Enrico Granata | 4c43add | 2017-03-15 18:01:34 -0700 | [diff] [blame^] | 1 | #!/usr/bin/env python3 |
| 2 | # |
| 3 | # Copyright (C) 2017 The Android Open Source Project |
| 4 | # |
| 5 | # Licensed under the Apache License, Version 2.0 (the "License"); |
| 6 | # you may not use this file except in compliance with the License. |
| 7 | # You may obtain a copy of the License at |
| 8 | # |
| 9 | # http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | # |
| 11 | # Unless required by applicable law or agreed to in writing, software |
| 12 | # distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | # See the License for the specific language governing permissions and |
| 15 | # limitations under the License. |
| 16 | # |
| 17 | |
| 18 | # A parser for enum types defined in HIDL. |
| 19 | # This script can parse HIDL files and generate a parse tree. |
| 20 | # To use, import and call parse("path/to/file.hal") |
| 21 | # It will return a Python dictionary with two keys: |
| 22 | # - header: an instance of Header |
| 23 | # - enums: a dictionary of EnumDecl objects by name |
| 24 | # This script cannot parse structs for now, but that would be easy to add. |
| 25 | |
| 26 | # It requires 'ply' (Python Lex/Yacc). |
| 27 | |
| 28 | import ply |
| 29 | |
| 30 | tokens = ('package', 'import', 'enum', |
| 31 | 'COLON', 'IDENTIFIER', 'COMMENT', 'NUMBER', 'HEX', 'OR', 'EQUALS', |
| 32 | 'LPAREN', 'RPAREN', 'LBRACE', 'RBRACE', 'DOT', 'SEMICOLON', 'VERSION', |
| 33 | 'COMMA', 'SHIFT') |
| 34 | |
| 35 | t_COLON = r':' |
| 36 | t_NUMBER = r'[0-9]+' |
| 37 | t_HEX = r'0x[0-9A-Fa-f]+' |
| 38 | t_OR = r'\|' |
| 39 | t_EQUALS = r'=' |
| 40 | t_LPAREN = r'\(' |
| 41 | t_RPAREN = r'\)' |
| 42 | t_SHIFT = r'<<' |
| 43 | |
| 44 | def t_COMMENT(t): |
| 45 | r'(/\*(.|\n)*?\*/)|(//.*)' |
| 46 | pass |
| 47 | |
| 48 | t_LBRACE = r'{' |
| 49 | t_RBRACE = r'}' |
| 50 | t_DOT = r'\.' |
| 51 | t_SEMICOLON = r';' |
| 52 | t_VERSION = r'@[0-9].[0-9]' |
| 53 | t_COMMA = r',' |
| 54 | t_ignore = ' \n\t' |
| 55 | |
| 56 | def t_IDENTIFIER(t): |
| 57 | r'[a-zA-Z_][a-zA-Z_0-9]*' |
| 58 | if t.value == 'package': |
| 59 | t.type = 'package' |
| 60 | elif t.value == 'import': |
| 61 | t.type = 'import' |
| 62 | elif t.value == 'enum': |
| 63 | t.type = 'enum' |
| 64 | return t |
| 65 | |
| 66 | def t_error(t): |
| 67 | t.type = t.value[0] |
| 68 | t.value = t.value[0] |
| 69 | t.lexer.skip(1) |
| 70 | return t |
| 71 | |
| 72 | import ply.lex as lex |
| 73 | lexer = lex.lex() |
| 74 | |
| 75 | class EnumHeader(object): |
| 76 | def __init__(self, name, base): |
| 77 | self.name = name |
| 78 | self.base = base |
| 79 | |
| 80 | def __str__(self): |
| 81 | return '%s%s' % (self.name, ' %s' % self.base if self.base else '') |
| 82 | |
| 83 | class EnumDecl(object): |
| 84 | def __init__(self, header, cases): |
| 85 | self.header = header |
| 86 | self.cases = cases |
| 87 | |
| 88 | def __str__(self): |
| 89 | return '%s {\n%s\n}' % (self.header, |
| 90 | '\n'.join(str(x) for x in self.cases)) |
| 91 | |
| 92 | def __repr__(self): |
| 93 | return self.__str__() |
| 94 | |
| 95 | class EnumCase(object): |
| 96 | def __init__(self, name, value): |
| 97 | self.name = name |
| 98 | self.value = value |
| 99 | |
| 100 | def __str__(self): |
| 101 | return '%s = %s' % (self.name, self.value) |
| 102 | |
| 103 | class PackageID(object): |
| 104 | def __init__(self, name, version): |
| 105 | self.name = name |
| 106 | self.version = version |
| 107 | |
| 108 | def __str__(self): |
| 109 | return '%s%s' % (self.name, self.version) |
| 110 | |
| 111 | class Package(object): |
| 112 | def __init__(self, package): |
| 113 | self.package = package |
| 114 | |
| 115 | def __str__(self): |
| 116 | return 'package %s' % self.package |
| 117 | |
| 118 | class Import(object): |
| 119 | def __init__(self, package): |
| 120 | self.package = package |
| 121 | |
| 122 | def __str__(self): |
| 123 | return 'import %s' % self.package |
| 124 | |
| 125 | class Header(object): |
