diff options
Diffstat (limited to 'src/mem/slicc/parser/parser.py')
-rw-r--r-- | src/mem/slicc/parser/parser.py | 563 |
1 files changed, 0 insertions, 563 deletions
diff --git a/src/mem/slicc/parser/parser.py b/src/mem/slicc/parser/parser.py deleted file mode 100644 index 7fecfd273..000000000 --- a/src/mem/slicc/parser/parser.py +++ /dev/null @@ -1,563 +0,0 @@ -# Copyright (c) 2009 The Hewlett-Packard Development Company -# 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. -# -# Authors: Nathan Binkert - -from ply import lex, yacc -import re - -t_ignore = '\t ' - -# C or C++ comment (ignore) -def t_c_comment(t): - r'/\*(.|\n)*?\*/' - t.lexer.lineno += t.value.count('\n') - -def t_cpp_comment(t): - r'//.*' - pass - -# Define a rule so we can track line numbers -def t_newline(t): - r'\n+' - t.lexer.lineno += len(t.value) - -reserved = { - 'global' : 'GLOBAL', - 'machine' : 'MACHINE', - 'in_port' : 'IN_PORT', - 'out_port' : 'OUT_PORT', - 'action' : 'ACTION', - 'transition' : 'TRANS', - 'structure' : 'STRUCT', - 'external_type' : 'EXTERN_TYPE', - 'enumeration' : 'ENUM', - 'peek' : 'PEEK', - 'enqueue' : 'ENQUEUE', - 'copy_head' : 'COPY_HEAD', - 'check_allocate' : 'CHECK_ALLOCATE', - 'check_stop_slots' : 'CHECK_STOP_SLOTS', - 'if' : 'IF', - 'else' : 'ELSE', - 'return' : 'RETURN', - 'THIS' : 'THIS', - 'CHIP' : 'CHIP', - 'void' : 'VOID', - 'new' : 'NEW', -} - -literals = ':[]{}(),=' - -tokens = [ 'EQ', 'NE', 'LT', 'GT', 'LE', 'GE', - 'LEFTSHIFT', 'RIGHTSHIFT', - 'NOT', 'AND', 'OR', - 'PLUS', 'DASH', 'STAR', 'SLASH', - 'DOUBLE_COLON', 'SEMICOLON', - 'ASSIGN', 'DOT', - 'IDENT', 'LIT_BOOL', 'FLOATNUMBER', 'NUMBER', 'STRING' ] -tokens += reserved.values() - -t_EQ = r'==' -t_NE = r'!=' -t_LT = r'<' -t_GT = r'>' -t_LE = r'<=' -t_GE = r'>=' -t_LEFTSHIFT = r'<<' -t_RIGHTSHIFT = r'>>' -t_NOT = r'!' -t_AND = r'&&' -t_OR = r'\|\|' -t_PLUS = r'\+' -t_DASH = r'-' -t_STAR = r'\*' -t_SLASH = r'/' -t_DOUBLE_COLON = r'::' -t_SEMICOLON = r';' -t_ASSIGN = r':=' -t_DOT = r'\.' - -class TokenError(Exception): - def __init__(self, msg, t): - super(TokenError, self).__init__(msg) - self.token = t - -class ParseError(Exception): - def __init__(self, msg, t): - super(ParseError, self).__init__(msg) - self.token = t - -def t_error(t): - raise TokenError("Illegal character", t) - -def t_IDENT(t): - r'[a-zA-Z_][a-zA-Z_0-9]*' - if t.value == 'true': - t.type = 'LIT_BOOL' - t.value = True - return t - - if t.value == 'false': - t.type = 'LIT_BOOL' - t.value = False - return t - - if t.value.startswith('LATENCY_'): - t.type = 'LATENCY' - return t - - t.type = reserved.get(t.value, 'IDENT') # Check for reserved words - return t - -def t_FLOATNUMBER(t): - '[0-9]+[.][0-9]+' - try: - t.value = float(t.value) - except ValueError: - raise TokenError("Illegal float", t) - return t - -def t_NUMBER(t): - r'[0-9]+' - try: - t.value = int(t.value) - except ValueError: - raise TokenError("Illegal number", t) - return t - -def t_STRING1(t): - r'\"[^"\n]*\"' - t.type = 'STRING' - return t - -def t_STRING2(t): - r"\'[^'\n]*\'" - t.type = 'STRING' - return t - - -def p_file(p): - "file : decl_l" - p[0] = [ x for x in p[1] if x is not None ] - -def p_error(t): - raise ParseError("Syntax error", t) - -def p_empty(p): - "empty :" - pass - -def p_decl_l(p): - "decl_l : decls" - p[0] = p[1] - -def p_decls(p): - """decls : decl decls - | empty""" - if len(p) == 3: - p[0] = [ p[1] ] + p[2] - elif len(p) == 2: - p[0] = [] - -def p_decl(p): - """decl : d_machine - | d_action - | d_in_port - | d_out_port - | t_trans - | d_extern - | d_global - | d_struct - | d_enum - | d_object - | d_func_decl - | d_func_def""" - p[0] = p[1] - -def p_d_machine(p): - """d_machine : MACHINE '(' ident pair_l ')' ':' param_l '{' decl_l '}'""" - - if len(p) == 9: - decl_l = p[7] - elif len(p) == 11: - decl_l = p[9] - decls = [ x for x in decl_l if x is not None ] - p[0] = Machine(p[3], decls) - -def p_d_action(p): - "d_action : ACTION '(' ident pair_l ')' statement_l" - p[0] = Action(p[3]) - -def p_d_in_port(p): - "d_in_port : IN_PORT '(' ident ',' type ',' var pair_l ')' statement_l" - p[0] = InPort(p[3]) - -def p_d_out_port(p): - "d_out_port : OUT_PORT '(' ident ',' type ',' var pair_l ')' SEMICOLON" - p[0] = OutPort(p[3]) - -def p_t_trans(p): - """t_trans : TRANS '(' ident_l ',' ident_l ',' ident pair_l ')' ident_l - | TRANS '(' ident_l ',' ident_l pair_l ')' ident_l""" - p[0] = Transition("transition") - -def p_d_extern(p): - """d_extern : EXTERN_TYPE '(' type pair_l ')' SEMICOLON - | EXTERN_TYPE '(' type pair_l ')' '{' type_methods '}'""" - p[0] = Extern(p[3]) - -def p_d_global(p): - "d_global : GLOBAL '(' type pair_l ')' '{' type_members '}'" - p[0] = Global(p[3]) - -def p_d_struct(p): - "d_struct : STRUCT '(' type pair_l ')' '{' type_members '}'" - p[0] = Struct(p[3]) - -def p_d_enum(p): - "d_enum : ENUM '(' type pair_l ')' '{' type_enums '}'" - p[0] = Enum(p[3]) - -def p_d_object(p): - "d_object : type ident pair_l SEMICOLON" - p[0] = Object(p[2]) - -def p_d_func_decl(p): - """d_func_decl : void ident '(' param_l ')' pair_l SEMICOLON - | type ident '(' param_l ')' pair_l SEMICOLON""" - pass - -def p_d_func_def(p): - """d_func_def : void ident '(' param_l ')' pair_l statement_l - | type ident '(' param_l ')' pair_l statement_l""" - p[0] = Function(p[2]) - -# Type fields -def p_type_members(p): - """type_members : type_member type_members - | empty""" - pass - -def p_type_member(p): - """type_member : type ident pair_l SEMICOLON - | type ident ASSIGN expr SEMICOLON""" - pass - -# Methods -def p_type_methods(p): - """type_methods : type_method type_methods - | empty""" - pass - -def p_type_method(p): - "type_method : type_or_void ident '(' type_l ')' pair_l SEMICOLON" - pass - -# Enum fields -def p_type_enums(p): - """type_enums : type_enum type_enums - | empty""" - pass - -def p_type_enum(p): - "type_enum : ident pair_l SEMICOLON" - pass - -# Type -def p_type_l(p): - """type_l : types - | empty""" - pass - -def p_types(p): - """types : type ',' types - | type""" - pass - -def p_type(p): - "type : ident" - p[0] = p[1] - -def p_void(p): - "void : VOID" - p[0] = None - -def p_type_or_void(p): - """type_or_void : type - | void""" - p[0] = p[1] - -# Formal Param -def p_param_l(p): - """param_l : params - | empty""" - pass - -def p_params(p): - """params : param ',' params - | param""" - pass - -def p_param(p): - "param : type ident" - pass - -# Idents and lists -def p_ident(p): - "ident : IDENT" - p[0] = p[1] - -def p_ident_l(p): - """ident_l : '{' idents '}' - | ident""" - p[0] = p[1] - -def p_idents(p): - """idents : ident SEMICOLON idents - | ident ',' idents - | ident idents - | empty""" - pass - -# Pair and pair lists -def p_pair_l(p): - """pair_l : ',' pairs - | empty""" - if len(p) == 3: - p[0] = p[2] - elif len(p) == 2: - p[0] = None - -def p_pairs(p): - """pairs : pair ',' pairs - | pair""" - if len(p) == 4: - p[3].append(p[1]) - p[0] = p[3] - elif len(p) == 2: - p[0] = [ p[1] ] - -def p_pair(p): - """pair : ident '=' STRING - | ident '=' ident - | STRING""" - if len(p) == 4: - p[0] = p[1], p[3] - elif len(p) == 2: - p[0] = "short", p[1] - -# Below are the rules for action descriptions -def p_statement_l(p): - "statement_l : '{' statements '}'" - pass - -def p_statements(p): - """statements : statement statements - | empty""" - pass - -def p_expr_l(p): - """expr_l : expr ',' expr_l - | expr - | empty""" - pass - -def p_statement(p): - """statement : expr SEMICOLON - | expr ASSIGN expr SEMICOLON - | ENQUEUE '(' var ',' type pair_l ')' statement_l - | PEEK '(' var ',' type ')' statement_l - | COPY_HEAD '(' var ',' var pair_l ')' SEMICOLON - | CHECK_ALLOCATE '(' var ')' SEMICOLON - | CHECK_STOP_SLOTS '(' var ',' STRING ',' STRING ')' SEMICOLON - | if_statement - | RETURN expr SEMICOLON""" - pass - -def p_if_statement(p): - """if_statement : IF '(' expr ')' statement_l ELSE statement_l - | IF '(' expr ')' statement_l - | IF '(' expr ')' statement_l ELSE if_statement""" - pass - -def p_expr(p): - """expr : var - | literal - | enumeration - | ident '(' expr_l ')' - | NEW type - | THIS DOT var '[' expr ']' DOT var DOT ident '(' expr_l ')' - | THIS DOT var '[' expr ']' DOT var DOT ident - | CHIP '[' expr ']' DOT var '[' expr ']' DOT var DOT ident '(' expr_l ')' - | CHIP '[' expr ']' DOT var '[' expr ']' DOT var DOT ident - | expr DOT ident - | expr DOT ident '(' expr_l ')' - | type DOUBLE_COLON ident '(' expr_l ')' - | expr '[' expr_l ']' - | expr STAR expr - | expr SLASH expr - | expr PLUS expr - | expr DASH expr - | expr LT expr - | expr GT expr - | expr LE expr - | expr GE expr - | expr EQ expr - | expr NE expr - | expr AND expr - | expr OR expr - | NOT expr - | expr RIGHTSHIFT expr - | expr LEFTSHIFT expr - | '(' expr ')'""" - pass - -def p_literal(p): - """literal : STRING - | NUMBER - | FLOATNUMBER - | LIT_BOOL""" - pass - -def p_enumeration(p): - "enumeration : ident ':' ident" - pass - -def p_var(p): - "var : ident" - pass - -lex.lex() -yacc.yacc(write_tables=0) - -slicc_generated_cc = set([ - 'ControllerFactory.cc', - 'MachineType.cc']) - -slicc_generated_hh = set([ - 'ControllerFactory.hh', - 'MachineType.hh', - 'Types.hh', - 'protocol_name.hh' ]) - -class Machine(object): - def __init__(self, name, decls): - self.name = name - self.decls = decls - - def add(self, hh, cc): - hh.add('%s_Controller.hh' % self.name) - hh.add('%s_Profiler.hh' % self.name) - - cc.add('%s_Controller.cc' % self.name) - cc.add('%s_Profiler.cc' % self.name) - cc.add('%s_Transitions.cc' % self.name) - cc.add('%s_Wakeup.cc' % self.name) - - for decl in self.decls: - decl.add(hh, cc, self.name) - -class Declaration(object): - hh = False - cc = False - def __init__(self, name): - self.name = name - - def add(self, hh, cc, name=None): - #print '>>>', type(self).__name__, self.name - if name: - name += '_' - else: - name = "" - if self.hh: - hh.add('%s%s.hh' % (name, self.name)) - if self.cc: - cc.add('%s%s.cc' % (name, self.name)) - -class Action(Declaration): pass -class InPort(Declaration): pass -class OutPort(Declaration): pass -class Transition(Declaration): pass -class Extern(Declaration): pass -class Global(Declaration): - hh = True - cc = True -class Struct(Declaration): - hh = True - cc = True -class Enum(Declaration): - hh = True - cc = True -class Object(Declaration): pass -class Function(Declaration): - cc = True - -def read_slicc(sources): - if not isinstance(sources, (list,tuple)): - sources = [ sources ] - - sm_files = [] - for source in sources: - for sm_file in file(source, "r"): - sm_file = sm_file.strip() - if not sm_file: - continue - if sm_file.startswith("#"): - continue - sm_files.append(sm_file) - - return sm_files - -def scan(filenames): - hh = slicc_generated_hh.copy() - cc = slicc_generated_cc.copy() - - for filename in filenames: - lex.lexer.lineno = 1 - try: - print "parsing ",filename - results = yacc.parse(file(filename, 'r').read()) - except (TokenError, ParseError), e: - sys.exit("%s: %s:%d" % (e, filename, e.token.lineno)) - - for result in results: - result.add(hh, cc) - - return list(hh), list(cc) - -if __name__ == '__main__': - import sys - - hh, cc = scan(read_slicc(sys.argv[1:])) - hh.sort() - cc.sort() - print 'Headers:' - for i in hh: - print ' %s' % i - - print 'Sources:' - for i in cc: - print ' %s' % i |