source: trunk/CifFile/YappsStarParser_1_0.py @ 4082

Last change on this file since 4082 was 4082, checked in by vondreele, 4 years ago
File size: 14.1 KB
Line 
1# To maximize python3/python2 compatibility
2from __future__ import print_function
3from __future__ import unicode_literals
4from __future__ import division
5from __future__ import absolute_import
6
7from .StarFile import StarBlock,StarFile,StarList,StarDict
8from io import StringIO
9# An alternative specification for the Cif Parser, based on Yapps2
10# by Amit Patel (http://theory.stanford.edu/~amitp/Yapps)
11#
12# helper code: we define our match tokens
13lastval = ''
14def monitor(location,value):
15    global lastval
16    #print 'At %s: %s' % (location,repr(value))
17    lastval = repr(value)
18    return value
19
20# Strip extras gets rid of leading and trailing whitespace, and
21# semicolons.
22def stripextras(value):
23     from .StarFile import remove_line_folding, remove_line_prefix
24     # we get rid of semicolons and leading/trailing terminators etc.
25     import re
26     jj = re.compile("[\n\r\f \t\v]*")
27     semis = re.compile("[\n\r\f \t\v]*[\n\r\f]\n*;")
28     cut = semis.match(value)
29     if cut:        #we have a semicolon-delimited string
30          nv = value[cut.end():len(value)-2]
31          try:
32             if nv[-1]=='\r': nv = nv[:-1]
33          except IndexError:    #empty data value
34             pass
35          # apply protocols
36          nv = remove_line_prefix(nv)
37          nv = remove_line_folding(nv)
38          return nv
39     else:
40          cut = jj.match(value)
41          if cut:
42               return stripstring(value[cut.end():])
43          return value
44
45# helper function to get rid of inverted commas etc.
46
47def stripstring(value):
48     if value:
49         if value[0]== '\'' and value[-1]=='\'':
50           return value[1:-1]
51         if value[0]=='"' and value[-1]=='"':
52           return value[1:-1]
53     return value
54
55# helper function to get rid of triple quotes
56def striptriple(value):
57    if value:
58        if value[:3] == '"""' and value[-3:] == '"""':
59            return value[3:-3]
60        if value[:3] == "'''" and value[-3:] == "'''":
61            return value[3:-3]
62    return value
63
64# helper function to populate a StarBlock given a list of names
65# and values .
66#
67# Note that there may be an empty list at the very end of our itemlists,
68# so we remove that if necessary.
69#
70
71def makeloop(target_block,loopdata):
72    loop_seq,itemlists = loopdata
73    if itemlists[-1] == []: itemlists.pop(-1)
74    # print 'Making loop with %s' % repr(itemlists)
75    step_size = len(loop_seq)
76    for col_no in range(step_size):
77       target_block.AddItem(loop_seq[col_no], itemlists[col_no::step_size],precheck=True)
78    # print 'Makeloop constructed %s' % repr(loopstructure)
79    # now construct the loop
80    try:
81        target_block.CreateLoop(loop_seq)  #will raise ValueError on problem
82    except ValueError:
83        error_string =  'Incorrect number of loop values for loop containing %s' % repr(loop_seq)
84        print(error_string, file=sys.stderr)
85        raise ValueError(error_string)
86
87# return an object with the appropriate amount of nesting
88def make_empty(nestlevel):
89    gd = []
90    for i in range(1,nestlevel):
91        gd = [gd]
92    return gd
93
94# this function updates a dictionary first checking for name collisions,
95# which imply that the CIF is invalid.  We need case insensitivity for
96# names.
97
98# Unfortunately we cannot check loop item contents against non-loop contents
99# in a non-messy way during parsing, as we may not have easy access to previous
100# key value pairs in the context of our call (unlike our built-in access to all
101# previous loops).
102# For this reason, we don't waste time checking looped items against non-looped
103# names during parsing of a data block.  This would only match a subset of the
104# final items.   We do check against ordinary items, however.
105#
106# Note the following situations:
107# (1) new_dict is empty -> we have just added a loop; do no checking
108# (2) new_dict is not empty -> we have some new key-value pairs
109#
110def cif_update(old_dict,new_dict,loops):
111    old_keys = map(lambda a:a.lower(),old_dict.keys())
112    if new_dict != {}:    # otherwise we have a new loop
113        #print 'Comparing %s to %s' % (repr(old_keys),repr(new_dict.keys()))
114        for new_key in new_dict.keys():
115            if new_key.lower() in old_keys:
116                raise CifError("Duplicate dataname or blockname %s in input file" % new_key)
117            old_dict[new_key] = new_dict[new_key]
118#
119# this takes two lines, so we couldn't fit it into a one line execution statement...
