bpo-40939: Remove the old parser (Part 2) (GH-21005)

Remove some remaining files and Makefile targets for the old parser
This commit is contained in:
Lysandros Nikolaou 2020-06-20 21:07:25 +03:00 committed by GitHub
parent 55460ee6dc
commit 314858e276
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
24 changed files with 5 additions and 3534 deletions

2
.gitattributes vendored
View File

@ -41,8 +41,6 @@ PC/readme.txt text eol=crlf
# Generated files
# https://github.com/github/linguist#generated-code
Include/graminit.h linguist-generated=true
Python/graminit.h linguist-generated=true
Modules/clinic/*.h linguist-generated=true
Objects/clinic/*.h linguist-generated=true
PC/clinic/*.h linguist-generated=true

View File

@ -6,19 +6,8 @@ extern "C" {
#endif
#include "Python-ast.h" /* mod_ty */
#include "node.h" /* node */
PyAPI_FUNC(int) PyAST_Validate(mod_ty);
PyAPI_FUNC(mod_ty) PyAST_FromNode(
const node *n,
PyCompilerFlags *flags,
const char *filename, /* decoded from the filesystem encoding */
PyArena *arena);
PyAPI_FUNC(mod_ty) PyAST_FromNodeObject(
const node *n,
PyCompilerFlags *flags,
PyObject *filename,
PyArena *arena);
/* _PyAST_ExprAsUnicode is defined in ast_unparse.c */
PyAPI_FUNC(PyObject *) _PyAST_ExprAsUnicode(expr_ty);

View File

@ -1,23 +0,0 @@
#ifndef Py_BITSET_H
#define Py_BITSET_H
#ifdef __cplusplus
extern "C" {
#endif
/* Bitset interface */
#define BYTE char
typedef BYTE *bitset;
#define testbit(ss, ibit) (((ss)[BIT2BYTE(ibit)] & BIT2MASK(ibit)) != 0)
#define BITSPERBYTE (8*sizeof(BYTE))
#define BIT2BYTE(ibit) ((ibit) / BITSPERBYTE)
#define BIT2SHIFT(ibit) ((ibit) % BITSPERBYTE)
#define BIT2MASK(ibit) (1 << BIT2SHIFT(ibit))
#ifdef __cplusplus
}
#endif
#endif /* !Py_BITSET_H */

View File

@ -8,10 +8,6 @@ extern "C" {
#endif
/* Public interface */
struct _node; /* Declare the existence of this type */
PyAPI_FUNC(PyCodeObject *) PyNode_Compile(struct _node *, const char *);
/* XXX (ncoghlan): Unprefixed type name in a public API! */
#define PyCF_MASK (CO_FUTURE_DIVISION | CO_FUTURE_ABSOLUTE_IMPORT | \
CO_FUTURE_WITH_STATEMENT | CO_FUTURE_PRINT_FUNCTION | \
CO_FUTURE_UNICODE_LITERALS | CO_FUTURE_BARRY_AS_BDFL | \

View File

@ -1,94 +0,0 @@
/* Generated by Parser/pgen */
#define single_input 256
#define file_input 257
#define eval_input 258
#define decorator 259
#define decorators 260
#define decorated 261
#define async_funcdef 262
#define funcdef 263
#define parameters 264
#define typedargslist 265
#define tfpdef 266
#define varargslist 267
#define vfpdef 268
#define stmt 269
#define simple_stmt 270
#define small_stmt 271
#define expr_stmt 272
#define annassign 273
#define testlist_star_expr 274
#define augassign 275
#define del_stmt 276
#define pass_stmt 277
#define flow_stmt 278
#define break_stmt 279
#define continue_stmt 280
#define return_stmt 281
#define yield_stmt 282
#define raise_stmt 283
#define import_stmt 284
#define import_name 285
#define import_from 286
#define import_as_name 287
#define dotted_as_name 288
#define import_as_names 289
#define dotted_as_names 290
#define dotted_name 291
#define global_stmt 292
#define nonlocal_stmt 293
#define assert_stmt 294
#define compound_stmt 295
#define async_stmt 296
#define if_stmt 297
#define while_stmt 298
#define for_stmt 299
#define try_stmt 300
#define with_stmt 301
#define with_item 302
#define except_clause 303
#define suite 304
#define namedexpr_test 305
#define test 306
#define test_nocond 307
#define lambdef 308
#define lambdef_nocond 309
#define or_test 310
#define and_test 311
#define not_test 312
#define comparison 313
#define comp_op 314
#define star_expr 315
#define expr 316
#define xor_expr 317
#define and_expr 318
#define shift_expr 319
#define arith_expr 320
#define term 321
#define factor 322
#define power 323
#define atom_expr 324
#define atom 325
#define testlist_comp 326
#define trailer 327
#define subscriptlist 328
#define subscript 329
#define sliceop 330
#define exprlist 331
#define testlist 332
#define dictorsetmaker 333
#define classdef 334
#define arglist 335
#define argument 336
#define comp_iter 337
#define sync_comp_for 338
#define comp_for 339
#define comp_if 340
#define encoding_decl 341
#define yield_expr 342
#define yield_arg 343
#define func_body_suite 344
#define func_type_input 345
#define func_type 346
#define typelist 347

