code_formatter.py revision 13709 
1# Copyright (c) 2006-2009 Nathan Binkert <nate@binkert.org>
2# All rights reserved.
3#
4# Redistribution and use in source and binary forms, with or without
5# modification, are permitted provided that the following conditions are
6# met: redistributions of source code must retain the above copyright
7# notice, this list of conditions and the following disclaimer;
8# redistributions in binary form must reproduce the above copyright
9# notice, this list of conditions and the following disclaimer in the
10# documentation and/or other materials provided with the distribution;
11# neither the name of the copyright holders nor the names of its
12# contributors may be used to endorse or promote products derived from
13# this software without specific prior written permission.
14#
15# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
16# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
17# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
18# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
19# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
20# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
21# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
22# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
23# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
25# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26
27from __future__ import print_function
28
29import __builtin__
30import inspect
31import os
32import re
33import string
34
35class lookup(object):
36    def __init__(self, formatter, frame, *args, **kwargs):
37        self.frame = frame
38        self.formatter = formatter
39        self.dict = self.formatter._dict
40        self.args = args
41        self.kwargs = kwargs
42        self.locals = {}
43
44    def __setitem__(self, item, val):
45        self.locals[item] = val
46
47    def __getitem__(self, item):
48        if item in self.locals:
49            return self.locals[item]
50
51        if item in self.kwargs:
52            return self.kwargs[item]
53
54        if item == '__file__':
55            return self.frame.f_code.co_filename
56
57        if item == '__line__':
58            return self.frame.f_lineno
59
60        if self.formatter.locals and item in self.frame.f_locals:
61            return self.frame.f_locals[item]
62
63        if item in self.dict:
64            return self.dict[item]
65
66        if self.formatter.globals and item in self.frame.f_globals:
67            return self.frame.f_globals[item]
68
69        if item in __builtin__.__dict__:
70            return __builtin__.__dict__[item]
71
72        try:
73            item = int(item)
74            return self.args[item]
75        except ValueError:
76            pass
77        raise IndexError("Could not find '%s'" % item)
78
79class code_formatter_meta(type):
80    pattern = r"""
81    (?:
82      %(delim)s(?P<escaped>%(delim)s)              | # escaped delimiter
83      ^(?P<indent>[ ]*)%(delim)s(?P<lone>%(ident)s)$ | # lone identifier
84      %(delim)s(?P<ident>%(ident)s)                | # identifier
85      %(delim)s%(lb)s(?P<b_ident>%(ident)s)%(rb)s  | # braced identifier
86      %(delim)s(?P<pos>%(pos)s)                    | # positional parameter
87      %(delim)s%(lb)s(?P<b_pos>%(pos)s)%(rb)s      | # braced positional
88      %(delim)s%(ldb)s(?P<eval>.*?)%(rdb)s         | # double braced expression
89      %(delim)s(?P<invalid>)                       # ill-formed delimiter exprs
90    )
91    """
92    def __init__(cls, name, bases, dct):
93        super(code_formatter_meta, cls).__init__(name, bases, dct)
94        if 'pattern' in dct:
95            pat = cls.pattern
96        else:
97            # tuple expansion to ensure strings are proper length
98            lb,rb = cls.braced
99            lb1,lb2,rb2,rb1 = cls.double_braced
100            pat = code_formatter_meta.pattern % {
101                'delim' : re.escape(cls.delim),
102                'ident' : cls.ident,
103                'pos' : cls.pos,
104                'lb' : re.escape(lb),
105                'rb' : re.escape(rb),
106                'ldb' : re.escape(lb1+lb2),
107                'rdb' : re.escape(rb2+rb1),
108                }
109        cls.pattern = re.compile(pat, re.VERBOSE | re.DOTALL | re.MULTILINE)
110
111class code_formatter(object):
112    __metaclass__ = code_formatter_meta
113
114    delim = r'$'
115    ident = r'[_A-z]\w*'
116    pos = r'[0-9]+'
117    braced = r'{}'
118    double_braced = r'{{}}'
119
120    globals = True
121    locals = True
122    fix_newlines = True
123    def __init__(self, *args, **kwargs):
124        self._data = []
125        self._dict = {}
126        self._indent_level = 0
127        self._indent_spaces = 4
128        self.globals = kwargs.pop('globals', type(self).globals)
129        self.locals = kwargs.pop('locals', type(self).locals)
130        self._fix_newlines = \
131                kwargs.pop('fix_newlines', type(self).fix_newlines)
132
133        if args:
134            self.__call__(args)
135
136    def indent(self, count=1):
137        self._indent_level += self._indent_spaces * count
138
139    def dedent(self, count=1):
140        assert self._indent_level >= (self._indent_spaces * count)
141        self._indent_level -= self._indent_spaces * count
142
143    def fix(self, status):
