dsl.py 13.9 KB
Newer Older
1
2
#!/usr/bin/python3

3
"""dsl.py - Support for domain specific notations for DHParser
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

Copyright 2016  by Eckhart Arnold (arnold@badw.de)
                Bavarian Academy of Sciences an Humanities (badw.de)

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied.  See the License for the specific language governing
permissions and limitations under the License.

Module ``DSLsupport`` contains various functions to support the
compilation of domain specific languages based on an EBNF-grammar.
"""

di68kap's avatar
di68kap committed
24
import collections
25
import os
26

27
28
29
30
31
try:
    import regex as re
except ImportError:
    import re

32
from .ebnf import EBNFGrammar, EBNFTransform, EBNFCompiler, grammar_changed
33
34
from .toolkit import load_if_file, is_python_code, compile_python_object
from .parsers import GrammarBase, CompilerBase, full_compilation, nil_scanner
35
from .syntaxtree import Node
36

Eckhart Arnold's avatar
Eckhart Arnold committed
37
38
39
40
41

__all__ = ['GrammarError',
           'CompilationError',
           'load_compiler_suite',
           'compileDSL',
42
           'run_compiler']
Eckhart Arnold's avatar
Eckhart Arnold committed
43
44


45
46
47
48
49
50
51
52
53
SECTION_MARKER = """\n
#######################################################################
#
# {marker}
#
#######################################################################
\n"""

RX_SECTION_MARKER = re.compile(SECTION_MARKER.format(marker=r'.*?SECTION.*?'))
54
RX_WHITESPACE = re.compile('\s*')
55
56
57
58
59
60

SYMBOLS_SECTION = "SYMBOLS SECTION - Can be edited. Changes will be preserved."
SCANNER_SECTION = "SCANNER SECTION - Can be edited. Changes will be preserved."
PARSER_SECTION = "PARSER SECTION - Don't edit! CHANGES WILL BE OVERWRITTEN!"
AST_SECTION = "AST SECTION - Can be edited. Changes will be preserved."
COMPILER_SECTION = "COMPILER SECTION - Can be edited. Changes will be preserved."
di68kap's avatar
di68kap committed
61
END_SECTIONS_MARKER = "END OF DHPARSER-SECTIONS"
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85


class GrammarError(Exception):
    """Raised when (already) the grammar of a domain specific language (DSL)
    contains errors.
    """

    def __init__(self, error_messages, grammar_src):
        self.error_messages = error_messages
        self.grammar_src = grammar_src


class CompilationError(Exception):
    """Raised when a string or file in a domain specific language (DSL)
    contains errors.
    """

    def __init__(self, error_messages, dsl_text, dsl_grammar, AST):
        self.error_messages = error_messages
        self.dsl_text = dsl_text
        self.dsl_grammar = dsl_grammar
        self.AST = AST

    def __str__(self):
86
        return '\n'.join(self.error_messages)
87
88


89
DHPARSER_IMPORTS = '''
di68kap's avatar
di68kap committed
90
from functools import partial
91
import sys
di68kap's avatar
di68kap committed
92
93
94
95
try:
    import regex as re
except ImportError:
    import re
96
from DHParser.toolkit import load_if_file    
97
from DHParser.parsers import GrammarBase, CompilerBase, nil_scanner, \\
di68kap's avatar
di68kap committed
98
99
    Lookbehind, Lookahead, Alternative, Pop, Required, Token, \\
    Optional, NegativeLookbehind, OneOrMore, RegExp, Retrieve, Sequence, RE, Capture, \\
100
101
    ZeroOrMore, Forward, NegativeLookahead, mixin_comment, full_compilation
from DHParser.syntaxtree import Node, remove_enclosing_delimiters, remove_children_if, \\
di68kap's avatar
di68kap committed
102
103
104
    reduce_single_child, replace_by_single_child, remove_whitespace, TOKEN_KEYWORD, \\
    no_operation, remove_expendables, remove_tokens, flatten, WHITESPACE_KEYWORD, \\
    is_whitespace, is_expendable
105
'''
di68kap's avatar
di68kap committed
106

107

108
109
110
111
DHPARSER_COMPILER = '''
def compile_{NAME}(source):
    """Compiles ``source`` and returns (result, errors, ast).
    """
112
    return full_compilation(source, {NAME}Scanner,
113
        {NAME}Grammar(), {NAME}Transform, {NAME}Compiler())
114
115
116
117
118
119
120
121
122
123
124
125
126
127

if __name__ == "__main__":
    if len(sys.argv) > 1:
        result, errors, ast = compile_{NAME}(sys.argv[1])
        if errors:
            for error in errors:
                print(error)
                sys.exit(1)
        else:
            print(result)
    else:
        print("Usage: {NAME}_compiler.py [FILENAME]")
'''