View File

@ -1,77 +0,0 @@
/* Grammar interface */
#ifndef Py_GRAMMAR_H
#define Py_GRAMMAR_H
#ifdef __cplusplus
extern "C" {
#endif
#include "bitset.h" /* Sigh... */
/* A label of an arc */
typedef struct {
int lb_type;
const char *lb_str;
} label;
#define EMPTY 0 /* Label number 0 is by definition the empty label */
/* A list of labels */
typedef struct {
int ll_nlabels;
const label *ll_label;
} labellist;
/* An arc from one state to another */
typedef struct {
short a_lbl; /* Label of this arc */
short a_arrow; /* State where this arc goes to */
} arc;
/* A state in a DFA */
typedef struct {
int s_narcs;
const arc *s_arc; /* Array of arcs */
/* Optional accelerators */
int s_lower; /* Lowest label index */
int s_upper; /* Highest label index */
int *s_accel; /* Accelerator */
int s_accept; /* Nonzero for accepting state */
} state;
/* A DFA */
typedef struct {
int d_type; /* Non-terminal this represents */
char *d_name; /* For printing */
int d_nstates;
state *d_state; /* Array of states */
bitset d_first;
} dfa;
/* A grammar */
typedef struct {
int g_ndfas;
const dfa *g_dfa; /* Array of DFAs */
const labellist g_ll;
int g_start; /* Start symbol of the grammar */
int g_accel; /* Set if accelerators present */
} grammar;
/* FUNCTIONS */
const dfa *PyGrammar_FindDFA(grammar *g, int type);
const char *PyGrammar_LabelRepr(label *lb);
void PyGrammar_AddAccelerators(grammar *g);
void PyGrammar_RemoveAccelerators(grammar *);
#ifdef __cplusplus
}
#endif
#endif /* !Py_GRAMMAR_H */

View File

@ -1,47 +0,0 @@
/* Parse tree node interface */
#ifndef Py_NODE_H
#define Py_NODE_H
#ifdef __cplusplus
extern "C" {
#endif
typedef struct _node {
short n_type;
char *n_str;
int n_lineno;
int n_col_offset;
int n_nchildren;
struct _node *n_child;
int n_end_lineno;
int n_end_col_offset;
} node;
PyAPI_FUNC(node *) PyNode_New(int type);
PyAPI_FUNC(int) PyNode_AddChild(node *n, int type,
char *str, int lineno, int col_offset,
int end_lineno, int end_col_offset);
PyAPI_FUNC(void) PyNode_Free(node *n);
#ifndef Py_LIMITED_API
PyAPI_FUNC(Py_ssize_t) _PyNode_SizeOf(node *n);
#endif
/* Node access functions */
#define NCH(n) ((n)->n_nchildren)
#define CHILD(n, i) (&(n)->n_child[i])
#define TYPE(n) ((n)->n_type)
#define STR(n) ((n)->n_str)
#define LINENO(n) ((n)->n_lineno)
/* Assert that the type of a node is what we expect */
#define REQ(n, type) assert(TYPE(n) == (type))
PyAPI_FUNC(void) PyNode_ListTree(node *);
void _PyNode_FinalizeEndPos(node *n); // helper also used in parsetok.c
#ifdef __cplusplus
}
#endif
#endif /* !Py_NODE_H */

