) {
	    char buf[350];
	    PyOS_snprintf(buf, sizeof(buf),
			  "unknown scope for %.100s in %.100s(%s) in %s\n"
			  "symbols: %s\nlocals: %s\nglobals: %s\n",
			  PyString_AS_STRING(name), 
			  PyString_AS_STRING(c->u->u_name), 
			  PyObject_REPR(c->u->u_ste->ste_id),
			  c->c_filename,
			  PyObject_REPR(c->u->u_ste->ste_symbols),
			  PyObject_REPR(c->u->u_varnames),
			  PyObject_REPR(c->u->u_names)
		);
	    Py_FatalError(buf);
	}

	return scope;
}

static int
compiler_lookup_arg(PyObject *dict, PyObject *name)
{
    PyObject *k, *v;
    k = PyTuple_Pack(2, name, name->ob_type);
    if (k == NULL)
	return -1;
    v = PyDict_GetItem(dict, k);
    Py_DECREF(k);
    if (v == NULL)
	return -1;
    return PyInt_AS_LONG(v);
}

static int
compiler_make_closure(struct compiler *c, PyCodeObject *co, int args)
{
	int i, free = PyCode_GetNumFree(co);
	if (free == 0) {
	    ADDOP_O(c, LOAD_CONST, (PyObject*)co, consts);
	    ADDOP_I(c, MAKE_FUNCTION, args);
	    return 1;
	}
	for (i = 0; i < free; ++i) {
		/* Bypass com_addop_varname because it will generate
		   LOAD_DEREF but LOAD_CLOSURE is needed. 
		*/
		PyObject *name = PyTuple_GET_ITEM(co->co_freevars, i);
		int arg, reftype;

		/* Special case: If a class contains a method with a
		   free variable that has the same name as a method,
		   the name will be considered free *and* local in the
		   class.  It should be handled by the closure, as
		   well as by the normal name loookup logic.
		*/
		reftype = get_ref_type(c, name);
		if (reftype == CELL)
			arg = compiler_lookup_arg(c->u->u_cellvars, name);
		else /* (reftype == FREE) */
			arg = compiler_lookup_arg(c->u->u_freevars, name);
		if (arg == -1) {
			printf("lookup %s in %s %d %d\n"
				"freevars of %s: %s\n",
				PyObject_REPR(name), 
				PyString_AS_STRING(c->u->u_name), 
				reftype, arg,
				PyString_AS_STRING(co->co_name),
				PyObject_REPR(co->co_freevars));
			Py_FatalError("compiler_make_closure()");
		}
		ADDOP_I(c, LOAD_CLOSURE, arg);
	}
	ADDOP_I(c, BUILD_TUPLE, free);
	ADDOP_O(c, LOAD_CONST, (PyObject*)co, consts);
	ADDOP_I(c, MAKE_CLOSURE, args);
	return 1;
}

static int
compiler_decorators(struct compiler *c, asdl_seq* decos)
{
	int i;

	if (!decos)
		return 1;

	for (i = 0; i < asdl_seq_LEN(decos); i++) {
		VISIT(c, expr, (expr_ty)asdl_seq_GET(decos, i));
	}
	return 1;
}

static int
compiler_arguments(struct compiler *c, arguments_ty args)
{
	int i;
	int n = asdl_seq_LEN(args->args);
	/* Correctly handle nested argument lists */
	for (i = 0; i < n; i++) {
		expr_ty arg = (expr_ty)asdl_seq_GET(args->args, i);
		if (arg->kind == Tuple_kind) {
			PyObject *id = PyString_FromFormat(".%d", i);
			if (id == NULL) {
				return 0;
			}
			if (!compiler_nameop(c, id, Load)) {
				Py_DECREF(id);
				return 0;
			}
			Py_DECREF(id);
			VISIT(c, expr, arg);
		}
	}
	return 1;
}

static int
compiler_function(struct compiler *c, stmt_ty s)
{
	PyCodeObject *co;
	PyObject *first_const = Py_None;
	arguments_ty args = s->v.FunctionDef.args;
	asdl_seq* decos = s->v.FunctionDef.decorators;
	stmt_ty st;
	int i, n, docstring;

	assert(s->kind == FunctionDef_kind);

	if (!compiler_decorators(c, decos))
		return 0;
	if (args->defaults)
		VISIT_SEQ(c, expr, args->defaults);
	if (!compiler_enter_scope(c, s->v.FunctionDef.name, (void *)s,
				  s->lineno))
		return 0;

	st = (stmt_ty)asdl_seq_GET(s->v.FunctionDef.body, 0);
	docstring = compiler_isdocstring(st);
	if (docstring)
	    first_const = st->v.Expr.value->v.Str.s;
	if (compiler_add_o(c, c->u->u_consts, first_const) < 0)	 {
	    compiler_exit_scope(c);
	    return 0;
	}

	/* unpack nested arguments */
	compiler_arguments(c, args);

	c->u->u_argcount = asdl_seq_LEN(args->args);
	n = asdl_seq_LEN(s->v.FunctionDef.body);
	/* if there was a docstring, we need to skip the first statement */
	for (i = docstring; i < n; i++) {
		st = (stmt_ty)asdl_seq_GET(s->v.FunctionDef.body, i);
		VISIT_IN_SCOPE(c, stmt, st);
	}
	co = assemble(c, 1);
	compiler_exit_scope(c);
	if (co == NULL)
		return 0;

	compiler_make_closure(c, co, asdl_seq_LEN(args->defaults));
	Py_DECREF(co);

	for (i = 0; i < asdl_seq_LEN(decos); i++) {
		ADDOP_I(c, CALL_FUNCTION, 1);
	}

	return compiler_nameop(c, s->v.FunctionDef.name, Store);
}

static int
compiler_class(struct compiler *c, stmt_ty s)
{
	int n;
	PyCodeObject *co;
	PyObject *str;
	/* push class name on stack, needed by BUILD_CLASS */
	ADDOP_O(c, LOAD_CONST, s->v.ClassDef.name, consts);
	/* push the tuple of base classes on the stack */
	n = asdl_seq_LEN(s->v.ClassDef.bases);
	if (n > 0)
		VISIT_SEQ(c, expr, s->v.ClassDef.bases);
	ADDOP_I(c, BUILD_TUPLE, n);
	if (!compiler_enter_scope(c, s->v.ClassDef.name, (void *)s,
				  s->lineno))
		return 0;
	c->u->u_private = s->v.ClassDef.name;
	Py_INCREF(c->u->u_private);
	str = PyString_InternFromString("__name__");
	if (!str || !compiler_nameop(c, str, Load)) {
		Py_XDECREF(str);
		compiler_exit_scope(c);
		return 0;
	}
	
	Py_DECREF(str);
	str = PyString_InternFromString("__module__");
	if (!str || !compiler_nameop(c, str, Store)) {
		Py_XDECREF(str);
		compiler_exit_scope(c);
		return 0;
	}
	Py_DECREF(str);

	if (!compiler_body(c, s->v.ClassDef.body)) {
		compiler_exit_scope(c);
		return 0;
	}

	ADDOP_IN_SCOPE(c, LOAD_LOCALS);
	ADDOP_IN_SCOPE(c, RETURN_VALUE);
	co = assemble(c, 1);
	compiler_exit_scope(c);
	if (co == NULL)
		return 0;

	compiler_make_closure(c, co, 0);
	Py_DECREF(co);

	ADDOP_I(c, CALL_FUNCTION, 0);
	ADDOP(c, BUILD_CLASS);
	if (!compiler_nameop(c, s->v.ClassDef.name, Store))
		return 0;
	return 1;
}

static int
compiler_ifexp(struct compiler *c, expr_ty e)
{
	basicblock *end, *next;
	
	assert(e->kind == IfExp_kind);
	end = compiler_new_block(c);
	if (end == NULL)
		return 0;
	next = compiler_new_block(c);
	if (next == NULL)
		return 0;
	VISIT(c, expr, e->v.IfExp.test);
	ADDOP_JREL(c, JUMP_IF_FALSE, next);
	ADDOP(c, POP_TOP);
	VISIT(c, expr, e->v.IfExp.body);
	ADDOP_JREL(c, JUMP_FORWARD, end);
	compiler_use_next_block(c, next);
	ADDOP(c, POP_TOP);
	VISIT(c, expr, e->v.IfExp.orelse);
	compiler_use_next_block(c, end);
	return 1;
}

static int
compiler_lambda(struct compiler *c, expr_ty e)
{
	PyCodeObject *co;
	static identifier name;
	arguments_ty args = e->v.Lambda.args;
	assert(e->kind == Lambda_kind);

	if (!name) {
		name = PyString_InternFromString("<lambda>");
		if (!name)
			return 0;
	}

	if (args->defaults)
		VISIT_SEQ(c, expr, args->defaults);
	if (!compiler_enter_scope(c, name, (void *)e, e->lineno))
		return 0;

	/* unpack nested arguments */
	compiler_arguments(c, args);
	
	c->u->u_argcount = asdl_seq_LEN(args->args);
	VISIT_IN_SCOPE(c, expr, e->v.Lambda.body);
	ADDOP_IN_SCOPE(c, RETURN_VALUE);
	co = assemble(c, 1);
	compiler_exit_scope(c);
	if (co == NULL)
		return 0;

	compiler_make_closure(c, co, asdl_seq_LEN(args->defaults));
	Py_DECREF(co);

	return 1;
}

static int
compiler_print(struct compiler *c, stmt_ty s)
{
	int i, n;
	bool dest;

	assert(s->kind == Print_kind);
	n = asdl_seq_LEN(s->v.Print.values);
	dest = false;
	if (s->v.Print.dest) {
		VISIT(c, expr, s->v.Print.dest);
		dest = true;
	}
	for (i = 0; i < n; i++) {
		expr_ty e = (expr_ty)asdl_seq_GET(s->v.Print.values, i);
		if (dest) {
			ADDOP(c, DUP_TOP);
			VISIT(c, expr, e);
			ADDOP(c, ROT_TWO);
			ADDOP(c, PRINT_ITEM_TO);
		}
		else {
			VISIT(c, expr, e);
			ADDOP(c, PRINT_ITEM);
		}
	}
	if (s->v.Print.nl) {
		if (dest)
			ADDOP(c, PRINT_NEWLINE_TO)
		else
			ADDOP(c, PRINT_NEWLINE)
	}
	else if (dest)
		ADDOP(c, POP_TOP);
	return 1;
}

static int
compiler_if(struct compiler *c, stmt_ty s)
{
	basicblock *end, *next;
	int constant;
	assert(s->kind == If_kind);
	end = compiler_new_block(c);
	if (end == NULL)
		return 0;
	next = compiler_new_block(c);
	if (next == NULL)
	    return 0;
	