| 126 | def __init__(self, package, imports): |
| 127 | self.package = package |
| 128 | self.imports = imports |
| 129 | |
| 130 | def __str__(self): |
| 131 | return str(self.package) + "\n" + \ |
| 132 | '\n'.join(str(x) for x in self.imports) |
| 133 | |
| 134 | # Error rule for syntax errors |
| 135 | def p_error(p): |
| 136 | print("Syntax error in input: %s" % p) |
| 137 | |
| 138 | def p_document(t): |
| 139 | 'document : header enum_decls' |
| 140 | enums = {} |
| 141 | for enum in t[2]: |
| 142 | enums[enum.header.name] = enum |
| 143 | t[0] = {'header' : t[1], 'enums' : enums} |
| 144 | |
| 145 | def p_enum_decls_1(t): |
| 146 | 'enum_decls : enum_decl' |
| 147 | t[0] = [t[1]] |
| 148 | def p_enum_decls_2(t): |
| 149 | 'enum_decls : enum_decls enum_decl' |
| 150 | t[0] = t[1] + [t[2]] |
| 151 | |
| 152 | def p_enum_cases_1(t): |
| 153 | 'enum_cases : enum_case' |
| 154 | t[0] = [t[1]] |
| 155 | def p_enum_cases_2(t): |
| 156 | 'enum_cases : enum_cases COMMA enum_case' |
| 157 | t[0] = t[1] + [t[3]] |
| 158 | |
| 159 | def p_enum_base_1(t): |
| 160 | 'enum_base : VERSION COLON COLON IDENTIFIER' |
| 161 | t[0] = '%s::%s' % (t[1], t[4]) |
| 162 | def p_enum_base_2(t): |
| 163 | 'enum_base : IDENTIFIER' |
| 164 | t[0] = t[1] |
| 165 | |
| 166 | def p_enum_header_1(t): |
| 167 | 'enum_header : enum IDENTIFIER' |
| 168 | t[0] = EnumHeader(t[2], None) |
| 169 | def p_enum_header_2(t): |
| 170 | 'enum_header : enum IDENTIFIER COLON enum_base' |
| 171 | t[0] = EnumHeader(t[2], t[4]) |
| 172 | |
| 173 | def p_enum_decl_1(t): |
| 174 | 'enum_decl : enum_header LBRACE enum_cases RBRACE SEMICOLON' |
| 175 | t[0] = EnumDecl(t[1], t[3]) |
| 176 | def p_enum_decl_2(t): |
| 177 | 'enum_decl : enum_header LBRACE enum_cases COMMA RBRACE SEMICOLON' |
| 178 | t[0] = EnumDecl(t[1], t[3]) |
| 179 | |
| 180 | def p_enum_value_1(t): |
| 181 | '''enum_value : NUMBER |
| 182 | | HEX |
| 183 | | IDENTIFIER''' |
| 184 | t[0] = t[1] |
| 185 | def p_enum_value_2(t): |
| 186 | 'enum_value : enum_value SHIFT NUMBER' |
| 187 | t[0] = '%s << %s' % (t[1], t[3]) |
| 188 | def p_enum_value_3(t): |
| 189 | 'enum_value : enum_value OR enum_value' |
| 190 | t[0] = "%s | %s" % (t[1], t[3]) |
| 191 | def p_enum_value_4(t): |
| 192 | 'enum_value : LPAREN enum_value RPAREN' |
| 193 | t[0] = t[2] |
| 194 | def p_enum_value_5(t): |
| 195 | 'enum_value : IDENTIFIER COLON IDENTIFIER' |
| 196 | t[0] = '%s:%s' % (t[1],t[3]) |
| 197 | |
| 198 | def p_enum_case(t): |
| 199 | 'enum_case : IDENTIFIER EQUALS enum_value' |
| 200 | t[0] = EnumCase(t[1], t[3]) |
| 201 | |
| 202 | def p_header_1(t): |
| 203 | 'header : package_decl' |
| 204 | t[0] = Header(t[1], []) |
| 205 | |
| 206 | def p_header_2(t): |
| 207 | 'header : package_decl import_decls' |
| 208 | t[0] = Header(t[1], t[2]) |
| 209 | |
| 210 | def p_import_decls_1(t): |
| 211 | 'import_decls : import_decl' |
| 212 | t[0] = [t[1]] |
| 213 | |
| 214 | def p_import_decls_2(t): |
| 215 | 'import_decls : import_decls import_decl' |
| 216 | t[0] = t[1] + [t[2]] |
| 217 | |
| 218 | def p_package_decl(t): |
| 219 | 'package_decl : package package_ID SEMICOLON' |
| 220 | t[0] = Package(t[2]) |
| 221 | |
| 222 | def p_import_decl(t): |
| 223 | 'import_decl : import package_ID SEMICOLON' |
| 224 | t[0] = Import(t[2]) |
| 225 | |
| 226 | def p_package_ID(t): |
| 227 | 'package_ID : dotted_identifier VERSION' |
| 228 | t[0] = PackageID(t[1], t[2]) |
| 229 | |
| 230 | def p_dotted_identifier_1(t): |
| 231 | 'dotted_identifier : IDENTIFIER' |
| 232 | t[0] = t[1] |
| 233 | def p_dotted_identifier_2(t): |
| 234 | 'dotted_identifier : dotted_identifier DOT IDENTIFIER' |
| 235 | t[0] = t[1] + '.' + t[3] |
| 236 | |
| 237 | class SilentLogger(object): |
| 238 | def warning(*args): |
| 239 | pass |
| 240 | |
| 241 | import ply.yacc as yacc |
| 242 | parser = yacc.yacc(debug=False, write_tables=False, errorlog=SilentLogger()) |
| 243 | import sys |
| 244 | |
| 245 | def parse(filename): |
| 246 | return parser.parse(open(filename, 'r').read()) |