python,scons: Move grammar.py and code_formatter.py into build_tools.

These are only used in a build, and so don't need to be built into gem5.
grammar.py is used by slicc and the fast model project file parser, and
code_formatter.py is only used by SConscripts.

Change-Id: Id43e62459d69f07fdb2ed125548a83e38bbb7590
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49396
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Hoa Nguyen <hoanguyen@ucdavis.edu>
Maintainer: Gabe Black <gabe.black@gmail.com>
This commit is contained in:
Gabe Black
2021-08-14 06:18:43 -07:00
parent 6cc3b4f6c1
commit cc75a47b84
14 changed files with 12 additions and 13 deletions

View File

@@ -0,0 +1,346 @@
# Copyright (c) 2006-2009 Nathan Binkert <nate@binkert.org>
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met: redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer;
# redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution;
# neither the name of the copyright holders nor the names of its
# contributors may be used to endorse or promote products derived from
# this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
try:
import builtins
except ImportError:
# Python 2 fallback
import __builtin__ as builtins
import inspect
import os
import re
class lookup(object):
def __init__(self, formatter, frame, *args, **kwargs):
self.frame = frame
self.formatter = formatter
self.dict = self.formatter._dict
self.args = args
self.kwargs = kwargs
self.locals = {}
def __setitem__(self, item, val):
self.locals[item] = val
def __getitem__(self, item):
if item in self.locals:
return self.locals[item]
if item in self.kwargs:
return self.kwargs[item]
if item == '__file__':
return self.frame.f_code.co_filename
if item == '__line__':
return self.frame.f_lineno
if self.formatter.locals and item in self.frame.f_locals:
return self.frame.f_locals[item]
if item in self.dict:
return self.dict[item]
if self.formatter.globals and item in self.frame.f_globals:
return self.frame.f_globals[item]
if item in builtins.__dict__:
return builtins.__dict__[item]
try:
item = int(item)
return self.args[item]
except ValueError:
pass
raise IndexError("Could not find '%s'" % item)
class code_formatter_meta(type):
pattern = r"""
(?:
%(delim)s(?P<escaped>%(delim)s) | # escaped delimiter
^(?P<indent>[ ]*)%(delim)s(?P<lone>%(ident)s)$ | # lone identifier
%(delim)s(?P<ident>%(ident)s) | # identifier
%(delim)s%(lb)s(?P<b_ident>%(ident)s)%(rb)s | # braced identifier
%(delim)s(?P<pos>%(pos)s) | # positional parameter
%(delim)s%(lb)s(?P<b_pos>%(pos)s)%(rb)s | # braced positional
%(delim)s%(ldb)s(?P<eval>.*?)%(rdb)s | # double braced expression
%(delim)s(?P<invalid>) # ill-formed delimiter exprs
)
"""
def __init__(cls, name, bases, dct):
super(code_formatter_meta, cls).__init__(name, bases, dct)
if 'pattern' in dct:
pat = cls.pattern
else:
# tuple expansion to ensure strings are proper length
lb,rb = cls.braced
lb1,lb2,rb2,rb1 = cls.double_braced
pat = code_formatter_meta.pattern % {
'delim' : re.escape(cls.delim),
'ident' : cls.ident,
'pos' : cls.pos,
'lb' : re.escape(lb),
'rb' : re.escape(rb),
'ldb' : re.escape(lb1+lb2),
'rdb' : re.escape(rb2+rb1),
}
cls.pattern = re.compile(pat, re.VERBOSE | re.DOTALL | re.MULTILINE)
class code_formatter(object, metaclass=code_formatter_meta):
delim = r'$'
ident = r'[_A-z]\w*'
pos = r'[0-9]+'
braced = r'{}'
double_braced = r'{{}}'
globals = True
locals = True
fix_newlines = True
def __init__(self, *args, **kwargs):
self._data = []
self._dict = {}
self._indent_level = 0
self._indent_spaces = 4
self.globals = kwargs.pop('globals', type(self).globals)
self.locals = kwargs.pop('locals', type(self).locals)
self._fix_newlines = \
kwargs.pop('fix_newlines', type(self).fix_newlines)
if args:
self.__call__(args)
def indent(self, count=1):
self._indent_level += self._indent_spaces * count
def dedent(self, count=1):
assert self._indent_level >= (self._indent_spaces * count)
self._indent_level -= self._indent_spaces * count
def fix(self, status):
previous = self._fix_newlines
self._fix_newlines = status
