Exemple #1
0
def selftest() -> bool:
    """Run a simple self-test of DHParser.
    """
    print("DHParser selftest...")
    print("\nSTAGE I:  Trying to compile EBNF-Grammar:\n")
    builtin_ebnf_parser = get_ebnf_grammar()
    docstring = str(builtin_ebnf_parser.__doc__)  # type: str
    ebnf_src = docstring[docstring.find('@'):]
    ebnf_transformer = get_ebnf_transformer()
    ebnf_compiler = get_ebnf_compiler('EBNF')
    result, errors, _ = compile_source(ebnf_src, None, builtin_ebnf_parser,
                                       ebnf_transformer, ebnf_compiler)
    generated_ebnf_parser = cast(str, result)

    if errors:
        print("Selftest FAILED :-(")
        print("\n\n".join(str(err) for err in errors))
        return False
    print(generated_ebnf_parser)
    print("\n\nSTAGE 2: Selfhosting-test: "
          "Trying to compile EBNF-Grammar with generated parser...\n")
    selfhosted_ebnf_parser = compileDSL(ebnf_src, None, generated_ebnf_parser,
                                        ebnf_transformer, ebnf_compiler)
    # ebnf_compiler.gen_transformer_skeleton()
    print(selfhosted_ebnf_parser)
    return True
Exemple #2
0
 def compile_EBNF(self, text: str):
     from DHParser.compile import compile_source
     from DHParser.ebnf import get_ebnf_preprocessor, get_ebnf_grammar, get_ebnf_transformer, \
         get_ebnf_compiler
     compiler = get_ebnf_compiler("EBNFServerAnalyse", text)
     result, messages, _ = compile_source(text, get_ebnf_preprocessor(),
                                          get_ebnf_grammar(),
                                          get_ebnf_transformer(), compiler)
     # TODO: return errors as well as (distilled) information about symbols for code propositions
     return None
Exemple #3
0
def compile_EBNF(text: str, diagnostics_signature: bytes) -> (str, bytes):
    from DHParser.compile import compile_source
    from DHParser.ebnf import get_ebnf_preprocessor, get_ebnf_grammar, get_ebnf_transformer, \
        get_ebnf_compiler
    from DHParser.toolkit import json_dumps
    compiler = get_ebnf_compiler("EBNFServerAnalyse", text)
    result, messages, _ = compile_source(text, get_ebnf_preprocessor(),
                                         get_ebnf_grammar(),
                                         get_ebnf_transformer(), compiler)
    # TODO: return errors as well as (distilled) information about symbols for code propositions
    signature = b''.join(msg.signature() for msg in messages)
    if signature != diagnostics_signature:
        # publish diagnostics only if the same diagnostics have not been reported earlier, already
        diagnostics = [msg.diagnosticObj() for msg in messages]
        return json_dumps(
            diagnostics), signature  # TODO: bytes is not a JSON-type proper!
    else:
        return '[]', signature
Exemple #4
0
def load_compiler_suite(compiler_suite: str) -> \
        Tuple[PreprocessorFactoryFunc, ParserFactoryFunc,
              TransformerFactoryFunc, CompilerFactoryFunc]:
    """
    Extracts a compiler suite from file or string `compiler_suite`
    and returns it as a tuple (preprocessor, parser, ast, compiler).