120def order_update(order_array,new_name):
121    order_array.append(new_name)
122    return new_name
123
124# and finally...turn a sequence into a python dict (thanks to Stackoverflow)
125def pairwise(iterable):
126    it = iter(iterable)
127    while 1:
128        yield next(it), next(it)
129
130
131# Begin -- grammar generated by Yapps
132import sys, re
133from . import yapps3_compiled_rt as yappsrt
134
135class StarParserScanner(yappsrt.Scanner):
136    def __init__(self, *args,**kwargs):
137        patterns = [
138         ('([ \t\n\r](?!;))|[ \t]', '([ \t\n\r](?!;))|[ \t]'),
139         ('(#.*[\n\r](?!;))|(#.*)', '(#.*[\n\r](?!;))|(#.*)'),
140         ('LBLOCK', '(L|l)(O|o)(O|o)(P|p)_'),
141         ('GLOBAL', '(G|g)(L|l)(O|o)(B|b)(A|a)(L|l)_'),
142         ('STOP', '(S|s)(T|t)(O|o)(P|p)_'),
143         ('save_heading', '(S|s)(A|a)(V|v)(E|e)_[][!%&\\(\\)*+,./:<=>?@0-9A-Za-z\\\\^`{}\\|~"#$\';_-]+'),
144         ('save_end', '(S|s)(A|a)(V|v)(E|e)_'),
145         ('data_name', '_[][!%&\\(\\)*+,./:<=>?@0-9A-Za-z\\\\^`{}\\|~"#$\';_-]+'),
146         ('data_heading', '(D|d)(A|a)(T|t)(A|a)_[][!%&\\(\\)*+,./:<=>?@0-9A-Za-z\\\\^`{}\\|~"#$\';_-]+'),
147         ('start_sc_line', '(\n|\r\n);([^\n\r])*(\r\n|\r|\n)+'),
148         ('sc_line_of_text', '[^;\r\n]([^\r\n])*(\r\n|\r|\n)+'),
149         ('end_sc_line', ';'),
150         ('data_value_1', '((?!(((S|s)(A|a)(V|v)(E|e)_[^\\s]*)|((G|g)(L|l)(O|o)(B|b)(A|a)(L|l)_[^\\s]*)|((S|s)(T|t)(O|o)(P|p)_[^\\s]*)|((D|d)(A|a)(T|t)(A|a)_[^\\s]*)))[^\\s"#$\'_][^\\s]*)|\'((\'(?=\\S))|([^\n\r\x0c\']))*\'+|"(("(?=\\S))|([^\n\r"]))*"+'),
151         ('END', '$'),
152        ]
153        yappsrt.Scanner.__init__(self,patterns,['([ \t\n\r](?!;))|[ \t]', '(#.*[\n\r](?!;))|(#.*)'],*args,**kwargs)
154
155class StarParser(yappsrt.Parser):
156    Context = yappsrt.Context
157    def input(self, prepared, _parent=None):
158        _context = self.Context(_parent, self._scanner, self._pos, 'input', [prepared])
159        _token = self._peek('END', 'data_heading')
160        if _token == 'data_heading':
161            dblock = self.dblock(prepared, _context)
162            allblocks = prepared;allblocks.merge_fast(dblock)
163            while self._peek('END', 'data_heading') == 'data_heading':
164                dblock = self.dblock(prepared, _context)
165                allblocks.merge_fast(dblock)
166            if self._peek() not in ['END', 'data_heading']:
167                raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['END', 'data_heading']))
168            END = self._scan('END')
169        else: # == 'END'
170            END = self._scan('END')
171            allblocks = prepared
172        return allblocks
173
174    def dblock(self, prepared, _parent=None):
175        _context = self.Context(_parent, self._scanner, self._pos, 'dblock', [prepared])
176        data_heading = self._scan('data_heading')
177        heading = data_heading[5:];thisbc=StarFile(characterset='unicode',standard=prepared.standard);newname = thisbc.NewBlock(heading,StarBlock(overwrite=False));act_block=thisbc[newname]
178        while self._peek('save_heading', 'LBLOCK', 'data_name', 'save_end', 'END', 'data_heading') in ['save_heading', 'LBLOCK', 'data_name']:
179            _token = self._peek('save_heading', 'LBLOCK', 'data_name')
180            if _token != 'save_heading':
181                dataseq = self.dataseq(thisbc[heading], _context)
182            else: # == 'save_heading'
183                save_frame = self.save_frame(_context)
184                thisbc.merge_fast(save_frame,parent=act_block)
185        if self._peek() not in ['save_heading', 'LBLOCK', 'data_name', 'save_end', 'END', 'data_heading']:
186            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['save_heading', 'LBLOCK', 'data_name', 'save_end', 'END', 'data_heading']))
187        thisbc[heading].setmaxnamelength(thisbc[heading].maxnamelength);return (monitor('dblock',thisbc))
188
189    def dataseq(self, starblock, _parent=None):
190        _context = self.Context(_parent, self._scanner, self._pos, 'dataseq', [starblock])
191        data = self.data(starblock, _context)
192        while self._peek('LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading') in ['LBLOCK', 'data_name']:
193            data = self.data(starblock, _context)
194        if self._peek() not in ['LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading']:
195            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading']))
196
197    def data(self, currentblock, _parent=None):