128
129
130
131
132
133
134
135
136
137
138
139
140

def get_grammar_instance(grammar):
    """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 GrammarBase-derived grammar class or an instance of
    such a class (i.e. a grammar object already).
    """
    if isinstance(grammar, str):
        # read grammar
        grammar_src = load_if_file(grammar)
        if is_python_code(grammar):
            parser_py, errors, AST = grammar_src, '', None
        else:
141
            parser_py, errors, AST = full_compilation(grammar_src, None,
142
                                                      EBNFGrammar(), EBNFTransform, EBNFCompiler())
143
        if errors:
144
            raise GrammarError('\n\n'.join(errors), grammar_src)
di68kap's avatar
di68kap committed
145
        parser_root = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w*Grammar$')()
146
147
148
149
150
151
152
153
154
155
156
    else:
        # assume that dsl_grammar is a ParserHQ-object or Grammar class
        grammar_src = ''
        if isinstance(grammar, GrammarBase):
            parser_root = grammar
        else:
            # assume `grammar` is a grammar class and get the root object
            parser_root = grammar()
    return parser_root, grammar_src


157
def compileDSL(text_or_file, dsl_grammar, ast_transformation, compiler,
158
159
               scanner=nil_scanner):
    """Compiles a text in a domain specific language (DSL) with an
160
161
    EBNF-specified grammar. Returns the compiled text or raises a
    compilation error.
162
163
164
    
    Raises:
        CompilationError if any errors occured during compilation
165
166
167
168
    """
    assert isinstance(text_or_file, str)
    assert isinstance(compiler, CompilerBase)
    parser_root, grammar_src = get_grammar_instance(dsl_grammar)
169
    src = load_if_file(text_or_file)
170
171
    result, errors, AST = full_compilation(src, scanner, parser_root,
                                           ast_transformation, compiler)
172
    if errors:  raise CompilationError(errors, src, grammar_src, AST)
173
174
175
    return result


176
def compileEBNF(ebnf_src, ebnf_grammar_obj=None, source_only=False):
177
178
179
180
181
182
183
    """Compiles an EBNF source file into a Grammar class.

    Please note: This functions returns a class which must be 
    instantiated before calling its parse()-method! Calling the method
    directly from the class (which is technically possible in python
    yields an error message complaining about a missing parameter,
    the cause of which may not be obvious at first sight. 
184
185
186
187
188
189
190
191

    Args:
        ebnf_src(str):  Either the file name of an EBNF grammar or
            the EBNF grammar itself as a string.
        ebnf_grammar_obj:  An existing instance of the 
            DHParser.EBNFcompiler.EBNFGrammar object. This can speed
            up compilation, because no new EBNFGrammar object needs to
            be instantiated.
192
193
        source_only (bool):  If True, the source code of the Grammar
            class is returned instead of the class itself.
194
195
    Returns:
        A Grammar class that can be instantiated for parsing a text
196
        which conforms to the language defined by ``ebnf_src``.
197
198
    """
    grammar = ebnf_grammar_obj or EBNFGrammar()
199
    grammar_src = compileDSL(ebnf_src, grammar, EBNFTransform, EBNFCompiler())
200
201
    return grammar_src if source_only else \
        compile_python_object(DHPARSER_IMPORTS + grammar_src, '\w*Grammar$')
202
203


204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
def load_compiler_suite(compiler_suite):
    """Extracts a compiler suite from file or string ``compiler suite``
    and returns it as a tuple (scanner, parser, ast, compiler).
    """
    global RX_SECTION_MARKER
    assert isinstance(compiler_suite, str)
    source = load_if_file(compiler_suite)
    if is_python_code(compiler_suite):
        try:
            intro, imports, scanner_py, parser_py, ast_py, compiler_py, outro = \
                RX_SECTION_MARKER.split(source)
        except ValueError as error:
            raise AssertionError('File "' + compiler_suite + '" seems to be corrupted. '
                                 'Please delete or repair file manually.')
        scanner = compile_python_object(imports + scanner_py, '\w*Scanner$')
        ast = compile_python_object(imports + ast_py, '\w*Pipeline$')
        compiler = compile_python_object(imports + compiler_py, '\w*Compiler$')
    else:
        # assume source is an ebnf grammar
        parser_py, errors, AST = full_compilation(
224
            source, None, EBNFGrammar(), EBNFTransform, EBNFCompiler())
225
226
227
        if errors:
            raise GrammarError('\n\n'.join(errors), source)
        scanner = nil_scanner
228
        ast = EBNFTransform
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
        compiler = EBNFCompiler()
    parser = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w*Grammar$')()

    return scanner, parser, ast, compiler


def suite_outdated(compiler_suite, grammar_source):
    """Returns ``True``  if the ``compile_suite`` needs to be updated.
     