View File

@ -1,110 +0,0 @@
/* Parser-tokenizer link interface */
#ifndef Py_LIMITED_API
#ifndef Py_PARSETOK_H
#define Py_PARSETOK_H
#ifdef __cplusplus
extern "C" {
#endif
#include "grammar.h" /* grammar */
#include "node.h" /* node */
typedef struct {
int error;
PyObject *filename;
int lineno;
int offset;
char *text; /* UTF-8-encoded string */
int token;
int expected;
} perrdetail;
#if 0
#define PyPARSE_YIELD_IS_KEYWORD 0x0001
#endif
#define PyPARSE_DONT_IMPLY_DEDENT 0x0002
#if 0
#define PyPARSE_WITH_IS_KEYWORD 0x0003
#define PyPARSE_PRINT_IS_FUNCTION 0x0004
#define PyPARSE_UNICODE_LITERALS 0x0008
#endif
#define PyPARSE_IGNORE_COOKIE 0x0010
#define PyPARSE_BARRY_AS_BDFL 0x0020
#define PyPARSE_TYPE_COMMENTS 0x0040
#define PyPARSE_ASYNC_HACKS 0x0080
PyAPI_FUNC(node *) PyParser_ParseString(const char *, grammar *, int,
perrdetail *);
PyAPI_FUNC(node *) PyParser_ParseFile (FILE *, const char *, grammar *, int,
const char *, const char *,
perrdetail *);
PyAPI_FUNC(node *) PyParser_ParseStringFlags(const char *, grammar *, int,
perrdetail *, int);
PyAPI_FUNC(node *) PyParser_ParseFileFlags(
FILE *fp,
const char *filename, /* decoded from the filesystem encoding */
const char *enc,
grammar *g,
int start,
const char *ps1,
const char *ps2,
perrdetail *err_ret,
int flags);
PyAPI_FUNC(node *) PyParser_ParseFileFlagsEx(
FILE *fp,
const char *filename, /* decoded from the filesystem encoding */
const char *enc,
grammar *g,
int start,
const char *ps1,
const char *ps2,
perrdetail *err_ret,
int *flags);
PyAPI_FUNC(node *) PyParser_ParseFileObject(
FILE *fp,
PyObject *filename,
const char *enc,
grammar *g,
int start,
const char *ps1,
const char *ps2,
perrdetail *err_ret,
int *flags);
PyAPI_FUNC(node *) PyParser_ParseStringFlagsFilename(
const char *s,
const char *filename, /* decoded from the filesystem encoding */
grammar *g,
int start,
perrdetail *err_ret,
int flags);
PyAPI_FUNC(node *) PyParser_ParseStringFlagsFilenameEx(
const char *s,
const char *filename, /* decoded from the filesystem encoding */
grammar *g,
int start,
perrdetail *err_ret,
int *flags);
PyAPI_FUNC(node *) PyParser_ParseStringObject(
const char *s,
PyObject *filename,
grammar *g,
int start,
perrdetail *err_ret,
int *flags);
/* Note that the following functions are defined in pythonrun.c,
not in parsetok.c */
PyAPI_FUNC(void) PyParser_SetError(perrdetail *);
PyAPI_FUNC(void) PyParser_ClearError(perrdetail *);
#ifdef __cplusplus
}
#endif
#endif /* !Py_PARSETOK_H */
#endif /* !Py_LIMITED_API */