return previous
def nofix(self):
previous = self._fix_newlines
self._fix_newlines = False
return previous
def clear():
self._data = []
def write(self, *args):
f = open(os.path.join(*args), "w")
name, extension = os.path.splitext(f.name)
# Add a comment to inform which file generated the generated file
# to make it easier to backtrack and modify generated code
frame = inspect.currentframe().f_back
if re.match('\.(cc|hh|c|h)', extension) is not None:
f.write(f'''/**
* DO NOT EDIT THIS FILE!
* File automatically generated by
* {frame.f_code.co_filename}:{frame.f_lineno}
*/
''')
elif re.match('\.py', extension) is not None:
f.write(f'''#
# DO NOT EDIT THIS FILE!
# File automatically generated by
# {frame.f_code.co_filename}:{frame.f_lineno}
#
''')
elif re.match('\.html', extension) is not None:
f.write(f'''<!--
DO NOT EDIT THIS FILE!
File automatically generated by
{frame.f_code.co_filename}:{frame.f_lineno}
-->
''')
for data in self._data:
f.write(data)
f.close()
def __str__(self):
data = ''.join(self._data)
self._data = [ data ]
return data
def __getitem__(self, item):
return self._dict[item]
def __setitem__(self, item, value):
self._dict[item] = value
def __delitem__(self, item):
del self._dict[item]
def __contains__(self, item):
return item in self._dict
def __iadd__(self, data):
self.append(data)
def append(self, data):
if isinstance(data, code_formatter):
self._data.extend(data._data)
else:
self._append(str(data))
def _append(self, data):
if not self._fix_newlines:
self._data.append(data)
return
initial_newline = not self._data or self._data[-1] == '\n'
for line in data.splitlines():
if line:
if self._indent_level:
self._data.append(' ' * self._indent_level)
self._data.append(line)
if line or not initial_newline:
self._data.append('\n')
initial_newline = False
def __call__(self, *args, **kwargs):
if not args:
self._data.append('\n')
return
format = args[0]
args = args[1:]
frame = inspect.currentframe().f_back
l = lookup(self, frame, *args, **kwargs)
def convert(match):
ident = match.group('lone')
# check for a lone identifier
if ident:
indent = match.group('indent') # must be spaces
lone = '%s' % (l[ident], )
def indent_lines(gen):
for line in gen:
yield indent
yield line
return ''.join(indent_lines(lone.splitlines(True)))
# check for an identifier, braced or not
ident = match.group('ident') or match.group('b_ident')
if ident is not None:
return '%s' % (l[ident], )
# check for a positional parameter, braced or not
pos = match.group('pos') or match.group('b_pos')
if pos is not None:
pos = int(pos)
if pos > len(args):
raise ValueError \
('Positional parameter #%d not found in pattern' % pos,
code_formatter.pattern)
return '%s' % (args[int(pos)], )
# check for a double braced expression
eval_expr = match.group('eval')
if eval_expr is not None:
result = eval(eval_expr, {}, l)
return '%s' % (result, )
# check for an escaped delimiter
if match.group('escaped') is not None:
return '$'
# At this point, we have to match invalid
if match.group('invalid') is None:
# didn't match invalid!
raise ValueError('Unrecognized named group in pattern',
code_formatter.pattern)
i = match.start('invalid')
if i == 0:
colno = 1
lineno = 1
else:
lines = format[:i].splitlines(True)
colno = i - sum(len(z) for z in lines)
lineno = len(lines)
raise ValueError('Invalid format string: line %d, col %d' %
(lineno, colno))
d = code_formatter.pattern.sub(convert, format)
self._append(d)
__all__ = [ "code_formatter" ]
if __name__ == '__main__':
from .code_formatter import code_formatter
f = code_formatter()
class Foo(dict):
def __init__(self, **kwargs):
self.update(kwargs)
def __getattr__(self, attr):
return self[attr]
x = "this is a test"
l = [ [Foo(x=[Foo(y=9)])] ]
y = code_formatter()
y('''
{
this_is_a_test();
}
''')
f(' $y')
f('''$__file__:$__line__
{''')
f("${{', '.join(str(x) for x in range(4))}}")
f('${x}')
f('$x')
f.indent()
for i in range(5):
f('$x')
f('$i')
f('$0', "zero")
f('$1 $0', "zero", "one")
f('${0}', "he went")
f('${0}asdf', "he went")
f.dedent()
f('''
${{l[0][0]["x"][0].y}}
}
''', 1, 9)
print(f, end=' ')

