code_formatter.py revision 7672:d609cd948ca0 
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
27import __builtin__
28import inspect
29import os
30import re
31import string
32
33class lookup(object):
34    def __init__(self, formatter, frame, *args, **kwargs):
35        self.frame = frame
36        self.formatter = formatter
37        self.dict = self.formatter._dict
38        self.args = args
39        self.kwargs = kwargs
40        self.locals = {}
41
42    def __setitem__(self, item, val):
43        self.locals[item] = val
44
45    def __getitem__(self, item):
46        if item in self.locals:
47            return self.locals[item]
48
49        if item in self.kwargs:
50            return self.kwargs[item]
51
52        if item == '__file__':
53            return self.frame.f_code.co_filename
54
55        if item == '__line__':
56            return self.frame.f_lineno
57
58        if self.formatter.locals and item in self.frame.f_locals:
59            return self.frame.f_locals[item]
60
61        if item in self.dict:
62            return self.dict[item]
63
64        if self.formatter.globals and item in self.frame.f_globals:
65            return self.frame.f_globals[item]
66
67        if item in __builtin__.__dict__:
68            return __builtin__.__dict__[item]
69
70        try:
71            item = int(item)
72            return self.args[item]
73        except ValueError:
74            pass
75        raise IndexError, "Could not find '%s'" % item
76
77class code_formatter_meta(type):
78    pattern = r"""
79    (?:
80      %(delim)s(?P<escaped>%(delim)s)              | # escaped delimiter
81      ^(?P<indent>[ ]*)%(delim)s(?P<lone>%(ident)s)$ | # lone identifier
82      %(delim)s(?P<ident>%(ident)s)                | # identifier
83      %(delim)s%(lb)s(?P<b_ident>%(ident)s)%(rb)s  | # braced identifier
84      %(delim)s(?P<pos>%(pos)s)                    | # positional parameter
85      %(delim)s%(lb)s(?P<b_pos>%(pos)s)%(rb)s      | # braced positional
86      %(delim)s%(ldb)s(?P<eval>.*?)%(rdb)s         | # double braced expression
87      %(delim)s(?P<invalid>)                       # ill-formed delimiter exprs
88    )
89    """
90    def __init__(cls, name, bases, dct):
91        super(code_formatter_meta, cls).__init__(name, bases, dct)
92        if 'pattern' in dct:
93            pat = cls.pattern
94        else:
95            # tuple expansion to ensure strings are proper length
96            lb,rb = cls.braced
97            lb1,lb2,rb2,rb1 = cls.double_braced
98            pat = code_formatter_meta.pattern % {
99                'delim' : re.escape(cls.delim),
100                'ident' : cls.ident,
101                'pos' : cls.pos,
102                'lb' : re.escape(lb),
103                'rb' : re.escape(rb),
104                'ldb' : re.escape(lb1+lb2),
105                'rdb' : re.escape(rb2+rb1),
106                }
107        cls.pattern = re.compile(pat, re.VERBOSE | re.DOTALL | re.MULTILINE)
108
109class code_formatter(object):
110    __metaclass__ = code_formatter_meta
111
112    delim = r'$'
113    ident = r'[_A-z]\w*'
114    pos = r'[0-9]+'
115    braced = r'{}'
116    double_braced = r'{{}}'
117
118    globals = True
119    locals = True
120    fix_newlines = True
121    def __init__(self, *args, **kwargs):
122        self._data = []
123        self._dict = {}
124        self._indent_level = 0
125        self._indent_spaces = 4
126        self.globals = kwargs.pop('globals', type(self).globals)
127        self.locals = kwargs.pop('locals', type(self).locals)
128        self._fix_newlines = \
129                kwargs.pop('fix_newlines', type(self).fix_newlines)
130
131        if args:
132            self.__call__(args)
133
134    def indent(self, count=1):
135        self._indent_level += self._indent_spaces * count
136
137    def dedent(self, count=1):
138        assert self._indent_level >= (self._indent_spaces * count)
139        self._indent_level -= self._indent_spaces * count
140
141    def fix(self, status):
142        previous = self._fix_newlines
143        self._fix_newlines = status
144        return previous
145
146    def nofix(self):
147        previous = self._fix_newlines
148        self._fix_newlines = False
149        return previous
150
151    def clear():
152        self._data = []
153
154    def write(self, *args):
155        f = file(os.path.join(*args), "w")
156        for data in self._data:
157            f.write(data)
158        f.close()
159
160    def __str__(self):
161        data = string.join(self._data, '')
162        self._data = [ data ]
163        return data
164
165    def __getitem__(self, item):
166        return self._dict[item]