    An update is needed, if either the grammar in the compieler suite 
    does not reflect the latest changes of ``grammar_source`` or if 
    sections from the compiler suite have diligently been overwritten
    with whitespace order to trigger their recreation. Note: Do not
    delete or overwrite the section marker itself. 

    Parameters:
        compiler_suite:  the parser class representing the grammar
            or the file name of a compiler suite containing the grammar
        grammar_source:  File name or string representation of the
            EBNF code of the grammar

    Returns (bool):
        True, if ``compiler_suite`` seems to be out of date.
    """
    try:
        scanner, grammar, ast, compiler = load_compiler_suite(compiler_suite)
        return grammar_changed(grammar, grammar_source)
    except ValueError:
        return True


260
261
262
263
def run_compiler(source_file, compiler_suite="", extension=".xml"):
    """Compiles the a source file with a given compiler and writes the
    result to a file.

Eckhart Arnold's avatar
Eckhart Arnold committed
264
265
266
267
268
269
270
271
272
273
    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 scanner, 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.). `run_compiler()`
    returns a list of error messages or an empty list if no errors
    occurred.
    """
274
    filepath = os.path.normpath(source_file)
275
276
    # with open(source_file, encoding="utf-8") as f:
    #     source = f.read()
277
    rootname = os.path.splitext(filepath)[0]
278
    compiler_name = os.path.basename(rootname)
279
280
    if compiler_suite:
        scanner, parser, trans, cclass = load_compiler_suite(compiler_suite)
281
        compiler1 = cclass()
282
283
284
    else:
        scanner = nil_scanner
        parser = EBNFGrammar()
285
        trans = EBNFTransform
286
287
        compiler1 = EBNFCompiler(compiler_name, source_file)
    result, errors, ast = full_compilation(source_file, scanner, parser, trans, compiler1)
288
289
290
    if errors:
        return errors

291
    elif trans == EBNFTransform:  # either an EBNF- or no compiler suite given
292
293
294
295
296
297
298
        f = None

        global SECTION_MARKER, RX_SECTION_MARKER, SCANNER_SECTION, PARSER_SECTION, \
            AST_SECTION, COMPILER_SECTION, END_SECTIONS_MARKER
        try:
            f = open(rootname + '_compiler.py', 'r', encoding="utf-8")
            source = f.read()
299
300
            sections = RX_SECTION_MARKER.split(source)
            intro, imports, scanner, parser, ast, compiler, outro = sections
301
        except (PermissionError, FileNotFoundError, IOError) as error:
302
            intro, imports, scanner, parser, ast, compiler, outro = '', '', '', '', '', '', ''
303
304
305
306
307
308
        except ValueError as error:
            raise ValueError('File "' + rootname + '_compiler.py" seems to be corrupted. '
                                                   'Please delete or repair file manually!')
        finally:
            if f:  f.close()

309
310
311
312
313
314
315
316
317
318
319
320
321
        if RX_WHITESPACE.fullmatch(intro):
            intro = '#!/usr/bin/python'
        if RX_WHITESPACE.fullmatch(outro):
            outro = DHPARSER_COMPILER.format(NAME=compiler_name)
        if RX_WHITESPACE.fullmatch(imports):
            imports = DHPARSER_IMPORTS
        if RX_WHITESPACE.fullmatch(scanner):
            scanner = compiler1.gen_scanner_skeleton()
        if RX_WHITESPACE.fullmatch(ast):
            ast = compiler1.gen_AST_skeleton()
        if RX_WHITESPACE.fullmatch(compiler):
            compiler = compiler1.gen_compiler_skeleton()

322
323
324
325
        try:
            f = open(rootname + '_compiler.py', 'w', encoding="utf-8")
            f.write(intro)
            f.write(SECTION_MARKER.format(marker=SYMBOLS_SECTION))
326
            f.write(imports)
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
            f.write(SECTION_MARKER.format(marker=SCANNER_SECTION))
            f.write(scanner)
            f.write(SECTION_MARKER.format(marker=PARSER_SECTION))
            f.write(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 "' + rootname + '_compiler.py" because of: '
                  + "\n# ".join(str(error).split('\n)')))
            print(result)
        finally:
            if f:  f.close()

    else:
        try:
            f = open(rootname + extension, 'w', encoding="utf-8")
            if isinstance(result, Node):
                f.write(result.as_xml())
            else:
                f.write(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 []