View File

@ -1,122 +0,0 @@
"""Non-terminal symbols of Python grammar (from "graminit.h")."""
# This file is automatically generated; please don't muck it up!
#
# To update the symbols in this file, 'cd' to the top directory of
# the python source tree after building the interpreter and run:
#
# python3 Tools/scripts/generate_symbol_py.py Include/graminit.h Lib/symbol.py
#
# or just
#
# make regen-symbol
import warnings
warnings.warn(
"The symbol module is deprecated and will be removed "
"in future versions of Python",
DeprecationWarning,
stacklevel=2,
)
#--start constants--
single_input = 256
file_input = 257
eval_input = 258
decorator = 259
decorators = 260
decorated = 261
async_funcdef = 262
funcdef = 263
parameters = 264
typedargslist = 265
tfpdef = 266
varargslist = 267
vfpdef = 268
stmt = 269
simple_stmt = 270
small_stmt = 271
expr_stmt = 272
annassign = 273
testlist_star_expr = 274
augassign = 275
del_stmt = 276
pass_stmt = 277
flow_stmt = 278
break_stmt = 279
continue_stmt = 280
return_stmt = 281
yield_stmt = 282
raise_stmt = 283
import_stmt = 284
import_name = 285
import_from = 286
import_as_name = 287
dotted_as_name = 288
import_as_names = 289
dotted_as_names = 290
dotted_name = 291
global_stmt = 292
nonlocal_stmt = 293
assert_stmt = 294
compound_stmt = 295
async_stmt = 296
if_stmt = 297
while_stmt = 298
for_stmt = 299
try_stmt = 300
with_stmt = 301
with_item = 302
except_clause = 303
suite = 304
namedexpr_test = 305
test = 306
test_nocond = 307
lambdef = 308
lambdef_nocond = 309
or_test = 310
and_test = 311
not_test = 312
comparison = 313
comp_op = 314
star_expr = 315
expr = 316
xor_expr = 317
and_expr = 318
shift_expr = 319
arith_expr = 320
term = 321
factor = 322
power = 323
atom_expr = 324
atom = 325
testlist_comp = 326
trailer = 327
subscriptlist = 328
subscript = 329
sliceop = 330
exprlist = 331
testlist = 332
dictorsetmaker = 333
classdef = 334
arglist = 335
argument = 336
comp_iter = 337
sync_comp_for = 338
comp_for = 339
comp_if = 340
encoding_decl = 341
yield_expr = 342
yield_arg = 343
func_body_suite = 344
func_type_input = 345
func_type = 346
typelist = 347
#--end constants--
sym_name = {}
for _name, _value in list(globals().items()):
if type(_value) is type(0):
sym_name[_value] = _name
del _name, _value

View File

@ -1,58 +0,0 @@
import unittest
from test import support
import os
import sys
import sysconfig
import subprocess
SYMBOL_FILE = support.findfile('symbol.py')
GEN_SYMBOL_FILE = os.path.join(os.path.dirname(__file__),
'..', '..', 'Tools', 'scripts',
'generate_symbol_py.py')
GRAMMAR_FILE = os.path.join(os.path.dirname(__file__),
'..', '..', 'Include', 'graminit.h')
TEST_PY_FILE = 'symbol_test.py'
class TestSymbolGeneration(unittest.TestCase):
def _copy_file_without_generated_symbols(self, source_file, dest_file):
with open(source_file) as fp:
lines = fp.readlines()
with open(dest_file, 'w') as fp:
fp.writelines(lines[:lines.index("#--start constants--\n") + 1])
fp.writelines(lines[lines.index("#--end constants--\n"):])
def _generate_symbols(self, grammar_file, target_symbol_py_file):
proc = subprocess.Popen([sys.executable,
GEN_SYMBOL_FILE,
grammar_file,
target_symbol_py_file], stderr=subprocess.PIPE)
stderr = proc.communicate()[1]
return proc.returncode, stderr
def compare_files(self, file1, file2):
with open(file1) as fp:
lines1 = fp.readlines()
with open(file2) as fp:
lines2 = fp.readlines()
self.assertEqual(lines1, lines2)
@unittest.skipUnless(sysconfig.is_python_build(),
'test only works from source build directory')
def test_real_grammar_and_symbol_file(self):
output = support.TESTFN
self.addCleanup(support.unlink, output)
self._copy_file_without_generated_symbols(SYMBOL_FILE, output)
exitcode, stderr = self._generate_symbols(GRAMMAR_FILE, output)
self.assertEqual(b'', stderr)
self.assertEqual(0, exitcode)
self.compare_files(SYMBOL_FILE, output)
if __name__ == "__main__":
unittest.main()