	constant = expr_constant(s->v.If.test);
	/* constant = 0: "if 0"
	 * constant = 1: "if 1", "if 2", ...
	 * constant = -1: rest */
	if (constant == 0) {
		if (s->v.If.orelse)
			VISIT_SEQ(c, stmt, s->v.If.orelse);
	} else if (constant == 1) {
		VISIT_SEQ(c, stmt, s->v.If.body);
	} else {
		VISIT(c, expr, s->v.If.test);
		ADDOP_JREL(c, JUMP_IF_FALSE, next);
		ADDOP(c, POP_TOP);
		VISIT_SEQ(c, stmt, s->v.If.body);
		ADDOP_JREL(c, JUMP_FORWARD, end);
		compiler_use_next_block(c, next);
		ADDOP(c, POP_TOP);
		if (s->v.If.orelse)
	    		VISIT_SEQ(c, stmt, s->v.If.orelse);
	}
	compiler_use_next_block(c, end);
	return 1;
}

static int
compiler_for(struct compiler *c, stmt_ty s)
{
	basicblock *start, *cleanup, *end;

	start = compiler_new_block(c);
	cleanup = compiler_new_block(c);
	end = compiler_new_block(c);
	if (start == NULL || end == NULL || cleanup == NULL)
		return 0;
	ADDOP_JREL(c, SETUP_LOOP, end);
	if (!compiler_push_fblock(c, LOOP, start))
		return 0;
	VISIT(c, expr, s->v.For.iter);
	ADDOP(c, GET_ITER);
	compiler_use_next_block(c, start);
	/* XXX(nnorwitz): is there a better way to handle this?
	   for loops are special, we want to be able to trace them
	   each time around, so we need to set an extra line number. */
	c->u->u_lineno_set = false;
	ADDOP_JREL(c, FOR_ITER, cleanup);
	VISIT(c, expr, s->v.For.target);
	VISIT_SEQ(c, stmt, s->v.For.body);
	ADDOP_JABS(c, JUMP_ABSOLUTE, start);
	compiler_use_next_block(c, cleanup);
	ADDOP(c, POP_BLOCK);
	compiler_pop_fblock(c, LOOP, start);
	VISIT_SEQ(c, stmt, s->v.For.orelse);
	compiler_use_next_block(c, end);
	return 1;
}

static int
compiler_while(struct compiler *c, stmt_ty s)
{
	basicblock *loop, *orelse, *end, *anchor = NULL;
	int constant = expr_constant(s->v.While.test);

	if (constant == 0)
		return 1;
	loop = compiler_new_block(c);
	end = compiler_new_block(c);
	if (constant == -1) {
		anchor = compiler_new_block(c);
		if (anchor == NULL)
			return 0;
	}
	if (loop == NULL || end == NULL)
		return 0;
	if (s->v.While.orelse) {
		orelse = compiler_new_block(c);
		if (orelse == NULL)
			return 0;
	}
	else
		orelse = NULL;

	ADDOP_JREL(c, SETUP_LOOP, end);
	compiler_use_next_block(c, loop);
	if (!compiler_push_fblock(c, LOOP, loop))
		return 0;
	if (constant == -1) {
		VISIT(c, expr, s->v.While.test);
		ADDOP_JREL(c, JUMP_IF_FALSE, anchor);
		ADDOP(c, POP_TOP);
	}
	VISIT_SEQ(c, stmt, s->v.While.body);
	ADDOP_JABS(c, JUMP_ABSOLUTE, loop);

	/* XXX should the two POP instructions be in a separate block
	   if there is no else clause ?
	*/

	if (constant == -1) {
		compiler_use_next_block(c, anchor);
		ADDOP(c, POP_TOP);
		ADDOP(c, POP_BLOCK);
	}
	compiler_pop_fblock(c, LOOP, loop);
	if (orelse != NULL) /* what if orelse is just pass? */
		VISIT_SEQ(c, stmt, s->v.While.orelse);
	compiler_use_next_block(c, end);

	return 1;
}

static int
compiler_continue(struct compiler *c)
{
	static const char LOOP_ERROR_MSG[] = "'continue' not properly in loop";
	int i;

	if (!c->u->u_nfblocks)
		return compiler_error(c, LOOP_ERROR_MSG);
	i = c->u->u_nfblocks - 1;
	switch (c->u->u_fblock[i].fb_type) {
	case LOOP:
		ADDOP_JABS(c, JUMP_ABSOLUTE, c->u->u_fblock[i].fb_block);
		break;
	case EXCEPT:
	case FINALLY_TRY:
		while (--i >= 0 && c->u->u_fblock[i].fb_type != LOOP)
			;
		if (i == -1)
			return compiler_error(c, LOOP_ERROR_MSG);
		ADDOP_JABS(c, CONTINUE_LOOP, c->u->u_fblock[i].fb_block);
		break;
	case FINALLY_END:
		return compiler_error(c,
			"'continue' not supported inside 'finally' clause");
	}

	return 1;
}

/* Code generated for "try: <body> finally: <finalbody>" is as follows:
   
		SETUP_FINALLY	L
		<code for body>
		POP_BLOCK
		LOAD_CONST	<None>
	L:	<code for finalbody>
		END_FINALLY
   
   The special instructions use the block stack.  Each block
   stack entry contains the instruction that created it (here
   SETUP_FINALLY), the level of the value stack at the time the
   block stack entry was created, and a label (here L).
   
   SETUP_FINALLY:
	Pushes the current value stack level and the label
	onto the block stack.
   POP_BLOCK:
	Pops en entry from the block stack, and pops the value
	stack until its level is the same as indicated on the
	block stack.  (The label is ignored.)
   END_FINALLY:
	Pops a variable number of entries from the *value* stack
	and re-raises the exception they specify.  The number of
	entries popped depends on the (pseudo) exception type.
   
   The block stack is unwound when an exception is raised:
   when a SETUP_FINALLY entry is found, the exception is pushed
   onto the value stack (and the exception condition is cleared),
   and the interpreter jumps to the label gotten from the block
   stack.
*/

static int
compiler_try_finally(struct compiler *c, stmt_ty s)
{
	basicblock *body, *end;
	body = compiler_new_block(c);
	end = compiler_new_block(c);
	if (body == NULL || end == NULL)
		return 0;

	ADDOP_JREL(c, SETUP_FINALLY, end);
	compiler_use_next_block(c, body);
	if (!compiler_push_fblock(c, FINALLY_TRY, body))
		return 0;
	VISIT_SEQ(c, stmt, s->v.TryFinally.body);
	ADDOP(c, POP_BLOCK);
	compiler_pop_fblock(c, FINALLY_TRY, body);

	ADDOP_O(c, LOAD_CONST, Py_None, consts);
	compiler_use_next_block(c, end);
	if (!compiler_push_fblock(c, FINALLY_END, end))
		return 0;
	VISIT_SEQ(c, stmt, s->v.TryFinally.finalbody);
	ADDOP(c, END_FINALLY);
	compiler_pop_fblock(c, FINALLY_END, end);

	return 1;
}

/*
   Code generated for "try: S except E1, V1: S1 except E2, V2: S2 ...":
   (The contents of the value stack is shown in [], with the top
   at the right; 'tb' is trace-back info, 'val' the exception's
   associated value, and 'exc' the exception.)
   
   Value stack		Label	Instruction	Argument
   []				SETUP_EXCEPT	L1
   []				<code for S>
   []				POP_BLOCK
   []				JUMP_FORWARD	L0
   
   [tb, val, exc]	L1:	DUP				)
   [tb, val, exc, exc]		<evaluate E1>			)
   [tb, val, exc, exc, E1]	COMPARE_OP	EXC_MATCH	) only if E1
   [tb, val, exc, 1-or-0]	JUMP_IF_FALSE	L2		)
   [tb, val, exc, 1]		POP				)
   [tb, val, exc]		POP
   [tb, val]			<assign to V1>	(or POP if no V1)
   [tb]				POP
   []				<code for S1>
				JUMP_FORWARD	L0
   
   [tb, val, exc, 0]	L2:	POP
   [tb, val, exc]		DUP
   .............................etc.......................

   [tb, val, exc, 0]	Ln+1:	POP
   [tb, val, exc]		END_FINALLY	# re-raise exception
   
   []			L0:	<next statement>
   
   Of course, parts are not generated if Vi or Ei is not present.
*/
static int
compiler_try_except(struct compiler *c, stmt_ty s)
{
	basicblock *body, *orelse, *except, *end;
	int i, n;

	body = compiler_new_block(c);
	except = compiler_new_block(c);
	orelse = compiler_new_block(c);
	end = compiler_new_block(c);
	if (body == NULL || except == NULL || orelse == NULL || end == NULL)
		return 0;
	ADDOP_JREL(c, SETUP_EXCEPT, except);
	compiler_use_next_block(c, body);
	if (!compiler_push_fblock(c, EXCEPT, body))
		return 0;
	VISIT_SEQ(c, stmt, s->v.TryExcept.body);
	ADDOP(c, POP_BLOCK);
	compiler_pop_fblock(c, EXCEPT, body);
	ADDOP_JREL(c, JUMP_FORWARD, orelse);
	n = asdl_seq_LEN(s->v.TryExcept.handlers);
	compiler_use_next_block(c, except);
	for (i = 0; i < n; i++) {
		excepthandler_ty handler = (excepthandler_ty)asdl_seq_GET(
						s->v.TryExcept.handlers, i);
		if (!handler->type && i < n-1)
		    return compiler_error(c, "default 'except:' must be last");
        c->u->u_lineno_set = false;
        c->u->u_lineno = handler->lineno;
		except = compiler_new_block(c);
		if (except == NULL)
			return 0;
		if (handler->type) {
			ADDOP(c, DUP_TOP);
			VISIT(c, expr, handler->type);
			ADDOP_I(c, COMPARE_OP, PyCmp_EXC_MATCH);
			ADDOP_JREL(c, JUMP_IF_FALSE, except);
			ADDOP(c, POP_TOP);
		}
		ADDOP(c, POP_TOP);
		if (handler->name) {
			VISIT(c, expr, handler->name);
		}
		else {
			ADDOP(c, POP_TOP);
		}
		ADDOP(c, POP_TOP);
		VISIT_SEQ(c, stmt, handler->body);
		ADDOP_JREL(c, JUMP_FORWARD, end);
		compiler_use_next_block(c, except);
		if (handler->type)
			ADDOP(c, POP_TOP);
	}
	ADDOP(c, END_FINALLY);
	compiler_use_next_block(c, orelse);
	VISIT_SEQ(c, stmt, s->v.TryExcept.orelse);
	compiler_use_next_block(c, end);
	return 1;
}

static int
compiler_import_as(struct compiler *c, identifier name, identifier asname)
{
	/* The IMPORT_NAME opcode was already generated.  This function
	   merely needs to bind the result to a name.