167
168    def __setitem__(self, item, value):
169        self._dict[item] = value
170
171    def __delitem__(self, item):
172        del self._dict[item]
173
174    def __contains__(self, item):
175        return item in self._dict
176
177    def __iadd__(self, data):
178        self.append(data)
179
180    def append(self, data):
181        if isinstance(data, code_formatter):
182            self._data.extend(data._data)
183        else:
184            self._append(str(data))
185
186    def _append(self, data):
187        if not self._fix_newlines:
188            self._data.append(data)
189            return
190
191        initial_newline = not self._data or self._data[-1] == '\n'
192        for line in data.splitlines():
193            if line:
194                if self._indent_level:
195                    self._data.append(' ' * self._indent_level)
196                self._data.append(line)
197
198            if line or not initial_newline:
199                self._data.append('\n')
200
201            initial_newline = False
202
203    def __call__(self, *args, **kwargs):
204        if not args:
205            self._data.append('\n')
206            return
207
208        format = args[0]
209        args = args[1:]
210
211        frame = inspect.currentframe().f_back
212
213        l = lookup(self, frame, *args, **kwargs)
214        def convert(match):
215            ident = match.group('lone')
216            # check for a lone identifier
217            if ident:
218                indent = match.group('indent') # must be spaces
219                lone = '%s' % (l[ident], )
220
221                def indent_lines(gen):
222                    for line in gen:
223                        yield indent
224                        yield line
225                return ''.join(indent_lines(lone.splitlines(True)))
226
227            # check for an identifier, braced or not
228            ident = match.group('ident') or match.group('b_ident')
229            if ident is not None:
230                return '%s' % (l[ident], )
231
232            # check for a positional parameter, braced or not
233            pos = match.group('pos') or match.group('b_pos')
234            if pos is not None:
235                pos = int(pos)
236                if pos > len(args):
237                    raise ValueError \
238                        ('Positional parameter #%d not found in pattern' % pos,
239                         code_formatter.pattern)
240                return '%s' % (args[int(pos)], )
241
242            # check for a double braced expression
243            eval_expr = match.group('eval')
244            if eval_expr is not None:
245                result = eval(eval_expr, {}, l)
246                return '%s' % (result, )
247
248            # check for an escaped delimiter
249            if match.group('escaped') is not None:
250                return '$'
251
252            # At this point, we have to match invalid
253            if match.group('invalid') is None:
254                # didn't match invalid!
255                raise ValueError('Unrecognized named group in pattern',
256                                 code_formatter.pattern)
257
258            i = match.start('invalid')
259            if i == 0:
260                colno = 1
261                lineno = 1
262            else:
263                lines = format[:i].splitlines(True)
264                colno = i - reduce(lambda x,y: x+y, (len(z) for z in lines))
265                lineno = len(lines)
266
267                raise ValueError('Invalid format string: line %d, col %d' %
268                                 (lineno, colno))
269
270        d = code_formatter.pattern.sub(convert, format)
271        self._append(d)
272
273__all__ = [ "code_formatter" ]
274
275if __name__ == '__main__':
276    from code_formatter import code_formatter
277    f = code_formatter()
278
279    class Foo(dict):
280        def __init__(self, **kwargs):
281            self.update(kwargs)
282        def __getattr__(self, attr):
283            return self[attr]
284
285    x = "this is a test"
286    l = [ [Foo(x=[Foo(y=9)])] ]
287
288    y = code_formatter()
289    y('''
290{
291    this_is_a_test();
292}
293''')
294    f('    $y')
295    f('''$__file__:$__line__
296{''')
297    f("${{', '.join(str(x) for x in xrange(4))}}")
298    f('${x}')
299    f('$x')
300    f.indent()
301    for i in xrange(5):
302        f('$x')
303        f('$i')
304        f('$0', "zero")
305        f('$1 $0', "zero", "one")
306        f('${0}', "he went")
307        f('${0}asdf', "he went")
308    f.dedent()
309
310    f('''
311    ${{l[0][0]["x"][0].y}}
312}
313''', 1, 9)
314
315    print f,
316