View File

@ -344,7 +344,6 @@ PYTHON_OBJS= \
Python/getcopyright.o \
Python/getplatform.o \
Python/getversion.o \
Python/graminit.o \
Python/hamt.o \
Python/hashtable.o \
Python/import.o \
@ -742,7 +741,7 @@ regen-importlib: Programs/_freeze_importlib
# Regenerate all generated files
regen-all: regen-opcode regen-opcode-targets regen-typeslots \
regen-token regen-symbol regen-ast regen-importlib clinic \
regen-token regen-ast regen-importlib clinic \
regen-pegen-metaparser regen-pegen
############################################################################
@ -881,15 +880,7 @@ regen-keyword:
$(srcdir)/Lib/keyword.py.new
$(UPDATE_FILE) $(srcdir)/Lib/keyword.py $(srcdir)/Lib/keyword.py.new
.PHONY: regen-symbol
regen-symbol: $(srcdir)/Include/graminit.h
# Regenerate Lib/symbol.py from Include/graminit.h
# using Tools/scripts/generate_symbol_py.py
$(PYTHON_FOR_REGEN) $(srcdir)/Tools/scripts/generate_symbol_py.py \
$(srcdir)/Include/graminit.h \
$(srcdir)/Lib/symbol.py
Python/compile.o Python/symtable.o Python/ast_unparse.o Python/ast.o Python/future.o Parser/parsetok.o: $(srcdir)/Include/graminit.h $(srcdir)/Include/Python-ast.h
Python/compile.o Python/symtable.o Python/ast_unparse.o Python/ast.o Python/future.o: $(srcdir)/Include/Python-ast.h
Python/getplatform.o: $(srcdir)/Python/getplatform.c
$(CC) -c $(PY_CORE_CFLAGS) -DPLATFORM='"$(MACHDEP)"' -o $@ $(srcdir)/Python/getplatform.c
@ -989,7 +980,6 @@ PYTHON_HEADERS= \
$(srcdir)/Include/abstract.h \
$(srcdir)/Include/asdl.h \
$(srcdir)/Include/ast.h \
$(srcdir)/Include/bitset.h \
$(srcdir)/Include/bltinmodule.h \
$(srcdir)/Include/boolobject.h \
$(srcdir)/Include/bytearrayobject.h \
@ -1027,7 +1017,6 @@ PYTHON_HEADERS= \
$(srcdir)/Include/modsupport.h \
$(srcdir)/Include/moduleobject.h \
$(srcdir)/Include/namespaceobject.h \
$(srcdir)/Include/node.h \
$(srcdir)/Include/object.h \
$(srcdir)/Include/objimpl.h \
$(srcdir)/Include/odictobject.h \

View File

@ -0,0 +1 @@
Remove the remaining files from the old parser and the :mod:`symbol` module.

View File

@ -115,7 +115,6 @@
<ClInclude Include="..\Include\abstract.h" />
<ClInclude Include="..\Include\asdl.h" />
<ClInclude Include="..\Include\ast.h" />
<ClInclude Include="..\Include\bitset.h" />
<ClInclude Include="..\Include\boolobject.h" />
<ClInclude Include="..\Include\bytearrayobject.h" />
<ClInclude Include="..\Include\bytesobject.h" />
@ -162,8 +161,6 @@
<ClInclude Include="..\Include\frameobject.h" />
<ClInclude Include="..\Include\funcobject.h" />
<ClInclude Include="..\Include\genobject.h" />
<ClInclude Include="..\Include\graminit.h" />
<ClInclude Include="..\Include\grammar.h" />
<ClInclude Include="..\Include\import.h" />
<ClInclude Include="..\Include\internal\pegen_interface.h" />
<ClInclude Include="..\Include\internal\pycore_abstract.h" />
@ -209,14 +206,12 @@
<ClInclude Include="..\Include\modsupport.h" />
<ClInclude Include="..\Include\moduleobject.h" />
<ClInclude Include="..\Include\namespaceobject.h" />
<ClInclude Include="..\Include\node.h" />
<ClInclude Include="..\Include\object.h" />
<ClInclude Include="..\Include\objimpl.h" />
<ClInclude Include="..\Include\odictobject.h" />
<ClInclude Include="..\Include\opcode.h" />
<ClInclude Include="..\Include\osdefs.h" />
<ClInclude Include="..\Include\osmodule.h" />
<ClInclude Include="..\Include\parsetok.h" />
<ClInclude Include="..\Include\patchlevel.h" />
<ClInclude Include="..\Include\picklebufobject.h" />
<ClInclude Include="..\Include\py_curses.h" />
@ -451,7 +446,6 @@
<ClCompile Include="..\Python\getopt.c" />
<ClCompile Include="..\Python\getplatform.c" />
<ClCompile Include="..\Python\getversion.c" />
<ClCompile Include="..\Python\graminit.c" />
<ClCompile Include="..\Python\hamt.c" />
<ClCompile Include="..\Python\hashtable.c" />
<ClCompile Include="..\Python\import.c" />