	   If there is a dot in name, we need to split it and emit a 
	   LOAD_ATTR for each name.
	*/
	const char *src = PyString_AS_STRING(name);
	const char *dot = strchr(src, '.');
	if (dot) {
		/* Consume the base module name to get the first attribute */
		src = dot + 1;
		while (dot) {
			/* NB src is only defined when dot != NULL */
			PyObject *attr;
			dot = strchr(src, '.');
			attr = PyString_FromStringAndSize(src, 
					    dot ? dot - src : strlen(src));
			if (!attr)
				return -1;
			ADDOP_O(c, LOAD_ATTR, attr, names);
			Py_DECREF(attr);
			src = dot + 1;
		}
	}
	return compiler_nameop(c, asname, Store);
}

static int
compiler_import(struct compiler *c, stmt_ty s)
{
	/* The Import node stores a module name like a.b.c as a single
	   string.  This is convenient for all cases except
	     import a.b.c as d
	   where we need to parse that string to extract the individual
	   module names.  
	   XXX Perhaps change the representation to make this case simpler?
	 */
	int i, n = asdl_seq_LEN(s->v.Import.names);

	for (i = 0; i < n; i++) {
		alias_ty alias = (alias_ty)asdl_seq_GET(s->v.Import.names, i);
		int r;
		PyObject *level;

		if (c->c_flags && (c->c_flags->cf_flags & CO_FUTURE_ABSOLUTE_IMPORT))
			level = PyInt_FromLong(0);
		else
			level = PyInt_FromLong(-1);

		if (level == NULL)
			return 0;

		ADDOP_O(c, LOAD_CONST, level, consts);
		Py_DECREF(level);
		ADDOP_O(c, LOAD_CONST, Py_None, consts);
		ADDOP_NAME(c, IMPORT_NAME, alias->name, names);

		if (alias->asname) {
			r = compiler_import_as(c, alias->name, alias->asname);
			if (!r)
			    return r;
		}
		else {
			identifier tmp = alias->name;
			const char *base = PyString_AS_STRING(alias->name);
			char *dot = strchr(base, '.');
			if (dot)
				tmp = PyString_FromStringAndSize(base, 
								 dot - base);
			r = compiler_nameop(c, tmp, Store);
			if (dot) {
				Py_DECREF(tmp);
			}
			if (!r)
				return r;
		}
	}
	return 1;
}

static int
compiler_from_import(struct compiler *c, stmt_ty s)
{
	int i, n = asdl_seq_LEN(s->v.ImportFrom.names);

	PyObject *names = PyTuple_New(n);
	PyObject *level;
	
	if (!names)
		return 0;

	if (s->v.ImportFrom.level == 0 && c->c_flags &&
	    !(c->c_flags->cf_flags & CO_FUTURE_ABSOLUTE_IMPORT))
		level = PyInt_FromLong(-1);
	else
		level = PyInt_FromLong(s->v.ImportFrom.level);

	if (!level) {
		Py_DECREF(names);
		return 0;
	}

	/* build up the names */
	for (i = 0; i < n; i++) {
		alias_ty alias = (alias_ty)asdl_seq_GET(s->v.ImportFrom.names, i);
		Py_INCREF(alias->name);
		PyTuple_SET_ITEM(names, i, alias->name);
	}

	if (s->lineno > c->c_future->ff_lineno) {
		if (!strcmp(PyString_AS_STRING(s->v.ImportFrom.module),
			    "__future__")) {
			Py_DECREF(level);
			Py_DECREF(names);
			return compiler_error(c, 
				      "from __future__ imports must occur "
				      "at the beginning of the file");

		}
	}

	ADDOP_O(c, LOAD_CONST, level, consts);
	Py_DECREF(level);
	ADDOP_O(c, LOAD_CONST, names, consts);
	Py_DECREF(names);
	ADDOP_NAME(c, IMPORT_NAME, s->v.ImportFrom.module, names);
	for (i = 0; i < n; i++) {
		alias_ty alias = (alias_ty)asdl_seq_GET(s->v.ImportFrom.names, i);
		identifier store_name;

		if (i == 0 && *PyString_AS_STRING(alias->name) == '*') {
			assert(n == 1);
			ADDOP(c, IMPORT_STAR);
			return 1;
		}
		    
		ADDOP_NAME(c, IMPORT_FROM, alias->name, names);
		store_name = alias->name;
		if (alias->asname)
			store_name = alias->asname;

		if (!compiler_nameop(c, store_name, Store)) {
			Py_DECREF(names);
			return 0;
		}
	}
	/* remove imported module */
	ADDOP(c, POP_TOP);
	return 1;
}

static int
compiler_assert(struct compiler *c, stmt_ty s)
{
	static PyObject *assertion_error = NULL;
	basicblock *end;

	if (Py_OptimizeFlag)
		return 1;
	if (assertion_error == NULL) {
		assertion_error = PyString_FromString("AssertionError");
		if (assertion_error == NULL)
			return 0;
	}
	VISIT(c, expr, s->v.Assert.test);
	end = compiler_new_block(c);
	if (end == NULL)
		return 0;
	ADDOP_JREL(c, JUMP_IF_TRUE, end);
	ADDOP(c, POP_TOP);
	ADDOP_O(c, LOAD_GLOBAL, assertion_error, names);
	if (s->v.Assert.msg) {
		VISIT(c, expr, s->v.Assert.msg);
		ADDOP_I(c, RAISE_VARARGS, 2);
	}
	else {
		ADDOP_I(c, RAISE_VARARGS, 1);
	}
	compiler_use_next_block(c, end);
	ADDOP(c, POP_TOP);
	return 1;
}

static int
compiler_visit_stmt(struct compiler *c, stmt_ty s)
{
	int i, n;

    /* Always assign a lineno to the next instruction for a stmt. */
	c->u->u_lineno = s->lineno;
	c->u->u_lineno_set = false;

	switch (s->kind) {
	case FunctionDef_kind:
		return compiler_function(c, s);
	case ClassDef_kind:
		return compiler_class(c, s);
	case Return_kind:
		if (c->u->u_ste->ste_type != FunctionBlock)
			return compiler_error(c, "'return' outside function");
		if (s->v.Return.value) {
			VISIT(c, expr, s->v.Return.value);
		}
		else
			ADDOP_O(c, LOAD_CONST, Py_None, consts);
		ADDOP(c, RETURN_VALUE);
		break;
	case Delete_kind:
		VISIT_SEQ(c, expr, s->v.Delete.targets)
		break;
	case Assign_kind:
		n = asdl_seq_LEN(s->v.Assign.targets);
		VISIT(c, expr, s->v.Assign.value);
		for (i = 0; i < n; i++) {
			if (i < n - 1)
				ADDOP(c, DUP_TOP);
			VISIT(c, expr,
			      (expr_ty)asdl_seq_GET(s->v.Assign.targets, i));
		}
		break;
	case AugAssign_kind:
		return compiler_augassign(c, s);
	case Print_kind:
		return compiler_print(c, s);
	case For_kind:
		return compiler_for(c, s);
	case While_kind:
		return compiler_while(c, s);
	case If_kind:
		return compiler_if(c, s);
	case Raise_kind:
		n = 0;
		if (s->v.Raise.type) {
			VISIT(c, expr, s->v.Raise.type);
			n++;
			if (s->v.Raise.inst) {
				VISIT(c, expr, s->v.Raise.inst);
				n++;
				if (s->v.Raise.tback) {
					VISIT(c, expr, s->v.Raise.tback);
					n++;
				}
			}
		}
		ADDOP_I(c, RAISE_VARARGS, n);
		break;
	case TryExcept_kind:
		return compiler_try_except(c, s);
	case TryFinally_kind:
		return compiler_try_finally(c, s);
	case Assert_kind:
		return compiler_assert(c, s);
	case Import_kind:
		return compiler_import(c, s);
	case ImportFrom_kind:
		return compiler_from_import(c, s);
	case Exec_kind:
		VISIT(c, expr, s->v.Exec.body);
		if (s->v.Exec.globals) {
			VISIT(c, expr, s->v.Exec.globals);
			if (s->v.Exec.locals) {
				VISIT(c, expr, s->v.Exec.locals);
			} else {
				ADDOP(c, DUP_TOP);
			}
		} else {
			ADDOP_O(c, LOAD_CONST, Py_None, consts);
			ADDOP(c, DUP_TOP);
		}
		ADDOP(c, EXEC_STMT);
		break;
	case Global_kind:
		break;
	case Expr_kind:
		if (c->c_interactive && c->c_nestlevel <= 1) {
			VISIT(c, expr, s->v.Expr.value);
			ADDOP(c, PRINT_EXPR);
		}
		else if (s->v.Expr.value->kind != Str_kind &&
			 s->v.Expr.value->kind != Num_kind) {
			VISIT(c, expr, s->v.Expr.value);
			ADDOP(c, POP_TOP);
		}
		break;
	case Pass_kind:
		break;
	case Break_kind:
		if (!c->u->u_nfblocks)
			return compiler_error(c, "'break' outside loop");
		ADDOP(c, BREAK_LOOP);
		break;
	case Continue_kind:
		return compiler_continue(c);
	case With_kind:
		return compiler_with(c, s);
	}
	return 1;
}

static int
unaryop(unaryop_ty op)
{
	switch (op) {
	case Invert:
		return UNARY_INVERT;
	case Not:
		return UNARY_NOT;
	case UAdd:
		return UNARY_POSITIVE;
	case USub:
		return UNARY_NEGATIVE;
	}
	return 0;
}

static int
binop(struct compiler *c, operator_ty op)
{
	switch (op) {
	case Add:
		return BINARY_ADD;
	case Sub:
		return BINARY_SUBTRACT;
	case Mult:
		return BINARY_MULTIPLY;
	case Div:
		if (c->c_flags && c->c_flags->cf_flags & CO_FUTURE_DIVISION)
			return BINARY_TRUE_DIVIDE;
		else
			return BINARY_DIVIDE;
	case Mod:
		return BINARY_MODULO;
	case Pow:
		return BINARY_POWER;
	case LShift:
		return BINARY_LSHIFT;
	case RShift:
		return BINARY_RSHIFT;
	case BitOr:
		return BINARY_OR;
	case BitXor:
		return BINARY_XOR;
	case BitAnd:
		return BINARY_AND;
	case FloorDiv:
		return BINARY_FLOOR_DIVIDE;
	}
	return 0;
}