    Returns:
        4-tuple (preprocessor function, parser class,
                 ast transformer function, compiler class)
    """
    global RX_SECTION_MARKER
    assert isinstance(compiler_suite, str)
    source = load_if_file(compiler_suite)
    dhpath = relative_path(os.path.dirname('.'), DHPARSER_PARENTDIR)
    imports = DHPARSER_IMPORTS.format(dhparser_parentdir=dhpath)
    if is_python_code(compiler_suite):
        sections = split_source(compiler_suite, source)
        _, imports, preprocessor_py, parser_py, ast_py, compiler_py, _ = sections
        # TODO: Compile in one step and pick parts from namespace later ?
        preprocessor = compile_python_object(imports + preprocessor_py,
                                             r'get_(?:\w+_)?preprocessor$')
        parser = compile_python_object(imports + parser_py, r'get_(?:\w+_)?grammar$')
        ast = compile_python_object(imports + ast_py, r'get_(?:\w+_)?transformer$')
    else:
        # Assume source is an ebnf grammar.
        # Is there really any reasonable application case for this?
        lg_dir = suspend_logging()
        compiler_py, messages, _ = compile_source(source, None, get_ebnf_grammar(),
                                                  get_ebnf_transformer(),
                                                  get_ebnf_compiler(compiler_suite, source))
        resume_logging(lg_dir)
        if has_errors(messages):
            raise DefinitionError(only_errors(messages), source)
        preprocessor = get_ebnf_preprocessor
        parser = get_ebnf_grammar
        ast = get_ebnf_transformer
    compiler = compile_python_object(imports + compiler_py, r'get_(?:\w+_)?compiler$')
    if callable(preprocessor) and callable(parser) and callable(Callable) and callable(compiler):
        return preprocessor, parser, ast, compiler
    raise ValueError('Could not generate compiler suite from source code!')
Exemple #5
0
def grammar_instance(grammar_representation) -> Tuple[Grammar, str]:
    """
    Returns a grammar object and the source code of the grammar, from
    the given `grammar`-data which can be either a file name, ebnf-code,
    python-code, a Grammar-derived grammar class or an instance of
    such a class (i.e. a grammar object already).
    """
    if isinstance(grammar_representation, str):
        # read grammar
        grammar_src = load_if_file(grammar_representation)
        if is_python_code(grammar_src):
            parser_py = grammar_src  # type: str
            messages = []            # type: List[Error]
        else:
            lg_dir = suspend_logging()
            result, messages, _ = compile_source(
                grammar_src, None,
                get_ebnf_grammar(), get_ebnf_transformer(), get_ebnf_compiler())
            parser_py = cast(str, result)
            resume_logging(lg_dir)
        if has_errors(messages):
            raise DefinitionError(only_errors(messages), grammar_src)
        imports = DHPARSER_IMPORTS.format(
            dhparser_parentdir=relative_path('.', DHPARSER_PARENTDIR))
        grammar_class = compile_python_object(imports + parser_py, r'\w+Grammar$')
        if inspect.isclass(grammar_class) and issubclass(grammar_class, Grammar):
            parser_root = grammar_class()
        else:
            raise ValueError('Could not compile or Grammar class!')
    else:
        # assume that dsl_grammar is a ParserHQ-object or Grammar class
        grammar_src = ''
        if isinstance(grammar_representation, Grammar):
            parser_root = grammar_representation
        else:
            # assume ``grammar_representation`` is a grammar class and get the root object
            parser_root = grammar_representation()
    return parser_root, grammar_src
Exemple #6
0
def compileDSL(text_or_file: str,
               preprocessor: Optional[PreprocessorFunc],
               dsl_grammar: Union[str, Grammar],
               ast_transformation: TransformationFunc,
               compiler: Compiler) -> Any:
    """
    Compiles a text in a domain specific language (DSL) with an
    EBNF-specified grammar. Returns the compiled text or raises a
    compilation error.

    Raises:
        CompilationError if any errors occurred during compilation
    """
    assert isinstance(text_or_file, str)
    assert isinstance(compiler, Compiler)

    parser, grammar_src = grammar_instance(dsl_grammar)
    result, messages, AST = compile_source(text_or_file, preprocessor, parser,
                                           ast_transformation, compiler)
    if has_errors(messages):
        src = load_if_file(text_or_file)
        raise CompilationError(only_errors(messages), src, grammar_src, AST, result)
    return result
Exemple #7
0
def compile_on_disk(source_file: str, compiler_suite="", extension=".xml") -> Iterable[Error]:
    """
    Compiles the a source file with a given compiler and writes the
    result to a file.

    If no ``compiler_suite`` is given it is assumed that the source
    file is an EBNF grammar. In this case the result will be a Python
    script containing a parser for that grammar as well as the
    skeletons for a preprocessor, AST transformation table, and compiler.
    If the Python script already exists only the parser name in the
    script will be updated. (For this to work, the different names
    need to be delimited section marker blocks.). `compile_on_disk()`
    returns a list of error messages or an empty list if no errors
    occurred.

    Parameters:
        source_file(str):  The file name of the source text to be
            compiled.
        compiler_suite(str):  The file name of the parser/compiler-suite
            (usually ending with 'Parser.py'), with which the source
            file shall be compiled. If this is left empty, the source
            file is assumed to be an EBNF-Grammar that will be compiled
            with the internal EBNF-Compiler.
        extension(str):  The result of the compilation (if successful)
            is written to a file with the same name but a different
            extension than the source file. This parameter sets the
            extension.

    Returns:
        A (potentially empty) list of error or warning messages.
    """
    filepath = os.path.normpath(source_file)
    f = None  # Optional[TextIO]
    with open(source_file, encoding="utf-8") as f:
        source = f.read()
    rootname = os.path.splitext(filepath)[0]
    dhpath = relative_path(os.path.dirname(rootname), DHPARSER_PARENTDIR)
    compiler_name = as_identifier(os.path.basename(rootname))
    if compiler_suite:
        sfactory, pfactory, tfactory, cfactory = load_compiler_suite(compiler_suite)
        compiler1 = cfactory()
    else:
        sfactory = get_ebnf_preprocessor  # PreprocessorFactoryFunc
        pfactory = get_ebnf_grammar       # ParserFactoryFunc
        tfactory = get_ebnf_transformer   # TransformerFactoryFunc
        cfactory = get_ebnf_compiler      # CompilerFactoryFunc
        compiler1 = cfactory()            # Compiler

    is_ebnf_compiler = False  # type: bool
    if isinstance(compiler1, EBNFCompiler):
        is_ebnf_compiler = True
        compiler1.set_grammar_name(compiler_name, source_file)