137
build_tools/grammar.py Normal file
View File

@@ -0,0 +1,137 @@
# Copyright (c) 2006-2011 Nathan Binkert <nate@binkert.org>
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met: redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer;
# redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution;
# neither the name of the copyright holders nor the names of its
# contributors may be used to endorse or promote products derived from
# this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
import os
import ply.lex
import ply.yacc
class ParseError(Exception):
def __init__(self, message, token=None):
Exception.__init__(self, message)
self.token = token
class Grammar(object):
def setupLexerFactory(self, **kwargs):
if 'module' in kwargs:
raise AttributeError("module is an illegal attribute")
self.lex_kwargs = kwargs
def setupParserFactory(self, **kwargs):
if 'module' in kwargs:
raise AttributeError("module is an illegal attribute")
if 'output' in kwargs:
dir,tab = os.path.split(output)
if not tab.endswith('.py'):
raise AttributeError('The output file must end with .py')
kwargs['outputdir'] = dir
kwargs['tabmodule'] = tab[:-3]
self.yacc_kwargs = kwargs
def __getattr__(self, attr):
if attr == 'lexers':
self.lexers = []
return self.lexers
if attr == 'lex_kwargs':
self.setupLexerFactory()
return self.lex_kwargs
if attr == 'yacc_kwargs':
self.setupParserFactory()
return self.yacc_kwargs
if attr == 'lex':
self.lex = ply.lex.lex(module=self, **self.lex_kwargs)
return self.lex
if attr == 'yacc':
self.yacc = ply.yacc.yacc(module=self, **self.yacc_kwargs)
return self.yacc
if attr == 'current_lexer':
if not self.lexers:
return None
return self.lexers[-1][0]
if attr == 'current_source':
if not self.lexers:
return '<none>'
return self.lexers[-1][1]
if attr == 'current_line':
if not self.lexers:
return -1
return self.current_lexer.lineno
raise AttributeError(
"'%s' object has no attribute '%s'" % (type(self), attr))
def parse_string(self, data, source='<string>', debug=None, tracking=0):
if not isinstance(data, str):
raise AttributeError(
"argument must be a string, was '%s'" % type(f))
lexer = self.lex.clone()
lexer.input(data)
self.lexers.append((lexer, source))
lrtab = ply.yacc.LRTable()
lrtab.lr_productions = self.yacc.productions
lrtab.lr_action = self.yacc.action
lrtab.lr_goto = self.yacc.goto
parser = ply.yacc.LRParser(lrtab, self.yacc.errorfunc)
result = parser.parse(lexer=lexer, debug=debug, tracking=tracking)
self.lexers.pop()
return result
def parse_file(self, f, **kwargs):
if isinstance(f, str):
source = f
f = open(f, 'r')
elif isinstance(f, file):
source = f.name
else:
raise AttributeError(
"argument must be either a string or file, was '%s'" % type(f))
return self.parse_string(f.read(), source, **kwargs)
def p_error(self, t):
if t:
msg = "Syntax error at %s:%d:%d\n>>%s<<" % \
(self.current_source, t.lineno, t.lexpos + 1, t.value)
else:
msg = "Syntax error at end of %s" % (self.current_source, )
raise ParseError(msg, t)
def t_error(self, t):
msg = "Illegal character %s @ %d:%d" % \
(repr(t.value[0]), t.lineno, t.lexpos)
raise ParseError(msg, t)