View File

@ -42,9 +42,6 @@
<ClInclude Include="..\Include\ast.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\bitset.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\boolobject.h">
<Filter>Include</Filter>
</ClInclude>
@ -183,12 +180,6 @@
<ClInclude Include="..\Include\genobject.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\graminit.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\grammar.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\import.h">
<Filter>Include</Filter>
</ClInclude>
@ -318,9 +309,6 @@
<ClInclude Include="..\Include\moduleobject.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\node.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\object.h">
<Filter>Include</Filter>
</ClInclude>
@ -336,9 +324,6 @@
<ClInclude Include="..\Include\osmodule.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\parsetok.h">
<Filter>Include</Filter>
</ClInclude>
<ClInclude Include="..\Include\patchlevel.h">
<Filter>Include</Filter>
</ClInclude>
@ -1010,9 +995,6 @@
<ClCompile Include="..\Python\getversion.c">
<Filter>Python</Filter>
</ClCompile>
<ClCompile Include="..\Python\graminit.c">
<Filter>Python</Filter>
</ClCompile>
<ClCompile Include="..\Python\hamt.h">
<Filter>Python</Filter>
</ClCompile>

View File

@ -133,10 +133,6 @@
</None>
<None Include="..\Grammar\Tokens">
</None>
<None Include="..\Include\graminit.h">
</None>
<None Include="..\Python\graminit.c">
</None>
<None Include="..\Include\token.h">
</None>
<None Include="..\Include\opcode.h">
@ -202,18 +198,12 @@
</Copy>
<Warning Text="Keywords updated. You will need to rebuild pythoncore to see the changes." Condition="'@(_Updated)' != ''" />
</Target>
<Target Name="_RegenSymbols" AfterTargets="_RegenKeywords">
<!-- Regenerate Lib/symbol.py from Include/graminit.h using Tools/scripts/generate_symbol_py.py-->
<Exec Command="&quot;$(PythonExe)&quot; $(PySourcePath)Tools\scripts\generate_symbol_py.py &quot;$(PySourcePath)Include\graminit.h&quot; &quot;$(PySourcePath)Lib\symbol.py&quot;" />
</Target>
<Target Name="_CleanFiles" BeforeTargets="CoreClean">
<ItemGroup>
<Clean Include="$(IntDir)keyword.py" />
<Clean Include="$(IntDir)opcode.h" />
<Clean Include="$(IntDir)Python-ast.c" />
<Clean Include="$(IntDir)Python-ast.h" />
<Clean Include="$(IntDir)graminit.h.new" />
<Clean Include="$(IntDir)graminit.c.new" />
</ItemGroup>
</Target>
</Project>

