dsl.py 18 KB
Newer Older
1
"""dsl.py - Support for domain specific notations for DHParser
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

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.
"""

import os
23

24
25
26
27
28
try:
    import regex as re
except ImportError:
    import re

29
from .ebnf import EBNFTransformer, grammar_changed, \
30
    get_ebnf_scanner, get_ebnf_grammar, get_ebnf_transformer, get_ebnf_compiler
Eckhart Arnold's avatar
Eckhart Arnold committed
31
from .toolkit import logging, load_if_file, is_python_code, compile_python_object
32
from .parsers import Grammar, CompilerBase, compile_source, nil_scanner
33
from .syntaxtree import Node
34

Eckhart Arnold's avatar
Eckhart Arnold committed
35
36
37
38
39

__all__ = ['GrammarError',
           'CompilationError',
           'load_compiler_suite',
           'compileDSL',
Eckhart Arnold's avatar
Eckhart Arnold committed
40
           'raw_compileEBNF',
41
           'compileEBNF',
42
           'parser_factory',
43
           'compile_on_disk']
Eckhart Arnold's avatar
Eckhart Arnold committed
44
45


46
47
48
49
50
51
52
53
54
SECTION_MARKER = """\n
#######################################################################
#
# {marker}
#
#######################################################################
\n"""

RX_SECTION_MARKER = re.compile(SECTION_MARKER.format(marker=r'.*?SECTION.*?'))
55
RX_WHITESPACE = re.compile('\s*')
56
57
58
59
60
61

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
62
END_SECTIONS_MARKER = "END OF DHPARSER-SECTIONS"
63
64


65
DHPARSER_IMPORTS = '''
di68kap's avatar
di68kap committed
66
from functools import partial
Eckhart Arnold's avatar
Eckhart Arnold committed
67
import os
68
import sys
di68kap's avatar
di68kap committed
69
70
71
72
try:
    import regex as re
except ImportError:
    import re
Eckhart Arnold's avatar
Eckhart Arnold committed
73
from DHParser.toolkit import logging, is_filename, load_if_file    
74
from DHParser.parsers import Grammar, CompilerBase, nil_scanner, \\
di68kap's avatar
di68kap committed
75
76
    Lookbehind, Lookahead, Alternative, Pop, Required, Token, \\
    Optional, NegativeLookbehind, OneOrMore, RegExp, Retrieve, Sequence, RE, Capture, \\
77
78
    ZeroOrMore, Forward, NegativeLookahead, mixin_comment, compile_source, \\
    nop_filter, counterpart_filter, accumulating_filter
79
80
81
from DHParser.syntaxtree import Node, traverse, remove_enclosing_delimiters, \\
    remove_children_if, reduce_single_child, replace_by_single_child, remove_whitespace, \\
    no_operation, remove_expendables, remove_tokens, flatten, is_whitespace, is_expendable, \\
82
    collapse, map_content, WHITESPACE_PTYPE, TOKEN_PTYPE
83
'''
di68kap's avatar
di68kap committed
84

85

Eckhart Arnold's avatar
Eckhart Arnold committed
86
DHPARSER_MAIN = '''
87
def compile_src(source):
88
89
    """Compiles ``source`` and returns (result, errors, ast).
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
90
    with logging("LOGS"):
91
        compiler = get_compiler()
Eckhart Arnold's avatar
Eckhart Arnold committed
92
93
94
        cname = compiler.__class__.__name__
        log_file_name = os.path.basename(os.path.splitext(source)[0]) \\
            if is_filename(source) < 0 else cname[:cname.find('.')] + '_out'    
95
96
97
        result = compile_source(source, get_scanner(), 
                                get_grammar(),
                                get_transformer(), compiler)
Eckhart Arnold's avatar
Eckhart Arnold committed
98
99
    return result

100
101
102
103
104
105
106
107
108
109
110

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:
111
        print("Usage: {NAME}Compiler.py [FILENAME]")
112
113
'''

114

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
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):
        return '\n'.join(self.error_messages)