    result, messages, _ = compile_source(source, sfactory(), pfactory(), tfactory(), compiler1)

    if has_errors(messages):
        return messages

    elif is_ebnf_compiler:
        # trans == get_ebnf_transformer or trans == EBNFTransformer:
        # either an EBNF- or no compiler suite given
        ebnf_compiler = cast(EBNFCompiler, compiler1)  # type: EBNFCompiler
        global SECTION_MARKER, RX_SECTION_MARKER, PREPROCESSOR_SECTION, PARSER_SECTION, \
            AST_SECTION, COMPILER_SECTION, END_SECTIONS_MARKER, RX_WHITESPACE
        f = None
        try:
            parser_name = rootname + 'Parser.py'
            f = open(parser_name, 'r', encoding="utf-8")
            source = f.read()
            sections = split_source(parser_name, source)
            intro, imports, preprocessor, _, ast, compiler, outro = sections
            ast_trans_python_src = imports + ast  # DHPARSER_IMPORTS.format(dhparser_parentdir=dhpath)
            ast_trans_table = dict()  # type: TransformationDict
            try:
                ast_trans_table = compile_python_object(ast_trans_python_src,
                                                        r'(?:\w+_)?AST_transformation_table$')
            except Exception as e:
                if isinstance(e, NameError):
                    err_str = 'NameError "{}" while compiling AST-Transformation. ' \
                              'Possibly due to a forgotten import at the beginning ' \
                              'of the AST-Block (!)'.format(str(e))
                else:
                    err_str = 'Exception {} while compiling AST-Transformation: {}' \
                              .format(str(type(e)), str(e))
                messages.append(Error(err_str, 0, CANNOT_VERIFY_TRANSTABLE_WARNING))
                if is_logging():
                    with open(os.path.join(log_dir(), rootname + '_AST_src.py'), 'w',
                              encoding='utf-8') as f:
                        f.write(ast_trans_python_src)
            messages.extend(ebnf_compiler.verify_transformation_table(ast_trans_table))
            # TODO: Verify compiler
        except (PermissionError, FileNotFoundError, IOError):
            intro, imports, preprocessor, _, ast, compiler, outro = '', '', '', '', '', '', ''
        finally:
            if f:
                f.close()
                f = None

        if RX_WHITESPACE.fullmatch(intro):
            intro = '#!/usr/bin/env python3'
        if RX_WHITESPACE.fullmatch(outro):
            outro = read_template('DSLParser.pyi').format(NAME=compiler_name)
        if RX_WHITESPACE.fullmatch(imports):
            imports = DHParser.ebnf.DHPARSER_IMPORTS.format(dhparser_parentdir=dhpath)
        if RX_WHITESPACE.fullmatch(preprocessor):
            preprocessor = ebnf_compiler.gen_preprocessor_skeleton()
        if RX_WHITESPACE.fullmatch(ast):
            ast = ebnf_compiler.gen_transformer_skeleton()
        if RX_WHITESPACE.fullmatch(compiler):
            compiler = ebnf_compiler.gen_compiler_skeleton()

        compilerscript = rootname + 'Parser.py'
        try:
            f = open(compilerscript, 'w', encoding="utf-8")
            f.write(intro)
            f.write(SECTION_MARKER.format(marker=SYMBOLS_SECTION))
            f.write(imports)
            f.write(SECTION_MARKER.format(marker=PREPROCESSOR_SECTION))
            f.write(preprocessor)
            f.write(SECTION_MARKER.format(marker=PARSER_SECTION))
            f.write(cast(str, result))
            f.write(SECTION_MARKER.format(marker=AST_SECTION))
            f.write(ast)
            f.write(SECTION_MARKER.format(marker=COMPILER_SECTION))
            f.write(compiler)
            f.write(SECTION_MARKER.format(marker=END_SECTIONS_MARKER))
            f.write(outro)
        except (PermissionError, FileNotFoundError, IOError) as error:
            print('# Could not write file "' + compilerscript + '" because of: '
                  + "\n# ".join(str(error).split('\n)')))
            print(result)
        finally:
            if f:
                f.close()

        if platform.system() != "Windows":
            # set file permissions so that the compilerscript can be executed
            st = os.stat(compilerscript)
            os.chmod(compilerscript, st.st_mode | stat.S_IEXEC)

    else:
        f = None
        try:
            f = open(rootname + extension, 'w', encoding="utf-8")
            if isinstance(result, Node):
                if extension.lower() == '.xml':
                    f.write(result.as_xml())
                else:
                    f.write(result.as_sxpr())
            elif isinstance(result, str):
                f.write(result)
            else:
                raise AssertionError('Illegal result type: ' + str(type(result)))
        except (PermissionError, FileNotFoundError, IOError) as error:
            print('# Could not write file "' + rootname + '.py" because of: '
                  + "\n# ".join(str(error).split('\n)')))
            print(result)
        finally:
            if f:
                f.close()

    return messages