View File

@ -1,189 +0,0 @@
/* Parse tree node implementation */
#include "Python.h"
#include "node.h"
#include "errcode.h"
node *
PyNode_New(int type)
{
node *n = (node *) PyObject_MALLOC(1 * sizeof(node));
if (n == NULL)
return NULL;
n->n_type = type;
n->n_str = NULL;
n->n_lineno = 0;
n->n_end_lineno = 0;
n->n_col_offset = 0;
n->n_end_col_offset = -1;
n->n_nchildren = 0;
n->n_child = NULL;
return n;
}
/* See comments at XXXROUNDUP below. Returns -1 on overflow. */
static int
fancy_roundup(int n)
{
/* Round up to the closest power of 2 >= n. */
int result = 256;
assert(n > 128);
while (result < n) {
result <<= 1;
if (result <= 0)
return -1;
}
return result;
}
/* A gimmick to make massive numbers of reallocs quicker. The result is
* a number >= the input. In PyNode_AddChild, it's used like so, when
* we're about to add child number current_size + 1:
*
* if XXXROUNDUP(current_size) < XXXROUNDUP(current_size + 1):
* allocate space for XXXROUNDUP(current_size + 1) total children
* else:
* we already have enough space
*
* Since a node starts out empty, we must have
*
* XXXROUNDUP(0) < XXXROUNDUP(1)
*
* so that we allocate space for the first child. One-child nodes are very
* common (presumably that would change if we used a more abstract form
* of syntax tree), so to avoid wasting memory it's desirable that
* XXXROUNDUP(1) == 1. That in turn forces XXXROUNDUP(0) == 0.
*
* Else for 2 <= n <= 128, we round up to the closest multiple of 4. Why 4?
* Rounding up to a multiple of an exact power of 2 is very efficient, and
* most nodes with more than one child have <= 4 kids.
*
* Else we call fancy_roundup() to grow proportionately to n. We've got an
* extreme case then (like test_longexp.py), and on many platforms doing
* anything less than proportional growth leads to exorbitant runtime
* (e.g., MacPython), or extreme fragmentation of user address space (e.g.,
* Win98).
*
* In a run of compileall across the 2.3a0 Lib directory, Andrew MacIntyre
* reported that, with this scheme, 89% of PyObject_REALLOC calls in
* PyNode_AddChild passed 1 for the size, and 9% passed 4. So this usually
* wastes very little memory, but is very effective at sidestepping
* platform-realloc disasters on vulnerable platforms.
*
* Note that this would be straightforward if a node stored its current
* capacity. The code is tricky to avoid that.
*/
#define XXXROUNDUP(n) ((n) <= 1 ? (n) : \
(n) <= 128 ? (int)_Py_SIZE_ROUND_UP((n), 4) : \
fancy_roundup(n))
void
_PyNode_FinalizeEndPos(node *n)
{
int nch = NCH(n);
node *last;
if (nch == 0) {
return;
}
last = CHILD(n, nch - 1);
_PyNode_FinalizeEndPos(last);
n->n_end_lineno = last->n_end_lineno;
n->n_end_col_offset = last->n_end_col_offset;
}
int
PyNode_AddChild(node *n1, int type, char *str, int lineno, int col_offset,
int end_lineno, int end_col_offset)
{
const int nch = n1->n_nchildren;
int current_capacity;
int required_capacity;
node *n;
// finalize end position of previous node (if any)
if (nch > 0) {
_PyNode_FinalizeEndPos(CHILD(n1, nch - 1));
}
if (nch == INT_MAX || nch < 0)
return E_OVERFLOW;
current_capacity = XXXROUNDUP(nch);
required_capacity = XXXROUNDUP(nch + 1);
if (current_capacity < 0 || required_capacity < 0)
return E_OVERFLOW;
if (current_capacity < required_capacity) {
if ((size_t)required_capacity > SIZE_MAX / sizeof(node)) {
return E_NOMEM;
}
n = n1->n_child;
n = (node *) PyObject_REALLOC(n,
required_capacity * sizeof(node));
if (n == NULL)
return E_NOMEM;
n1->n_child = n;
}
n = &n1->n_child[n1->n_nchildren++];
n->n_type = type;
n->n_str = str;
n->n_lineno = lineno;
n->n_col_offset = col_offset;
n->n_end_lineno = end_lineno; // this and below will be updates after all children are added.
n->n_end_col_offset = end_col_offset;
n->n_nchildren = 0;
n->n_child = NULL;
return 0;
}
/* Forward */
static void freechildren(node *);
static Py_ssize_t sizeofchildren(node *n);
void
PyNode_Free(node *n)
{
if (n != NULL) {
freechildren(n);
PyObject_FREE(n);
}
}
Py_ssize_t
_PyNode_SizeOf(node *n)
{
Py_ssize_t res = 0;
if (n != NULL)
res = sizeof(node) + sizeofchildren(n);
return res;
}
static void
freechildren(node *n)
{
int i;
for (i = NCH(n); --i >= 0; )
freechildren(CHILD(n, i));
if (n->n_child != NULL)
PyObject_FREE(n->n_child);
if (STR(n) != NULL)
PyObject_FREE(STR(n));
}
static Py_ssize_t
sizeofchildren(node *n)
{
Py_ssize_t res = 0;
int i;
for (i = NCH(n); --i >= 0; )
res += sizeofchildren(CHILD(n, i));
if (n->n_child != NULL)
/* allocated size of n->n_child array */
res += XXXROUNDUP(NCH(n)) * sizeof(node);
if (STR(n) != NULL)
res += strlen(STR(n)) + 1;
return res;
}