140
def grammar_instance(grammar_representation):
141
142
    """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,
143
    python-code, a Grammar-derived grammar class or an instance of
144
145
    such a class (i.e. a grammar object already).
    """
146
    if isinstance(grammar_representation, str):
147
        # read grammar
148
149
        grammar_src = load_if_file(grammar_representation)
        if is_python_code(grammar_representation):
150
151
            parser_py, errors, AST = grammar_src, '', None
        else:
Eckhart Arnold's avatar
Eckhart Arnold committed
152
153
            with logging(False):
                parser_py, errors, AST = compile_source(grammar_src, None,
Eckhart Arnold's avatar
Eckhart Arnold committed
154
                    get_ebnf_grammar(), get_ebnf_transformer(), get_ebnf_compiler())
155
        if errors:
156
            raise GrammarError('\n\n'.join(errors), grammar_src)
157
        parser_root = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w+Grammar$')()
158
159
160
    else:
        # assume that dsl_grammar is a ParserHQ-object or Grammar class
        grammar_src = ''
161
        if isinstance(grammar_representation, Grammar):
162
            parser_root = grammar_representation
163
        else:
Eckhart Arnold's avatar
Eckhart Arnold committed
164
165
            # assume ``grammar_representation`` is a grammar class and get the root object
            # TODO: further case: grammar_representation is a method
166
            parser_root = grammar_representation()
167
168
169
    return parser_root, grammar_src


170
def compileDSL(text_or_file, scanner, dsl_grammar, ast_transformation, compiler):
171
    """Compiles a text in a domain specific language (DSL) with an
172
173
    EBNF-specified grammar. Returns the compiled text or raises a
    compilation error.
174
175
    
    Raises:
176
        CompilationError if any errors occurred during compilation
177
178
179
    """
    assert isinstance(text_or_file, str)
    assert isinstance(compiler, CompilerBase)
Eckhart Arnold's avatar
Eckhart Arnold committed
180

181
    parser_root, grammar_src = grammar_instance(dsl_grammar)
Eckhart Arnold's avatar
Eckhart Arnold committed
182
183
184
185
186
    result, errors, AST = compile_source(text_or_file, scanner, parser_root,
                                         ast_transformation, compiler)
    if errors:
        src = load_if_file(text_or_file)
        raise CompilationError(errors, src, grammar_src, AST)
187
188
189
    return result


Eckhart Arnold's avatar
Eckhart Arnold committed
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
def raw_compileEBNF(ebnf_src, branding="DSL"):
    """Compiles an EBNF grammar file and returns the compiler object
    that was used and which can now be queried for the result as well
    as skeleton code for scanner, transformer and compiler objects.
    
    Args:
        ebnf_src(str):  Either the file name of an EBNF grammar or
            the EBNF grammar itself as a string.
        branding (str):  Branding name for the compiler suite source
            code. 
    Returns:
        An instance of class ``ebnf.EBNFCompiler``
    Raises:
        CompilationError if any errors occurred during compilation      
    """
    grammar = get_ebnf_grammar()
    compiler = get_ebnf_compiler(branding , ebnf_src)
    compileDSL(ebnf_src, nil_scanner, grammar, EBNFTransformer, compiler)
    return compiler


def compileEBNF(ebnf_src, branding="DSL"):
    """Compiles an EBNF source file and returns the source code of a
    compiler suite with skeletons for scanner, transformer and 
    compiler.
215
216
217
218

    Args:
        ebnf_src(str):  Either the file name of an EBNF grammar or
            the EBNF grammar itself as a string.
219
220
        branding (str):  Branding name for the compiler suite source
            code. 
221
    Returns:
222
        The complete compiler suite skeleton as Python source code.
223
224
    Raises:
        CompilationError if any errors occurred during compilation        
225
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
226
    compiler = raw_compileEBNF(ebnf_src, branding)
227
228
229
    src = ["#/usr/bin/python\n",
           SECTION_MARKER.format(marker=SYMBOLS_SECTION), DHPARSER_IMPORTS,
           SECTION_MARKER.format(marker=SCANNER_SECTION), compiler.gen_scanner_skeleton(),
Eckhart Arnold's avatar
Eckhart Arnold committed
230
           SECTION_MARKER.format(marker=PARSER_SECTION), compiler.result,
231
232
233
           SECTION_MARKER.format(marker=AST_SECTION), compiler.gen_transformer_skeleton(),
           SECTION_MARKER.format(marker=COMPILER_SECTION), compiler.gen_compiler_skeleton(),
           SECTION_MARKER.format(marker=SYMBOLS_SECTION), DHPARSER_MAIN.format(NAME=branding)]
234
235
236
    return '\n'.join(src)