static int
cmpop(cmpop_ty op)
{
	switch (op) {
	case Eq:
		return PyCmp_EQ;
	case NotEq:
		return PyCmp_NE;
	case Lt:
		return PyCmp_LT;
	case LtE:
		return PyCmp_LE;
	case Gt:
		return PyCmp_GT;
	case GtE:
		return PyCmp_GE;
	case Is:
		return PyCmp_IS;
	case IsNot:
		return PyCmp_IS_NOT;
	case In:
		return PyCmp_IN;
	case NotIn:
		return PyCmp_NOT_IN;
	}
	return PyCmp_BAD;
}

static int
inplace_binop(struct compiler *c, operator_ty op)
{
	switch (op) {
	case Add:
		return INPLACE_ADD;
	case Sub:
		return INPLACE_SUBTRACT;
	case Mult:
		return INPLACE_MULTIPLY;
	case Div:
		if (c->c_flags && c->c_flags->cf_flags & CO_FUTURE_DIVISION)
			return INPLACE_TRUE_DIVIDE;
		else
			return INPLACE_DIVIDE;
	case Mod:
		return INPLACE_MODULO;
	case Pow:
		return INPLACE_POWER;
	case LShift:
		return INPLACE_LSHIFT;
	case RShift:
		return INPLACE_RSHIFT;
	case BitOr:
		return INPLACE_OR;
	case BitXor:
		return INPLACE_XOR;
	case BitAnd:
		return INPLACE_AND;
	case FloorDiv:
		return INPLACE_FLOOR_DIVIDE;
	}
	PyErr_Format(PyExc_SystemError,
		     "inplace binary op %d should not be possible", op);
	return 0;
}

static int
compiler_nameop(struct compiler *c, identifier name, expr_context_ty ctx)
{
	int op, scope, arg;
	enum { OP_FAST, OP_GLOBAL, OP_DEREF, OP_NAME } optype;

	PyObject *dict = c->u->u_names;
	PyObject *mangled;
	/* XXX AugStore isn't used anywhere! */

	/* First check for assignment to __debug__. Param? */
	if ((ctx == Store || ctx == AugStore || ctx == Del)
	    && !strcmp(PyString_AS_STRING(name), "__debug__")) {
		return compiler_error(c, "can not assign to __debug__");
	}

	mangled = _Py_Mangle(c->u->u_private, name);
	if (!mangled)
		return 0;

	op = 0;
	optype = OP_NAME;
	scope = PyST_GetScope(c->u->u_ste, mangled);
	switch (scope) {
	case FREE:
		dict = c->u->u_freevars;
		optype = OP_DEREF;
		break;
	case CELL:
		dict = c->u->u_cellvars;
		optype = OP_DEREF;
		break;
	case LOCAL:
		if (c->u->u_ste->ste_type == FunctionBlock)
			optype = OP_FAST;
		break;
	case GLOBAL_IMPLICIT:
		if (c->u->u_ste->ste_type == FunctionBlock &&
			!c->u->u_ste->ste_unoptimized)
			optype = OP_GLOBAL;
		break;
	case GLOBAL_EXPLICIT:
		optype = OP_GLOBAL;
		break;
	default:
		/* scope can be 0 */
		break;
	}

	/* XXX Leave assert here, but handle __doc__ and the like better */
	assert(scope || PyString_AS_STRING(name)[0] == '_');

	switch (optype) {
	case OP_DEREF:
		switch (ctx) {
		case Load: op = LOAD_DEREF; break;
		case Store: op = STORE_DEREF; break;
		case AugLoad:
		case AugStore:
			break;
		case Del:
			PyErr_Format(PyExc_SyntaxError,
				     "can not delete variable '%s' referenced "
				     "in nested scope",
				     PyString_AS_STRING(name));
			Py_DECREF(mangled);
			return 0;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
					"param invalid for deref variable");
			return 0;
		}
		break;
	case OP_FAST:
		switch (ctx) {
		case Load: op = LOAD_FAST; break;
		case Store: op = STORE_FAST; break;
		case Del: op = DELETE_FAST; break;
		case AugLoad:
		case AugStore:
			break;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
					"param invalid for local variable");
			return 0;
		}
		ADDOP_O(c, op, mangled, varnames);
		Py_DECREF(mangled);
		return 1;
	case OP_GLOBAL:
		switch (ctx) {
		case Load: op = LOAD_GLOBAL; break;
		case Store: op = STORE_GLOBAL; break;
		case Del: op = DELETE_GLOBAL; break;
		case AugLoad:
		case AugStore:
			break;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
					"param invalid for global variable");
			return 0;
		}
		break;
	case OP_NAME:
		switch (ctx) {
		case Load: op = LOAD_NAME; break;
		case Store: op = STORE_NAME; break;
		case Del: op = DELETE_NAME; break;
		case AugLoad:
		case AugStore:
			break;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
					"param invalid for name variable");
			return 0;
		}
		break;
	}

	assert(op);
	arg = compiler_add_o(c, dict, mangled);
	Py_DECREF(mangled);
	if (arg < 0)
		return 0;
	return compiler_addop_i(c, op, arg);
}

static int
compiler_boolop(struct compiler *c, expr_ty e)
{
	basicblock *end;
	int jumpi, i, n;
	asdl_seq *s;

	assert(e->kind == BoolOp_kind);
	if (e->v.BoolOp.op == And)
		jumpi = JUMP_IF_FALSE;
	else
		jumpi = JUMP_IF_TRUE;
	end = compiler_new_block(c);
	if (end == NULL)
		return 0;
	s = e->v.BoolOp.values;
	n = asdl_seq_LEN(s) - 1;
	assert(n >= 0);
	for (i = 0; i < n; ++i) {
		VISIT(c, expr, (expr_ty)asdl_seq_GET(s, i));
		ADDOP_JREL(c, jumpi, end);
		ADDOP(c, POP_TOP)
	}
	VISIT(c, expr, (expr_ty)asdl_seq_GET(s, n));
	compiler_use_next_block(c, end);
	return 1;
}

static int
compiler_list(struct compiler *c, expr_ty e)
{
	int n = asdl_seq_LEN(e->v.List.elts);
	if (e->v.List.ctx == Store) {
		ADDOP_I(c, UNPACK_SEQUENCE, n);
	}
	VISIT_SEQ(c, expr, e->v.List.elts);
	if (e->v.List.ctx == Load) {
		ADDOP_I(c, BUILD_LIST, n);
	}
	return 1;
}

static int
compiler_tuple(struct compiler *c, expr_ty e)
{
	int n = asdl_seq_LEN(e->v.Tuple.elts);
	if (e->v.Tuple.ctx == Store) {
		ADDOP_I(c, UNPACK_SEQUENCE, n);
	}
	VISIT_SEQ(c, expr, e->v.Tuple.elts);
	if (e->v.Tuple.ctx == Load) {
		ADDOP_I(c, BUILD_TUPLE, n);
	}
	return 1;
}

static int
compiler_compare(struct compiler *c, expr_ty e)
{
	int i, n;
	basicblock *cleanup = NULL;

	/* XXX the logic can be cleaned up for 1 or multiple comparisons */
	VISIT(c, expr, e->v.Compare.left);
	n = asdl_seq_LEN(e->v.Compare.ops);
	assert(n > 0);
	if (n > 1) {
		cleanup = compiler_new_block(c);
		if (cleanup == NULL)
		    return 0;
		VISIT(c, expr, 
                        (expr_ty)asdl_seq_GET(e->v.Compare.comparators, 0));
	}
	for (i = 1; i < n; i++) {
		ADDOP(c, DUP_TOP);
		ADDOP(c, ROT_THREE);
		ADDOP_I(c, COMPARE_OP,
			cmpop((cmpop_ty)(asdl_seq_GET(
                                                  e->v.Compare.ops, i - 1))));
		ADDOP_JREL(c, JUMP_IF_FALSE, cleanup);
		NEXT_BLOCK(c);
		ADDOP(c, POP_TOP);
		if (i < (n - 1))
		    VISIT(c, expr, 
                            (expr_ty)asdl_seq_GET(e->v.Compare.comparators, i));
	}
	VISIT(c, expr, (expr_ty)asdl_seq_GET(e->v.Compare.comparators, n - 1));
	ADDOP_I(c, COMPARE_OP,
	       cmpop((cmpop_ty)(asdl_seq_GET(e->v.Compare.ops, n - 1))));
	if (n > 1) {
		basicblock *end = compiler_new_block(c);
		if (end == NULL)
		    return 0;
		ADDOP_JREL(c, JUMP_FORWARD, end);
		compiler_use_next_block(c, cleanup);
		ADDOP(c, ROT_TWO);
		ADDOP(c, POP_TOP);
		compiler_use_next_block(c, end);
	}
	return 1;
}
#undef CMPCAST

static int
compiler_call(struct compiler *c, expr_ty e)
{
	int n, code = 0;

	VISIT(c, expr, e->v.Call.func);
	n = asdl_seq_LEN(e->v.Call.args);
	VISIT_SEQ(c, expr, e->v.Call.args);
	if (e->v.Call.keywords) {
		VISIT_SEQ(c, keyword, e->v.Call.keywords);
		n |= asdl_seq_LEN(e->v.Call.keywords) << 8;
	}
	if (e->v.Call.starargs) {
		VISIT(c, expr, e->v.Call.starargs);
		code |= 1;
	}
	if (e->v.Call.kwargs) {
		VISIT(c, expr, e->v.Call.kwargs);
		code |= 2;
	}
	switch (code) {
	case 0:
		ADDOP_I(c, CALL_FUNCTION, n);
		break;
	case 1:
		ADDOP_I(c, CALL_FUNCTION_VAR, n);
		break;
	case 2:
		ADDOP_I(c, CALL_FUNCTION_KW, n);
		break;
	case 3:
		ADDOP_I(c, CALL_FUNCTION_VAR_KW, n);
		break;
	}
	return 1;
}

static int
compiler_listcomp_generator(struct compiler *c, PyObject *tmpname,
			    asdl_seq *generators, int gen_index, 
			    expr_ty elt)
{
	/* generate code for the iterator, then each of the ifs,
	   and then write to the element */

	comprehension_ty l;
	basicblock *start, *anchor, *skip, *if_cleanup;
	int i, n;

	start = compiler_new_block(c);
	skip = compiler_new_block(c);
	if_cleanup = compiler_new_block(c);
	anchor = compiler_new_block(c);

	if (start == NULL || skip == NULL || if_cleanup == NULL ||
		anchor == NULL)
	    return 0;

	l = (comprehension_ty)asdl_seq_GET(generators, gen_index);
	VISIT(c, expr, l->iter);
	ADDOP(c, GET_ITER);
	compiler_use_next_block(c, start);
	ADDOP_JREL(c, FOR_ITER, anchor);
	NEXT_BLOCK(c);
	VISIT(c, expr, l->target);

