dsl.py 15.5 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
           'compile_on_disk']
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
    ZeroOrMore, Forward, NegativeLookahead, mixin_comment, full_compilation
101
102
103
104
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, \\
    WHITESPACE_KEYWORD, TOKEN_KEYWORD
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, scanner, dsl_grammar, ast_transformation, compiler):
158
    """Compiles a text in a domain specific language (DSL) with an
159
160
    EBNF-specified grammar. Returns the compiled text or raises a
    compilation error.
161
162
163
    
    Raises:
        CompilationError if any errors occured during compilation
164
165
166
167
    """
    assert isinstance(text_or_file, str)
    assert isinstance(compiler, CompilerBase)
    parser_root, grammar_src = get_grammar_instance(dsl_grammar)
168
    src = load_if_file(text_or_file)
169
170
    result, errors, AST = full_compilation(src, scanner, parser_root,
                                           ast_transformation, compiler)
171
    if errors:  raise CompilationError(errors, src, grammar_src, AST)
172
173
174
    return result


175
def compileEBNF(ebnf_src, ebnf_grammar_obj=None, source_only=False):
176
177
178
179
180
181
182
    """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. 
183
184
185
186
187
188
189
190

    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.
191
192
        source_only (bool):  If True, the source code of the Grammar
            class is returned instead of the class itself.
193
194
    Returns:
        A Grammar class that can be instantiated for parsing a text
195
        which conforms to the language defined by ``ebnf_src``.
196
197
    """
    grammar = ebnf_grammar_obj or EBNFGrammar()
198
    grammar_src = compileDSL(ebnf_src, nil_scanner, grammar, EBNFTransform, EBNFCompiler())
199
200
    return grammar_src if source_only else \
        compile_python_object(DHPARSER_IMPORTS + grammar_src, '\w*Grammar$')
201
202


203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
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$')
218
        ast = compile_python_object(imports + ast_py, '\w*Transform$')
219
220
221
222
        compiler = compile_python_object(imports + compiler_py, '\w*Compiler$')
    else:
        # assume source is an ebnf grammar
        parser_py, errors, AST = full_compilation(
223
            source, None, EBNFGrammar(), EBNFTransform, EBNFCompiler())
224
225
226
        if errors:
            raise GrammarError('\n\n'.join(errors), source)
        scanner = nil_scanner
227
        ast = EBNFTransform
228
229
        compiler = EBNFCompiler
    parser = compile_python_object(DHPARSER_IMPORTS + parser_py, '\w*Grammar$')
230
231
232
233
234
235
236
237
238
239
240
241
242

    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. 

243
    Args:
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
        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


259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
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)
    return compileDSL(text_or_file, scanner, parser(), ast, compiler())


def compile_on_disk(source_file, compiler_suite="", extension=".xml"):
283
284
285
    """Compiles the a source file with a given compiler and writes the
    result to a file.

Eckhart Arnold's avatar
Eckhart Arnold committed
286
287
288
289
290
291
    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
292
    need to be delimited section marker blocks.). `compile_on_disk()`
Eckhart Arnold's avatar
Eckhart Arnold committed
293
294
    returns a list of error messages or an empty list if no errors
    occurred.
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
    
    Parameters:
        source_file(str):  The file name of the source text to be
            compiled.
        compiler_suite(str):  The file name of the compiler suite
            (usually ending with '_compiler.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 list of error messages or an empty list if there were no 
        errors. 
Eckhart Arnold's avatar
Eckhart Arnold committed
312
    """
313
    filepath = os.path.normpath(source_file)
314
315
    # with open(source_file, encoding="utf-8") as f:
    #     source = f.read()
316
    rootname = os.path.splitext(filepath)[0]
317
    compiler_name = os.path.basename(rootname)
318
    if compiler_suite:
319
320
        scanner, pclass, trans, cclass = load_compiler_suite(compiler_suite)
        parser = pclass()
321
        compiler1 = cclass()
322
323
324
    else:
        scanner = nil_scanner
        parser = EBNFGrammar()
325
        trans = EBNFTransform
326
327
        compiler1 = EBNFCompiler(compiler_name, source_file)
    result, errors, ast = full_compilation(source_file, scanner, parser, trans, compiler1)
328
329
330
    if errors:
        return errors

331
    elif trans == EBNFTransform:  # either an EBNF- or no compiler suite given
332
333
        global SECTION_MARKER, RX_SECTION_MARKER, SCANNER_SECTION, PARSER_SECTION, \
            AST_SECTION, COMPILER_SECTION, END_SECTIONS_MARKER
334
        f = None
335
336
337
        try:
            f = open(rootname + '_compiler.py', 'r', encoding="utf-8")
            source = f.read()
338
339
            sections = RX_SECTION_MARKER.split(source)
            intro, imports, scanner, parser, ast, compiler, outro = sections
340
        except (PermissionError, FileNotFoundError, IOError) as error:
341
            intro, imports, scanner, parser, ast, compiler, outro = '', '', '', '', '', '', ''
342
343
344
345
        except ValueError as error:
            raise ValueError('File "' + rootname + '_compiler.py" seems to be corrupted. '
                                                   'Please delete or repair file manually!')
        finally:
346
347
348
            if f:
                f.close()
                f = None
349

350
351
352
353
354
355
356
357
358
359
360
361
362
        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()

363
364
365
366
        try:
            f = open(rootname + '_compiler.py', 'w', encoding="utf-8")
            f.write(intro)
            f.write(SECTION_MARKER.format(marker=SYMBOLS_SECTION))
367
            f.write(imports)
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
            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 []