237
def parser_factory(ebnf_src, branding="DSL"):
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
    """Compiles an EBNF grammar and returns a grammar-parser factory
    function for that grammar.

    Args:
        ebnf_src(str):  Either the file name of an EBNF grammar or
            the EBNF grammar itself as a string.
        branding (str or bool):  Branding name for the compiler
            suite source code. 
    
    Returns:
        A factory function for a grammar-parser for texts in the
        language defined by ``ebnf_src``.
    """
    grammar_src = compileDSL(ebnf_src, nil_scanner, get_ebnf_grammar(),
                             get_ebnf_transformer(), get_ebnf_compiler(branding))
253
    return compile_python_object(DHPARSER_IMPORTS + grammar_src, 'get_(?:\w+_)?grammar$')
254
255


256
257
258
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).
Eckhart Arnold's avatar
Eckhart Arnold committed
259
260
261
    
    Returns:
        4-tuple (scanner function, parser class, ast transformer function, compiler class)
262
263
264
265
266
267
268
269
270
271
272
    """
    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.')
273
        # TODO: Compile in one step and pick parts from namespace later ?
274
275
276
        scanner = compile_python_object(imports + scanner_py, 'get_(?:\w+_)?scanner$')
        parser = compile_python_object(imports + parser_py, 'get_(?:\w+_)?grammar$')
        ast = compile_python_object(imports + ast_py, 'get_(?:\w+_)?transformer$')
277
    else:
278
        # assume source is an ebnf grammar. Is there really any reasonable application case for this?
Eckhart Arnold's avatar
Eckhart Arnold committed
279
        with logging(False):
280
            compile_py, errors, AST = compile_source(source, None,
Eckhart Arnold's avatar
Eckhart Arnold committed
281
                get_ebnf_grammar(), get_ebnf_transformer(), get_ebnf_compiler())
282
283
        if errors:
            raise GrammarError('\n\n'.join(errors), source)
284
285
        scanner = get_ebnf_scanner
        ast = get_ebnf_transformer
286
    compiler = compile_python_object(imports + compiler_py, 'get_(?:\w+_)?compiler$')
287
288
289
290

    return scanner, parser, ast, compiler


Eckhart Arnold's avatar
Eckhart Arnold committed
291
def is_outdated(compiler_suite, grammar_source):
292
293
294
295
296
297
298
299
    """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. 

300
    Args:
301
302
303
304
305
306
307
308
309
310
        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)
311
        return grammar_changed(grammar(), grammar_source)
312
313
314
315
    except ValueError:
        return True


316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
def run_compiler(text_or_file, compiler_suite):
    """Compiles a source with a given compiler suite.

    Args:
        text_or_file (str):  Either the file name of the source code or
            the source code directly. (Which is determined by 
            heuristics. If ``text_or_file`` contains at least on
            linefeed then it is always assumed to be a source text and
            not a file name.)
        compiler_suite(str):  File name of the compiler suite to be
            used.
        
    Returns:
        The result of the compilation, the form and type of which 
        depends entirely on the compiler.
        
    Raises:
        CompilerError
    """
    scanner, parser, ast, compiler = load_compiler_suite(compiler_suite)
336
    return compileDSL(text_or_file, scanner(), parser(), ast(), compiler())
337
338
339


def compile_on_disk(source_file, compiler_suite="", extension=".xml"):
340
341
342
    """Compiles the a source file with a given compiler and writes the
    result to a file.