	/* XXX this needs to be cleaned up...a lot! */
	n = asdl_seq_LEN(l->ifs);
	for (i = 0; i < n; i++) {
		expr_ty e = (expr_ty)asdl_seq_GET(l->ifs, i);
		VISIT(c, expr, e);
		ADDOP_JREL(c, JUMP_IF_FALSE, if_cleanup);
		NEXT_BLOCK(c);
		ADDOP(c, POP_TOP);
	} 

	if (++gen_index < asdl_seq_LEN(generators))
	    if (!compiler_listcomp_generator(c, tmpname, 
					     generators, gen_index, elt))
		return 0;

	/* only append after the last for generator */
	if (gen_index >= asdl_seq_LEN(generators)) {
	    if (!compiler_nameop(c, tmpname, Load))
		return 0;
	    VISIT(c, expr, elt);
	    ADDOP(c, LIST_APPEND);

	    compiler_use_next_block(c, skip);
	}
	for (i = 0; i < n; i++) {
		ADDOP_I(c, JUMP_FORWARD, 1);
		if (i == 0)
		    compiler_use_next_block(c, if_cleanup);
		ADDOP(c, POP_TOP);
	} 
	ADDOP_JABS(c, JUMP_ABSOLUTE, start);
	compiler_use_next_block(c, anchor);
	/* delete the append method added to locals */
	if (gen_index == 1)
	    if (!compiler_nameop(c, tmpname, Del))
		return 0;
	
	return 1;
}

static int
compiler_listcomp(struct compiler *c, expr_ty e)
{
	identifier tmp;
	int rc = 0;
	static identifier append;
	asdl_seq *generators = e->v.ListComp.generators;

	assert(e->kind == ListComp_kind);
	if (!append) {
		append = PyString_InternFromString("append");
		if (!append)
			return 0;
	}
	tmp = compiler_new_tmpname(c);
	if (!tmp)
		return 0;
	ADDOP_I(c, BUILD_LIST, 0);
	ADDOP(c, DUP_TOP);
	if (compiler_nameop(c, tmp, Store))
	    rc = compiler_listcomp_generator(c, tmp, generators, 0, 
					     e->v.ListComp.elt);
	Py_DECREF(tmp);
	return rc;
}

static int
compiler_genexp_generator(struct compiler *c,
			  asdl_seq *generators, int gen_index, 
			  expr_ty elt)
{
	/* generate code for the iterator, then each of the ifs,
	   and then write to the element */

	comprehension_ty ge;
	basicblock *start, *anchor, *skip, *if_cleanup, *end;
	int i, n;

	start = compiler_new_block(c);
	skip = compiler_new_block(c);
	if_cleanup = compiler_new_block(c);
	anchor = compiler_new_block(c);
	end = compiler_new_block(c);

	if (start == NULL || skip == NULL || if_cleanup == NULL ||
	    anchor == NULL || end == NULL)
		return 0;

	ge = (comprehension_ty)asdl_seq_GET(generators, gen_index);
	ADDOP_JREL(c, SETUP_LOOP, end);
	if (!compiler_push_fblock(c, LOOP, start))
		return 0;

	if (gen_index == 0) {
		/* Receive outermost iter as an implicit argument */
		c->u->u_argcount = 1;
		ADDOP_I(c, LOAD_FAST, 0);
	}
	else {
		/* Sub-iter - calculate on the fly */
		VISIT(c, expr, ge->iter);
		ADDOP(c, GET_ITER);
	}
	compiler_use_next_block(c, start);
	ADDOP_JREL(c, FOR_ITER, anchor);
	NEXT_BLOCK(c);
	VISIT(c, expr, ge->target);

	/* XXX this needs to be cleaned up...a lot! */
	n = asdl_seq_LEN(ge->ifs);
	for (i = 0; i < n; i++) {
		expr_ty e = (expr_ty)asdl_seq_GET(ge->ifs, i);
		VISIT(c, expr, e);
		ADDOP_JREL(c, JUMP_IF_FALSE, if_cleanup);
		NEXT_BLOCK(c);
		ADDOP(c, POP_TOP);
	} 

	if (++gen_index < asdl_seq_LEN(generators))
		if (!compiler_genexp_generator(c, generators, gen_index, elt))
			return 0;

	/* only append after the last 'for' generator */
	if (gen_index >= asdl_seq_LEN(generators)) {
		VISIT(c, expr, elt);
		ADDOP(c, YIELD_VALUE);
		ADDOP(c, POP_TOP);

		compiler_use_next_block(c, skip);
	}
	for (i = 0; i < n; i++) {
		ADDOP_I(c, JUMP_FORWARD, 1);
		if (i == 0)
			compiler_use_next_block(c, if_cleanup);

		ADDOP(c, POP_TOP);
	} 
	ADDOP_JABS(c, JUMP_ABSOLUTE, start);
	compiler_use_next_block(c, anchor);
	ADDOP(c, POP_BLOCK);
	compiler_pop_fblock(c, LOOP, start);
	compiler_use_next_block(c, end);

	return 1;
}

static int
compiler_genexp(struct compiler *c, expr_ty e)
{
	static identifier name;
	PyCodeObject *co;
	expr_ty outermost_iter = ((comprehension_ty)
				 (asdl_seq_GET(e->v.GeneratorExp.generators,
					       0)))->iter;

	if (!name) {
		name = PyString_FromString("<genexpr>");
		if (!name)
			return 0;
	}

	if (!compiler_enter_scope(c, name, (void *)e, e->lineno))
		return 0;
	compiler_genexp_generator(c, e->v.GeneratorExp.generators, 0,
				  e->v.GeneratorExp.elt);
	co = assemble(c, 1);
	compiler_exit_scope(c);
	if (co == NULL)
		return 0;

	compiler_make_closure(c, co, 0);
	Py_DECREF(co);

	VISIT(c, expr, outermost_iter);
	ADDOP(c, GET_ITER);
	ADDOP_I(c, CALL_FUNCTION, 1);

	return 1;
}

static int
compiler_visit_keyword(struct compiler *c, keyword_ty k)
{
	ADDOP_O(c, LOAD_CONST, k->arg, consts);
	VISIT(c, expr, k->value);
	return 1;
}

/* Test whether expression is constant.	 For constants, report
   whether they are true or false.

   Return values: 1 for true, 0 for false, -1 for non-constant.
 */

static int
expr_constant(expr_ty e)
{
	switch (e->kind) {
	case Num_kind:
		return PyObject_IsTrue(e->v.Num.n);
	case Str_kind:
		return PyObject_IsTrue(e->v.Str.s);
	case Name_kind:
		/* __debug__ is not assignable, so we can optimize
		 * it away in if and while statements */
		if (strcmp(PyString_AS_STRING(e->v.Name.id),
		           "__debug__") == 0)
			   return ! Py_OptimizeFlag;
		/* fall through */
	default:
		return -1;
	}
}

/*
   Implements the with statement from PEP 343.

   The semantics outlined in that PEP are as follows:  

   with EXPR as VAR:
       BLOCK
  
   It is implemented roughly as:
  
   context = EXPR
   exit = context.__exit__  # not calling it
   value = context.__enter__()
   try:
       VAR = value  # if VAR present in the syntax
       BLOCK
   finally:
       if an exception was raised:
	   exc = copy of (exception, instance, traceback)
       else:
	   exc = (None, None, None)
       exit(*exc)
 */
static int
compiler_with(struct compiler *c, stmt_ty s)
{
    static identifier enter_attr, exit_attr;
    basicblock *block, *finally;
    identifier tmpexit, tmpvalue = NULL;

    assert(s->kind == With_kind);

    if (!enter_attr) {
	enter_attr = PyString_InternFromString("__enter__");
	if (!enter_attr)
	    return 0;
    }
    if (!exit_attr) {
	exit_attr = PyString_InternFromString("__exit__");
	if (!exit_attr)
	    return 0;
    }

    block = compiler_new_block(c);
    finally = compiler_new_block(c);
    if (!block || !finally)
	return 0;

    /* Create a temporary variable to hold context.__exit__ */
    tmpexit = compiler_new_tmpname(c);
    if (tmpexit == NULL)
	return 0;
    PyArena_AddPyObject(c->c_arena, tmpexit);

    if (s->v.With.optional_vars) {
	/* Create a temporary variable to hold context.__enter__().
	   We need to do this rather than preserving it on the stack
	   because SETUP_FINALLY remembers the stack level.
	   We need to do the assignment *inside* the try/finally
	   so that context.__exit__() is called when the assignment
	   fails.  But we need to call context.__enter__() *before*
	   the try/finally so that if it fails we won't call
	   context.__exit__().
	*/
	tmpvalue = compiler_new_tmpname(c);
	if (tmpvalue == NULL)
	    return 0;
	PyArena_AddPyObject(c->c_arena, tmpvalue);
    }

    /* Evaluate EXPR */
    VISIT(c, expr, s->v.With.context_expr);

    /* Squirrel away context.__exit__  */
    ADDOP(c, DUP_TOP);
    ADDOP_O(c, LOAD_ATTR, exit_attr, names);
    if (!compiler_nameop(c, tmpexit, Store))
	return 0;

    /* Call context.__enter__() */
    ADDOP_O(c, LOAD_ATTR, enter_attr, names);
    ADDOP_I(c, CALL_FUNCTION, 0);

    if (s->v.With.optional_vars) {
	/* Store it in tmpvalue */
	if (!compiler_nameop(c, tmpvalue, Store))
	    return 0;
    }
    else {
	/* Discard result from context.__enter__() */
	ADDOP(c, POP_TOP);
    }

    /* Start the try block */
    ADDOP_JREL(c, SETUP_FINALLY, finally);

    compiler_use_next_block(c, block);
    if (!compiler_push_fblock(c, FINALLY_TRY, block)) {
	return 0;
    }

    if (s->v.With.optional_vars) {
	/* Bind saved result of context.__enter__() to VAR */
	if (!compiler_nameop(c, tmpvalue, Load) ||
	    !compiler_nameop(c, tmpvalue, Del))
	  return 0;
	VISIT(c, expr, s->v.With.optional_vars);
    }

    /* BLOCK code */
    VISIT_SEQ(c, stmt, s->v.With.body);

    /* End of try block; start the finally block */
    ADDOP(c, POP_BLOCK);
    compiler_pop_fblock(c, FINALLY_TRY, block);

    ADDOP_O(c, LOAD_CONST, Py_None, consts);
    compiler_use_next_block(c, finally);
    if (!compiler_push_fblock(c, FINALLY_END, finally))
	return 0;

    /* Finally block starts; push tmpexit and issue our magic opcode. */
    if (!compiler_nameop(c, tmpexit, Load) ||
	!compiler_nameop(c, tmpexit, Del))
	return 0;
    ADDOP(c, WITH_CLEANUP);