View File

@ -5,7 +5,6 @@
*/
#include "Python.h"
#include "Python-ast.h"
#include "node.h"
#include "ast.h"
#include "token.h"
#include "pythonrun.h"

View File

@ -1,8 +1,6 @@
#include "Python.h"
#include "Python-ast.h"
#include "node.h"
#include "token.h"
#include "graminit.h"
#include "code.h"
#include "symtable.h"
#include "ast.h"

File diff suppressed because it is too large Load Diff

View File

@ -3,7 +3,6 @@
#include "Python.h"
#include "Python-ast.h"
#include "node.h"
#include "ast.h"
#include "code.h"
#include "symtable.h"

View File

@ -20,13 +20,10 @@
#include "pycore_pystate.h" // _PyInterpreterState_GET()
#include "pycore_sysmodule.h" // _PySys_Audit()
#include "node.h" // node
#include "token.h" // INDENT
#include "parsetok.h" // perrdetail
#include "errcode.h" // E_EOF
#include "code.h" // PyCodeObject
#include "symtable.h" // PySymtable_BuildObject()
#include "ast.h" // PyAST_FromNodeObject()
#include "marshal.h" // PyMarshal_ReadLongFromFile()
#include "pegen_interface.h" // PyPegen_ASTFrom*

View File

@ -1,53 +0,0 @@
#! /usr/bin/env python3
# This script generates the symbol.py source file.
import sys
import re
def main(inFileName="Include/graminit.h", outFileName="Lib/symbol.py"):
try:
fp = open(inFileName)
except OSError as err:
sys.stderr.write("I/O error: %s\n" % str(err))
sys.exit(1)
with fp:
lines = fp.read().split("\n")
prog = re.compile(
"#define[ \t][ \t]*([A-Z0-9][A-Z0-9_]*)[ \t][ \t]*([0-9][0-9]*)",
re.IGNORECASE)
tokens = {}
for line in lines:
match = prog.match(line)
if match:
name, val = match.group(1, 2)
val = int(val)
tokens[val] = name # reverse so we can sort them...
keys = sorted(tokens.keys())
# load the output skeleton from the target:
try:
fp = open(outFileName)
except OSError as err:
sys.stderr.write("I/O error: %s\n" % str(err))
sys.exit(2)
with fp:
format = fp.read().split("\n")
try:
start = format.index("#--start constants--") + 1
end = format.index("#--end constants--")
except ValueError:
sys.stderr.write("target does not contain format markers")
sys.exit(3)
lines = []
for val in keys:
lines.append("%s = %d" % (tokens[val], val))
format[start:end] = lines
try:
fp = open(outFileName, 'w')
except OSError as err:
sys.stderr.write("I/O error: %s\n" % str(err))
sys.exit(4)
with fp:
fp.write("\n".join(format))
if __name__ == '__main__':
main(*sys.argv[1:])

2
configure vendored
View File

@ -2739,7 +2739,7 @@ if test "$srcdir" != . -a "$srcdir" != "$(pwd)"; then
# If we're building out-of-tree, we need to make sure the following
# resources get picked up before their $srcdir counterparts.
# Objects/ -> typeslots.inc
# Include/ -> Python-ast.h, graminit.h
# Include/ -> Python-ast.h
# Python/ -> importlib.h
# (A side effect of this is that these resources will automatically be
# regenerated when building out-of-tree, regardless of whether or not

View File

@ -16,7 +16,7 @@ if test "$srcdir" != . -a "$srcdir" != "$(pwd)"; then
# If we're building out-of-tree, we need to make sure the following
# resources get picked up before their $srcdir counterparts.
# Objects/ -> typeslots.inc
# Include/ -> Python-ast.h, graminit.h
# Include/ -> Python-ast.h
# Python/ -> importlib.h
# (A side effect of this is that these resources will automatically be
# regenerated when building out-of-tree, regardless of whether or not