Eckhart Arnold's avatar
Eckhart Arnold committed
343
344
345
346
347
348
    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
349
    need to be delimited section marker blocks.). `compile_on_disk()`
Eckhart Arnold's avatar
Eckhart Arnold committed
350
351
    returns a list of error messages or an empty list if no errors
    occurred.
352
353
354
355
356
    
    Parameters:
        source_file(str):  The file name of the source text to be
            compiled.
        compiler_suite(str):  The file name of the compiler suite
357
            (usually ending with 'Compiler.py'), with which the source
358
359
360
361
362
363
364
365
366
367
368
            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 list of error messages or an empty list if there were no 
        errors. 
Eckhart Arnold's avatar
Eckhart Arnold committed
369
    """
370
    filepath = os.path.normpath(source_file)
371
372
    # with open(source_file, encoding="utf-8") as f:
    #     source = f.read()
373
    rootname = os.path.splitext(filepath)[0]
374
    compiler_name = os.path.basename(rootname)
375
    if compiler_suite:
376
        scanner, parser, trans, cfactory = load_compiler_suite(compiler_suite)
377
    else:
378
379
380
381
382
383
        scanner = get_ebnf_scanner
        parser = get_ebnf_grammar
        trans = get_ebnf_transformer
        cfactory = get_ebnf_compiler
    compiler1 = cfactory(compiler_name, source_file)
    result, errors, ast = compile_source(source_file, scanner(), parser(), trans(), compiler1)
384
385
386
    if errors:
        return errors

387
    elif cfactory == get_ebnf_compiler:  # trans == get_ebnf_transformer or trans == EBNFTransformer:  # either an EBNF- or no compiler suite given
388
        global SECTION_MARKER, RX_SECTION_MARKER, SCANNER_SECTION, PARSER_SECTION, \
Eckhart Arnold's avatar
Eckhart Arnold committed
389
390
            AST_SECTION, COMPILER_SECTION, END_SECTIONS_MARKER, RX_WHITESPACE, \
            DHPARSER_MAIN, DHPARSER_IMPORTS
391
        f = None
392
        try:
393
            f = open(rootname + 'Compiler.py', 'r', encoding="utf-8")
394
            source = f.read()
395
396
            sections = RX_SECTION_MARKER.split(source)
            intro, imports, scanner, parser, ast, compiler, outro = sections
397
        except (PermissionError, FileNotFoundError, IOError) as error:
398
            intro, imports, scanner, parser, ast, compiler, outro = '', '', '', '', '', '', ''
399
        except ValueError as error:
400
            raise ValueError('File "' + rootname + 'Compiler.py" seems to be corrupted. '
401
402
                                                   'Please delete or repair file manually!')
        finally:
403
404
405
            if f:
                f.close()
                f = None
406

407
408
409
        if RX_WHITESPACE.fullmatch(intro):
            intro = '#!/usr/bin/python'
        if RX_WHITESPACE.fullmatch(outro):
Eckhart Arnold's avatar
Eckhart Arnold committed
410
            outro = DHPARSER_MAIN.format(NAME=compiler_name)
411
412
413
        if RX_WHITESPACE.fullmatch(imports):
            imports = DHPARSER_IMPORTS
        if RX_WHITESPACE.fullmatch(scanner):
414
            scanner = compiler1.gen_scanner_skeleton()
415
        if RX_WHITESPACE.fullmatch(ast):
416
            ast = compiler1.gen_transformer_skeleton()
417
        if RX_WHITESPACE.fullmatch(compiler):
418
            compiler = compiler1.gen_compiler_skeleton()
419

420
        try:
421
            f = open(rootname + 'Compiler.py', 'w', encoding="utf-8")
422
423
            f.write(intro)
            f.write(SECTION_MARKER.format(marker=SYMBOLS_SECTION))
424
            f.write(imports)
425
426
427
            f.write(SECTION_MARKER.format(marker=SCANNER_SECTION))
            f.write(scanner)
            f.write(SECTION_MARKER.format(marker=PARSER_SECTION))
428
            f.write(result)
429
430
431
432
433
434
435
            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:
436
            print('# Could not write file "' + rootname + 'Compiler.py" because of: '
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
                  + "\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 []