    /* Finally block ends. */
    ADDOP(c, END_FINALLY);
    compiler_pop_fblock(c, FINALLY_END, finally);
    return 1;
}

static int
compiler_visit_expr(struct compiler *c, expr_ty e)
{
	int i, n;

    /* If expr e has a different line number than the last expr/stmt,
       set a new line number for the next instruction.
       */
	if (e->lineno > c->u->u_lineno) {
		c->u->u_lineno = e->lineno;
		c->u->u_lineno_set = false;
	}
	switch (e->kind) {
	case BoolOp_kind:
		return compiler_boolop(c, e);
	case BinOp_kind:
		VISIT(c, expr, e->v.BinOp.left);
		VISIT(c, expr, e->v.BinOp.right);
		ADDOP(c, binop(c, e->v.BinOp.op));
		break;
	case UnaryOp_kind:
		VISIT(c, expr, e->v.UnaryOp.operand);
		ADDOP(c, unaryop(e->v.UnaryOp.op));
		break;
	case Lambda_kind:
		return compiler_lambda(c, e);
	case IfExp_kind:
		return compiler_ifexp(c, e);
	case Dict_kind:
		/* XXX get rid of arg? */
		ADDOP_I(c, BUILD_MAP, 0);
		n = asdl_seq_LEN(e->v.Dict.values);
		/* We must arrange things just right for STORE_SUBSCR.
		   It wants the stack to look like (value) (dict) (key) */
		for (i = 0; i < n; i++) {
			ADDOP(c, DUP_TOP);
			VISIT(c, expr, 
                                (expr_ty)asdl_seq_GET(e->v.Dict.values, i));
			ADDOP(c, ROT_TWO);
			VISIT(c, expr, 
                                (expr_ty)asdl_seq_GET(e->v.Dict.keys, i));
			ADDOP(c, STORE_SUBSCR);
		}
		break;
	case ListComp_kind:
		return compiler_listcomp(c, e);
	case GeneratorExp_kind:
		return compiler_genexp(c, e);
	case Yield_kind:
		if (c->u->u_ste->ste_type != FunctionBlock)
			return compiler_error(c, "'yield' outside function");
		/*
		for (i = 0; i < c->u->u_nfblocks; i++) {
			if (c->u->u_fblock[i].fb_type == FINALLY_TRY)
				return compiler_error(
					c, "'yield' not allowed in a 'try' "
					"block with a 'finally' clause");
		}
		*/
		if (e->v.Yield.value) {
			VISIT(c, expr, e->v.Yield.value);
		}
		else {
			ADDOP_O(c, LOAD_CONST, Py_None, consts);
		}
		ADDOP(c, YIELD_VALUE);
		break;
	case Compare_kind:
		return compiler_compare(c, e);
	case Call_kind:
		return compiler_call(c, e);
	case Repr_kind:
		VISIT(c, expr, e->v.Repr.value);
		ADDOP(c, UNARY_CONVERT);
		break;
	case Num_kind:
		ADDOP_O(c, LOAD_CONST, e->v.Num.n, consts);
		break;
	case Str_kind:
		ADDOP_O(c, LOAD_CONST, e->v.Str.s, consts);
		break;
	/* The following exprs can be assignment targets. */
	case Attribute_kind:
		if (e->v.Attribute.ctx != AugStore)
			VISIT(c, expr, e->v.Attribute.value);
		switch (e->v.Attribute.ctx) {
		case AugLoad:
			ADDOP(c, DUP_TOP);
			/* Fall through to load */
		case Load:
			ADDOP_NAME(c, LOAD_ATTR, e->v.Attribute.attr, names);
			break;
		case AugStore:
			ADDOP(c, ROT_TWO);
			/* Fall through to save */
		case Store:
			ADDOP_NAME(c, STORE_ATTR, e->v.Attribute.attr, names);
			break;
		case Del:
			ADDOP_NAME(c, DELETE_ATTR, e->v.Attribute.attr, names);
			break;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
					"param invalid in attribute expression");
			return 0;
		}
		break;
	case Subscript_kind:
		switch (e->v.Subscript.ctx) {
		case AugLoad:
			VISIT(c, expr, e->v.Subscript.value);
			VISIT_SLICE(c, e->v.Subscript.slice, AugLoad);
			break;
		case Load:
			VISIT(c, expr, e->v.Subscript.value);
			VISIT_SLICE(c, e->v.Subscript.slice, Load);
			break;
		case AugStore:
			VISIT_SLICE(c, e->v.Subscript.slice, AugStore);
			break;
		case Store:
			VISIT(c, expr, e->v.Subscript.value);
			VISIT_SLICE(c, e->v.Subscript.slice, Store);
			break;
		case Del:
			VISIT(c, expr, e->v.Subscript.value);
			VISIT_SLICE(c, e->v.Subscript.slice, Del);
			break;
		case Param:
		default:
			PyErr_SetString(PyExc_SystemError,
				"param invalid in subscript expression");
			return 0;
		}
		break;
	case Name_kind:
		return compiler_nameop(c, e->v.Name.id, e->v.Name.ctx);
	/* child nodes of List and Tuple will have expr_context set */
	case List_kind:
		return compiler_list(c, e);
	case Tuple_kind:
		return compiler_tuple(c, e);
	}
	return 1;
}

static int
compiler_augassign(struct compiler *c, stmt_ty s)
{
	expr_ty e = s->v.AugAssign.target;
	expr_ty auge;

	assert(s->kind == AugAssign_kind);

	switch (e->kind) {
	case Attribute_kind:
		auge = Attribute(e->v.Attribute.value, e->v.Attribute.attr,
				 AugLoad, e->lineno, e->col_offset, c->c_arena);
		if (auge == NULL)
		    return 0;
		VISIT(c, expr, auge);
		VISIT(c, expr, s->v.AugAssign.value);
		ADDOP(c, inplace_binop(c, s->v.AugAssign.op));
		auge->v.Attribute.ctx = AugStore;
		VISIT(c, expr, auge);
		break;
	case Subscript_kind:
		auge = Subscript(e->v.Subscript.value, e->v.Subscript.slice,
				 AugLoad, e->lineno, e->col_offset, c->c_arena);
		if (auge == NULL)
		    return 0;
		VISIT(c, expr, auge);
		VISIT(c, expr, s->v.AugAssign.value);
		ADDOP(c, inplace_binop(c, s->v.AugAssign.op));
		auge->v.Subscript.ctx = AugStore;
		VISIT(c, expr, auge);
		break;
	case Name_kind:
		if (!compiler_nameop(c, e->v.Name.id, Load))
		    return 0;
		VISIT(c, expr, s->v.AugAssign.value);
		ADDOP(c, inplace_binop(c, s->v.AugAssign.op));
		return compiler_nameop(c, e->v.Name.id, Store);
	default:
		PyErr_Format(PyExc_SystemError, 
			"invalid node type (%d) for augmented assignment",
			e->kind);
		return 0;
	}
	return 1;
}

static int
compiler_push_fblock(struct compiler *c, enum fblocktype t, basicblock *b)
{
	struct fblockinfo *f;
	if (c->u->u_nfblocks >= CO_MAXBLOCKS)
		return 0;
	f = &c->u->u_fblock[c->u->u_nfblocks++];
	f->fb_type = t;
	f->fb_block = b;
	return 1;
}

static void
compiler_pop_fblock(struct compiler *c, enum fblocktype t, basicblock *b)
{
	struct compiler_unit *u = c->u;
	assert(u->u_nfblocks > 0);
	u->u_nfblocks--;
	assert(u->u_fblock[u->u_nfblocks].fb_type == t);
	assert(u->u_fblock[u->u_nfblocks].fb_block == b);
}

/* Raises a SyntaxError and returns 0.
   If something goes wrong, a different exception may be raised.
*/

static int
compiler_error(struct compiler *c, const char *errstr)
{
	PyObject *loc;
	PyObject *u = NULL, *v = NULL;

	loc = PyErr_ProgramText(c->c_filename, c->u->u_lineno);
	if (!loc) {
		Py_INCREF(Py_None);
		loc = Py_None;
	}
	u = Py_BuildValue("(ziOO)", c->c_filename, c->u->u_lineno,
			  Py_None, loc);
	if (!u)
		goto exit;
	v = Py_BuildValue("(zO)", errstr, u);
	if (!v)
		goto exit;
	PyErr_SetObject(PyExc_SyntaxError, v);
 exit:
	Py_DECREF(loc);
	Py_XDECREF(u);
	Py_XDECREF(v);
	return 0;
}

static int
compiler_handle_subscr(struct compiler *c, const char *kind, 
		       expr_context_ty ctx) 
{
	int op = 0;

	/* XXX this code is duplicated */
	switch (ctx) {
		case AugLoad: /* fall through to Load */
		case Load:    op = BINARY_SUBSCR; break;
		case AugStore:/* fall through to Store */
		case Store:   op = STORE_SUBSCR; break;
		case Del:     op = DELETE_SUBSCR; break;
		case Param:
			PyErr_Format(PyExc_SystemError, 
				     "invalid %s kind %d in subscript\n", 
				     kind, ctx);
			return 0;
	}
	if (ctx == AugLoad) {
		ADDOP_I(c, DUP_TOPX, 2);
	}
	else if (ctx == AugStore) {
		ADDOP(c, ROT_THREE);
	}
	ADDOP(c, op);
	return 1;
}

static int
compiler_slice(struct compiler *c, slice_ty s, expr_context_ty ctx)
{
	int n = 2;
	assert(s->kind == Slice_kind);

	/* only handles the cases where BUILD_SLICE is emitted */
	if (s->v.Slice.lower) {
		VISIT(c, expr, s->v.Slice.lower);
	}
	else {
		ADDOP_O(c, LOAD_CONST, Py_None, consts);
	}
		
	if (s->v.Slice.upper) {
		VISIT(c, expr, s->v.Slice.upper);
	}
	else {
		ADDOP_O(c, LOAD_CONST, Py_None, consts);
	}

	if (s->v.Slice.step) {
		n++;
		VISIT(c, expr, s->v.Slice.step);
	}
	ADDOP_I(c, BUILD_SLICE, n);
	return 1;
}

static int
compiler_simple_slice(struct compiler *c, slice_ty s, expr_context_ty ctx)
{
	int op = 0, slice_offset = 0, stack_count = 0;

	assert(s->v.Slice.step == NULL);
	if (s->v.Slice.lower) {
		slice_offset++;
		stack_count++;
		if (ctx != AugStore) 
			VISIT(c, expr, s->v.Slice.lower);
	}
	if (s->v.Slice.upper) {
		slice_offset += 2;
		stack_count++;
		if (ctx != AugStore) 
			VISIT(c, expr, s->v.Slice.upper);
	}

	if (ctx == AugLoad) {
		switch (stack_count) {
		case 0: ADDOP(c, DUP_TOP); break;
		case 1: ADDOP_I(c, DUP_TOPX, 2); break;
		case 2: ADDOP_I(c, DUP_TOPX, 3); break;
		}
	}
	else if (ctx == AugStore) {
		switch (stack_count) {
		case 0: ADDOP(c, ROT_TWO); break;
		case 1: ADDOP(c, ROT_THREE); break;
		case 2: ADDOP(c, ROT_FOUR); break;
		}
	}

	switch (ctx) {
	case AugLoad: /* fall through to Load */
	case Load: op = SLICE; break;
	case AugStore:/* fall through to Store */
	case Store: op = STORE_SLICE; break;
	case Del: op = DELETE_SLICE; break;
	case Param:
	default:
		PyErr_SetString(PyExc_SystemError,
				"param invalid in simple slice");
		return 0;
	}