144        previous = self._fix_newlines
145        self._fix_newlines = status
146        return previous
147
148    def nofix(self):
149        previous = self._fix_newlines
150        self._fix_newlines = False
151        return previous
152
153    def clear():
154        self._data = []
155
156    def write(self, *args):
157        f = open(os.path.join(*args), "w")
158        for data in self._data:
159            f.write(data)
160        f.close()
161
162    def __str__(self):
163        data = string.join(self._data, '')
164        self._data = [ data ]
165        return data
166
167    def __getitem__(self, item):
168        return self._dict[item]
169
170    def __setitem__(self, item, value):
171        self._dict[item] = value
172
173    def __delitem__(self, item):
174        del self._dict[item]
175
176    def __contains__(self, item):
177        return item in self._dict
178
179    def __iadd__(self, data):
180        self.append(data)
181
182    def append(self, data):
183        if isinstance(data, code_formatter):
184            self._data.extend(data._data)
185        else:
186            self._append(str(data))
187
188    def _append(self, data):
189        if not self._fix_newlines:
190            self._data.append(data)
191            return
192
193        initial_newline = not self._data or self._data[-1] == '\n'
194        for line in data.splitlines():
195            if line:
196                if self._indent_level:
197                    self._data.append(' ' * self._indent_level)
198                self._data.append(line)
199
200            if line or not initial_newline:
201                self._data.append('\n')
202
203            initial_newline = False
204
205    def __call__(self, *args, **kwargs):
206        if not args:
207            self._data.append('\n')
208            return
209
210        format = args[0]
211        args = args[1:]
212
213        frame = inspect.currentframe().f_back
214
215        l = lookup(self, frame, *args, **kwargs)
216        def convert(match):
217            ident = match.group('lone')
218            # check for a lone identifier
219            if ident:
220                indent = match.group('indent') # must be spaces
221                lone = '%s' % (l[ident], )
222
223                def indent_lines(gen):
224                    for line in gen:
225                        yield indent
226                        yield line
227                return ''.join(indent_lines(lone.splitlines(True)))
228
229            # check for an identifier, braced or not
230            ident = match.group('ident') or match.group('b_ident')
231            if ident is not None:
232                return '%s' % (l[ident], )
233
234            # check for a positional parameter, braced or not
235            pos = match.group('pos') or match.group('b_pos')
236            if pos is not None:
237                pos = int(pos)
238                if pos > len(args):
239                    raise ValueError \
240                        ('Positional parameter #%d not found in pattern' % pos,
241                         code_formatter.pattern)
242                return '%s' % (args[int(pos)], )
243
244            # check for a double braced expression
245            eval_expr = match.group('eval')
246            if eval_expr is not None:
247                result = eval(eval_expr, {}, l)
248                return '%s' % (result, )
249
250            # check for an escaped delimiter
251            if match.group('escaped') is not None:
252                return '$'
253
254            # At this point, we have to match invalid
255            if match.group('invalid') is None:
256                # didn't match invalid!
257                raise ValueError('Unrecognized named group in pattern',
258                                 code_formatter.pattern)
259
260            i = match.start('invalid')
261            if i == 0:
262                colno = 1
263                lineno = 1
264            else:
265                lines = format[:i].splitlines(True)
266                colno = i - reduce(lambda x,y: x+y, (len(z) for z in lines))
267                lineno = len(lines)
268
269                raise ValueError('Invalid format string: line %d, col %d' %
270                                 (lineno, colno))
271
272        d = code_formatter.pattern.sub(convert, format)
273        self._append(d)
274
275__all__ = [ "code_formatter" ]
276
277if __name__ == '__main__':
278    from code_formatter import code_formatter
279    f = code_formatter()
280
281    class Foo(dict):
282        def __init__(self, **kwargs):
283            self.update(kwargs)
284        def __getattr__(self, attr):
285            return self[attr]
286
287    x = "this is a test"
288    l = [ [Foo(x=[Foo(y=9)])] ]
289
290    y = code_formatter()
291    y('''
292{
293    this_is_a_test();
294}
295''')
296    f('    $y')
297    f('''$__file__:$__line__
298{''')
299    f("${{', '.join(str(x) for x in range(4))}}")
300    f('${x}')
301    f('$x')
302    f.indent()
303    for i in range(5):
304        f('$x')
305        f('$i')
306        f('$0', "zero")
307        f('$1 $0', "zero", "one")
308        f('${0}', "he went")
309        f('${0}asdf', "he went")
310    f.dedent()
311
312    f('''
313    ${{l[0][0]["x"][0].y}}
314}
315''', 1, 9)
316
317    print(f, end=' ')
318