	ADDOP(c, op + slice_offset);
	return 1;
}

static int
compiler_visit_nested_slice(struct compiler *c, slice_ty s, 
			    expr_context_ty ctx)
{
	switch (s->kind) {
	case Ellipsis_kind:
		ADDOP_O(c, LOAD_CONST, Py_Ellipsis, consts);
		break;
	case Slice_kind:
		return compiler_slice(c, s, ctx);
	case Index_kind:
		VISIT(c, expr, s->v.Index.value);
		break;
	case ExtSlice_kind:
	default:
		PyErr_SetString(PyExc_SystemError,
				"extended slice invalid in nested slice");
		return 0;
	}
	return 1;
}


static int
compiler_visit_slice(struct compiler *c, slice_ty s, expr_context_ty ctx)
{
	char * kindname = NULL;
	switch (s->kind) {
	case Index_kind:
		kindname = "index";
		if (ctx != AugStore) {
			VISIT(c, expr, s->v.Index.value);
		}
		break;
	case Ellipsis_kind:
		kindname = "ellipsis";
		if (ctx != AugStore) {
			ADDOP_O(c, LOAD_CONST, Py_Ellipsis, consts);
		}
		break;
	case Slice_kind:
		kindname = "slice";
		if (!s->v.Slice.step) 
			return compiler_simple_slice(c, s, ctx);
		if (ctx != AugStore) {
			if (!compiler_slice(c, s, ctx))
				return 0;
		}
		break;
	case ExtSlice_kind:
		kindname = "extended slice";
		if (ctx != AugStore) {
			int i, n = asdl_seq_LEN(s->v.ExtSlice.dims);
			for (i = 0; i < n; i++) {
				slice_ty sub = (slice_ty)asdl_seq_GET(
                                        s->v.ExtSlice.dims, i);
				if (!compiler_visit_nested_slice(c, sub, ctx))
					return 0;
			}
			ADDOP_I(c, BUILD_TUPLE, n);
		}
		break;
	default:
		PyErr_Format(PyExc_SystemError,
			     "invalid subscript kind %d", s->kind);
		return 0;
	}
	return compiler_handle_subscr(c, kindname, ctx);
}

/* do depth-first search of basic block graph, starting with block.
   post records the block indices in post-order.

   XXX must handle implicit jumps from one block to next
*/

static void
dfs(struct compiler *c, basicblock *b, struct assembler *a)
{
	int i;
	struct instr *instr = NULL;

	if (b->b_seen)
		return;
	b->b_seen = 1;
	if (b->b_next != NULL)
		dfs(c, b->b_next, a);
	for (i = 0; i < b->b_iused; i++) {
		instr = &b->b_instr[i];
		if (instr->i_jrel || instr->i_jabs)
			dfs(c, instr->i_target, a);
	}
	a->a_postorder[a->a_nblocks++] = b;
}

static int
stackdepth_walk(struct compiler *c, basicblock *b, int depth, int maxdepth)
{
	int i;
	struct instr *instr;
	if (b->b_seen || b->b_startdepth >= depth)
		return maxdepth;
	b->b_seen = 1;
	b->b_startdepth = depth;
	for (i = 0; i < b->b_iused; i++) {
		instr = &b->b_instr[i];
		depth += opcode_stack_effect(instr->i_opcode, instr->i_oparg);
		if (depth > maxdepth)
			maxdepth = depth;
		assert(depth >= 0); /* invalid code or bug in stackdepth() */
		if (instr->i_jrel || instr->i_jabs) {
			maxdepth = stackdepth_walk(c, instr->i_target,
						   depth, maxdepth);
			if (instr->i_opcode == JUMP_ABSOLUTE ||
			    instr->i_opcode == JUMP_FORWARD) {
				goto out; /* remaining code is dead */
			}
		}
	}
	if (b->b_next)
		maxdepth = stackdepth_walk(c, b->b_next, depth, maxdepth);
out:
	b->b_seen = 0;
	return maxdepth;
}

/* Find the flow path that needs the largest stack.  We assume that
 * cycles in the flow graph have no net effect on the stack depth.
 */
static int
stackdepth(struct compiler *c)
{
	basicblock *b, *entryblock;
	entryblock = NULL;
	for (b = c->u->u_blocks; b != NULL; b = b->b_list) {
		b->b_seen = 0;
		b->b_startdepth = INT_MIN;
		entryblock = b;
	}
	if (!entryblock)
		return 0;
	return stackdepth_walk(c, entryblock, 0, 0);
}

static int
assemble_init(struct assembler *a, int nblocks, int firstlineno)
{
	memset(a, 0, sizeof(struct assembler));
	a->a_lineno = firstlineno;
	a->a_bytecode = PyString_FromStringAndSize(NULL, DEFAULT_CODE_SIZE);
	if (!a->a_bytecode)
		return 0;
	a->a_lnotab = PyString_FromStringAndSize(NULL, DEFAULT_LNOTAB_SIZE);
	if (!a->a_lnotab)
		return 0;
	a->a_postorder = (basicblock **)PyObject_Malloc(
					    sizeof(basicblock *) * nblocks);
	if (!a->a_postorder) {
		PyErr_NoMemory();
		return 0;
	}
	return 1;
}

static void
assemble_free(struct assembler *a)
{
	Py_XDECREF(a->a_bytecode);
	Py_XDECREF(a->a_lnotab);
	if (a->a_postorder)
		PyObject_Free(a->a_postorder);
}

/* Return the size of a basic block in bytes. */

static int
instrsize(struct instr *instr)
{
	if (!instr->i_hasarg)
		return 1;
	if (instr->i_oparg > 0xffff)
		return 6;
	return 3;
}

static int
blocksize(basicblock *b)
{
	int i;
	int size = 0;

	for (i = 0; i < b->b_iused; i++)
		size += instrsize(&b->b_instr[i]);
	return size;
}

/* All about a_lnotab.

c_lnotab is an array of unsigned bytes disguised as a Python string.
It is used to map bytecode offsets to source code line #s (when needed
for tracebacks).

The array is conceptually a list of
    (bytecode offset increment, line number increment)
pairs.	The details are important and delicate, best illustrated by example:

    byte code offset	source code line number
	0		    1
	6		    2
       50		    7
      350		  307
      361		  308

The first trick is that these numbers aren't stored, only the increments
from one row to the next (this doesn't really work, but it's a start):

    0, 1,  6, 1,  44, 5,  300, 300,  11, 1

The second trick is that an unsigned byte can't hold negative values, or
values larger than 255, so (a) there's a deep assumption that byte code
offsets and their corresponding line #s both increase monotonically, and (b)
if at least one column jumps by more than 255 from one row to the next, more
than one pair is written to the table. In case #b, there's no way to know
from looking at the table later how many were written.	That's the delicate
part.  A user of c_lnotab desiring to find the source line number
corresponding to a bytecode address A should do something like this

    lineno = addr = 0
    for addr_incr, line_incr in c_lnotab:
	addr += addr_incr
	if addr > A:
	    return lineno
	lineno += line_incr

In order for this to work, when the addr field increments by more than 255,
the line # increment in each pair generated must be 0 until the remaining addr
increment is < 256.  So, in the example above, assemble_lnotab (it used
to be called com_set_lineno) should not (as was actually done until 2.2)
expand 300, 300 to 255, 255, 45, 45, 
            but to 255,   0, 45, 255, 0, 45.
*/

static int
assemble_lnotab(struct assembler *a, struct instr *i)
{
	int d_bytecode, d_lineno;
	int len;
	unsigned char *lnotab;

	d_bytecode = a->a_offset - a->a_lineno_off;
	d_lineno = i->i_lineno - a->a_lineno;

	assert(d_bytecode >= 0);
	assert(d_lineno >= 0);

	/* XXX(nnorwitz): is there a better way to handle this?
	   for loops are special, we want to be able to trace them
	   each time around, so we need to set an extra line number. */
	if (d_lineno == 0 && i->i_opcode != FOR_ITER)
		return 1;

	if (d_bytecode > 255) {
		int j, nbytes, ncodes = d_bytecode / 255;
		nbytes = a->a_lnotab_off + 2 * ncodes;
		len = PyString_GET_SIZE(a->a_lnotab);
		if (nbytes >= len) {
			if (len * 2 < nbytes)
				len = nbytes;
			else
				len *= 2;
			if (_PyString_Resize(&a->a_lnotab, len) < 0)
				return 0;
		}
		lnotab = (unsigned char *)
			   PyString_AS_STRING(a->a_lnotab) + a->a_lnotab_off;
		for (j = 0; j < ncodes; j++) {
			*lnotab++ = 255;
			*lnotab++ = 0;
		}
		d_bytecode -= ncodes * 255;
		a->a_lnotab_off += ncodes * 2;
	}
	assert(d_bytecode <= 255);
	if (d_lineno > 255) {
		int j, nbytes, ncodes = d_lineno / 255;
		nbytes = a->a_lnotab_off + 2 * ncodes;
		len = PyString_GET_SIZE(a->a_lnotab);
		if (nbytes >= len) {
			if (len * 2 < nbytes)
				len = nbytes;
			else
				len *= 2;
			if (_PyString_Resize(&a->a_lnotab, len) < 0)
				return 0;
		}
		lnotab = (unsigned char *)
			   PyString_AS_STRING(a->a_lnotab) + a->a_lnotab_off;
		*lnotab++ = d_bytecode;
		*lnotab++ = 255;
		d_bytecode = 0;
		for (j = 1; j < ncodes; j++) {
			*lnotab++ = 0;
			*lnotab++ = 255;
		}
		d_lineno -= ncodes * 255;
		a->a_lnotab_off += ncodes * 2;
	}

	len = PyString_GET_SIZE(a->a_lnotab);
	if (a->a_lnotab_off + 2 >= len) {
		if (_PyString_Resize(&a->a_lnotab, len * 2) < 0)
			return 0;
	}
	lnotab = (unsigned char *)
			PyString_AS_STRING(a->a_lnotab) + a->a_lnotab_off;

	a->a_lnotab_off += 2;
	if (d_bytecode) {
		*lnotab++ = d_bytecode;
		*lnotab++ = d_lineno;
	}
	else {	/* First line of a block; def stmt, etc. */
		*lnotab++ = 0;
		*lnotab++ = d_lineno;
	}
	a->a_lineno = i->i_lineno;
	a->a_lineno_off = a->a_offset;
	return 1;
}

/* assemble_emit()
   Extend the bytecode with a new instruction.
   Update lnotab if necessary.
*/

static int
assemble_emit(struct assembler *a, struct instr *i)
{
	int size, arg = 0, ext = 0;
	Py_ssize_t len = PyString_GET_SIZE(a->a_bytecode);
	char *code;

	size = instrsize(i);
	if (i->i_hasarg) {
		arg = i->i_oparg;
		ext = arg >> 16;
	}
	if (i->i_lineno && !assemble_lnotab(a, i))
		return 0;
	if (a->a_offset + size >= len) {
		if (_PyString_Resize(&a->a_bytecode, len * 2) < 0)
		    return 0;
	}
	code = PyString_AS_STRING(a->a_bytecode) + a->a_offset;
	a->a_offset += size;
	if (size == 6) {
		assert(i->i_hasarg);
		*code++ = (char)EXTENDED_ARG;
		*code++ = ext & 0xff;
		*code++ = ext >> 8;
		arg &= 0xffff;
	}
	*code++ = i->i_opcode;
	if (i->i_hasarg) {
		assert(size == 3 || size == 6);
		*code++ = arg & 0xff;
		*code++ = arg >> 8;
	}
	return 1;
}

static void
assemble_jump_offsets(struct assembler *a, struct compiler *c)
{
	basicblock *b;
	int bsize, totsize, extended_arg_count, last_extended_arg_count = 0;
	int i;

	/* Compute the size of each block and fixup jump args.
	   Replace block pointer with position in bytecode. */
start:
	totsize = 0;
	for (i = a->a_nblocks - 1; i >= 0; i--) {
		b = a->a_postorder[i];
		bsize = blocksize(b);
		b->b_offset = totsize;
		totsize += bsize;
	}
	extended_arg_count = 0;
	for (b = c->u->u_blocks; b != NULL; b = b->b_list) {
		bsize = b->b_offset;
		for (i = 0; i < b->b_iused; i++) {
			struct instr *instr = &b->b_instr[i];
			/* Relative jumps are computed relative to
			   the instruction pointer after fetching
			   the jump instruction.
			*/
			bsize += instrsize(instr);
			if (instr->i_jabs)
				instr->i_oparg = instr->i_target->b_offset;
			else if (instr->i_jrel) {
				int delta = instr->i_target->b_offset - bsize;
				instr->i_oparg = delta;
			}
			else
				continue;
			if (instr->i_oparg > 0xffff)
				extended_arg_count++;
		}
	}

	/* XXX: This is an awful hack that could hurt performance, but
		on the bright side it should work until we come up
		with a better solution.

		In the meantime, should the goto be dropped in favor
		of a loop?

		The issue is that in the first loop blocksize() is called
		which calls instrsize() which requires i_oparg be set
		appropriately.	There is a bootstrap problem because
		i_oparg is calculated in the second loop above.

		So we loop until we stop seeing new EXTENDED_ARGs.
		The only EXTENDED_ARGs that could be popping up are
		ones in jump instructions.  So this should converge
		fairly quickly.
	*/
	if (last_extended_arg_count != extended_arg_count) {
		last_extended_arg_count = extended_arg_count;
		goto start;
	}
}

static PyObject *
dict_keys_inorder(PyObject *dict, int offset)
{
	PyObject *tuple, *k, *v;
	Py_ssize_t i, pos = 0, size = PyDict_Size(dict);

	tuple = PyTuple_New(size);
	if (tuple == NULL)
		return NULL;
	while (PyDict_Next(dict, &pos, &k, &v)) {
		i = PyInt_AS_LONG(v);
		k = PyTuple_GET_ITEM(k, 0);
		Py_INCREF(k);
		assert((i - offset) < size);
		assert((i - offset) >= 0);
		PyTuple_SET_ITEM(tuple, i - offset, k);
	}
	return tuple;
}

static int
compute_code_flags(struct compiler *c)
{
	PySTEntryObject *ste = c->u->u_ste;
	int flags = 0, n;
	if (ste->ste_type != ModuleBlock)
		flags |= CO_NEWLOCALS;
	if (ste->ste_type == FunctionBlock) {
		if (!ste->ste_unoptimized)
			flags |= CO_OPTIMIZED;
		if (ste->ste_nested)
			flags |= CO_NESTED;
		if (ste->ste_generator)
			flags |= CO_GENERATOR;
	}
	if (ste->ste_varargs)
		flags |= CO_VARARGS;
	if (ste->ste_varkeywords)
		flags |= CO_VARKEYWORDS;
	if (ste->ste_generator)
		flags |= CO_GENERATOR;

	/* (Only) inherit compilerflags in PyCF_MASK */
	flags |= (c->c_flags->cf_flags & PyCF_MASK);

	n = PyDict_Size(c->u->u_freevars);
	if (n < 0)
	    return -1;
	if (n == 0) {
	    n = PyDict_Size(c->u->u_cellvars);
	    if (n < 0)
		return -1;
	    if (n == 0) {
		flags |= CO_NOFREE;
	    }
	}

	return flags;
}

static PyCodeObject *
makecode(struct compiler *c, struct assembler *a)
{
	PyObject *tmp;
	PyCodeObject *co = NULL;
	PyObject *consts = NULL;
	PyObject *names = NULL;
	PyObject *varnames = NULL;
	PyObject *filename = NULL;
	PyObject *name = NULL;
	PyObject *freevars = NULL;
	PyObject *cellvars = NULL;
	PyObject *bytecode = NULL;
	int nlocals, flags;

	tmp = dict_keys_inorder(c->u->u_consts, 0);
	if (!tmp)
		goto error;
	consts = PySequence_List(tmp); /* optimize_code requires a list */
	Py_DECREF(tmp);

	names = dict_keys_inorder(c->u->u_names, 0);
	varnames = dict_keys_inorder(c->u->u_varnames, 0);
	if (!consts || !names || !varnames)
		goto error;
      
	cellvars = dict_keys_inorder(c->u->u_cellvars, 0);
	if (!cellvars)
	    goto error;
	freevars = dict_keys_inorder(c->u->u_freevars, PyTuple_Size(cellvars));
	if (!freevars)
	    goto error;
	filename = PyString_FromString(c->c_filename);
	if (!filename)
		goto error;

	nlocals = PyDict_Size(c->u->u_varnames);
	flags = compute_code_flags(c);
	if (flags < 0)
		goto error;

	bytecode = optimize_code(a->a_bytecode, consts, names, a->a_lnotab);
	if (!bytecode)
		goto error;

	tmp = PyList_AsTuple(consts); /* PyCode_New requires a tuple */
	if (!tmp)
		goto error;
	Py_DECREF(consts);
	consts = tmp;

	co = PyCode_New(c->u->u_argcount, nlocals, stackdepth(c), flags,
			bytecode, consts, names, varnames,
			freevars, cellvars,
			filename, c->u->u_name,
			c->u->u_firstlineno,
			a->a_lnotab);
 error:
	Py_XDECREF(consts);
	Py_XDECREF(names);
	Py_XDECREF(varnames);
	Py_XDECREF(filename);
	Py_XDECREF(name);
	Py_XDECREF(freevars);
	Py_XDECREF(cellvars);
	Py_XDECREF(bytecode);
	return co;
}


/* For debugging purposes only */
#if 0
static void
dump_instr(const struct instr *i)
{
	const char *jrel = i->i_jrel ? "jrel " : "";
	const char *jabs = i->i_jabs ? "jabs " : "";
	char arg[128];

	*arg = '\0';
	if (i->i_hasarg)
		sprintf(arg, "arg: %d ", i->i_oparg);

	fprintf(stderr, "line: %d, opcode: %d %s%s%s\n", 
			i->i_lineno, i->i_opcode, arg, jabs, jrel);
}

static void
dump_basicblock(const basicblock *b)
{
	const char *seen = b->b_seen ? "seen " : "";
	const char *b_return = b->b_return ? "return " : "";
	fprintf(stderr, "used: %d, depth: %d, offset: %d %s%s\n",
		b->b_iused, b->b_startdepth, b->b_offset, seen, b_return);
	if (b->b_instr) {
		int i;
		for (i = 0; i < b->b_iused; i++) {
			fprintf(stderr, "  [%02d] ", i);
			dump_instr(b->b_instr + i);
		}
	}
}
#endif

static PyCodeObject *
assemble(struct compiler *c, int addNone)
{
	basicblock *b, *entryblock;
	struct assembler a;
	int i, j, nblocks;
	PyCodeObject *co = NULL;

	/* Make sure every block that falls off the end returns None.
	   XXX NEXT_BLOCK() isn't quite right, because if the last
	   block ends with a jump or return b_next shouldn't set.
	 */
	if (!c->u->u_curblock->b_return) {
		NEXT_BLOCK(c);
		if (addNone)
			ADDOP_O(c, LOAD_CONST, Py_None, consts);
		ADDOP(c, RETURN_VALUE);
	}

	nblocks = 0;
	entryblock = NULL;
	for (b = c->u->u_blocks; b != NULL; b = b->b_list) {
		nblocks++;
		entryblock = b; 
	}

	/* Set firstlineno if it wasn't explicitly set. */
	if (!c->u->u_firstlineno) {
		if (entryblock && entryblock->b_instr)
			c->u->u_firstlineno = entryblock->b_instr->i_lineno;
		else
			c->u->u_firstlineno = 1;
	}
	if (!assemble_init(&a, nblocks, c->u->u_firstlineno))
		goto error;
	dfs(c, entryblock, &a);

	/* Can't modify the bytecode after computing jump offsets. */
	assemble_jump_offsets(&a, c);

	/* Emit code in reverse postorder from dfs. */
	for (i = a.a_nblocks - 1; i >= 0; i--) {
		b = a.a_postorder[i];
		for (j = 0; j < b->b_iused; j++)
			if (!assemble_emit(&a, &b->b_instr[j]))
				goto error;
	}

	if (_PyString_Resize(&a.a_lnotab, a.a_lnotab_off) < 0)
		goto error;
	if (_PyString_Resize(&a.a_bytecode, a.a_offset) < 0)
		goto error;

	co = makecode(c, &a);
 error:
	assemble_free(&a);
	return co;
}