198        _context = self.Context(_parent, self._scanner, self._pos, 'data', [currentblock])
199        _token = self._peek('LBLOCK', 'data_name')
200        if _token == 'LBLOCK':
201            top_loop = self.top_loop(_context)
202            makeloop(currentblock,top_loop)
203        else: # == 'data_name'
204            datakvpair = self.datakvpair(_context)
205            currentblock.AddItem(datakvpair[0],datakvpair[1],precheck=True)
206
207    def datakvpair(self, _parent=None):
208        _context = self.Context(_parent, self._scanner, self._pos, 'datakvpair', [])
209        data_name = self._scan('data_name')
210        data_value = self.data_value(_context)
211        return [data_name,data_value]
212
213    def data_value(self, _parent=None):
214        _context = self.Context(_parent, self._scanner, self._pos, 'data_value', [])
215        _token = self._peek('data_value_1', 'start_sc_line')
216        if _token == 'data_value_1':
217            data_value_1 = self._scan('data_value_1')
218            thisval = stripstring(data_value_1)
219        else: # == 'start_sc_line'
220            sc_lines_of_text = self.sc_lines_of_text(_context)
221            thisval = stripextras(sc_lines_of_text)
222        return monitor('data_value',thisval)
223
224    def sc_lines_of_text(self, _parent=None):
225        _context = self.Context(_parent, self._scanner, self._pos, 'sc_lines_of_text', [])
226        start_sc_line = self._scan('start_sc_line')
227        lines = StringIO();lines.write(start_sc_line)
228        while self._peek('end_sc_line', 'sc_line_of_text') == 'sc_line_of_text':
229            sc_line_of_text = self._scan('sc_line_of_text')
230            lines.write(sc_line_of_text)
231        if self._peek() not in ['end_sc_line', 'sc_line_of_text']:
232            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['sc_line_of_text', 'end_sc_line']))
233        end_sc_line = self._scan('end_sc_line')
234        lines.write(end_sc_line);return monitor('sc_line_of_text',lines.getvalue())
235
236    def top_loop(self, _parent=None):
237        _context = self.Context(_parent, self._scanner, self._pos, 'top_loop', [])
238        LBLOCK = self._scan('LBLOCK')
239        loopfield = self.loopfield(_context)
240        loopvalues = self.loopvalues(_context)
241        return loopfield,loopvalues
242
243    def loopfield(self, _parent=None):
244        _context = self.Context(_parent, self._scanner, self._pos, 'loopfield', [])
245        toploop=[]
246        while self._peek('data_name', 'data_value_1', 'start_sc_line') == 'data_name':
247            data_name = self._scan('data_name')
248            toploop.append(data_name)
249        if self._peek() not in ['data_name', 'data_value_1', 'start_sc_line']:
250            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['data_name', 'data_value_1', 'start_sc_line']))
251        return toploop
252
253    def loopvalues(self, _parent=None):
254        _context = self.Context(_parent, self._scanner, self._pos, 'loopvalues', [])
255        data_value = self.data_value(_context)
256        dataloop=[data_value]
257        while self._peek('data_value_1', 'start_sc_line', 'LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading') in ['data_value_1', 'start_sc_line']:
258            data_value = self.data_value(_context)
259            dataloop.append(monitor('loopval',data_value))
260        if self._peek() not in ['data_value_1', 'start_sc_line', 'LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading']:
261            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['data_value_1', 'start_sc_line', 'LBLOCK', 'data_name', 'save_heading', 'save_end', 'END', 'data_heading']))
262        return dataloop
263
264    def save_frame(self, _parent=None):
265        _context = self.Context(_parent, self._scanner, self._pos, 'save_frame', [])
266        save_heading = self._scan('save_heading')
267        savehead = save_heading[5:];savebc = StarFile();newname=savebc.NewBlock(savehead,StarBlock(overwrite=False));act_block=savebc[newname]
268        while self._peek('save_end', 'save_heading', 'LBLOCK', 'data_name', 'END', 'data_heading') in ['save_heading', 'LBLOCK', 'data_name']:
269            _token = self._peek('save_heading', 'LBLOCK', 'data_name')
270            if _token != 'save_heading':
271                dataseq = self.dataseq(savebc[savehead], _context)
272            else: # == 'save_heading'
273                save_frame = self.save_frame(_context)
274                savebc.merge_fast(save_frame,parent=act_block)
275        if self._peek() not in ['save_end', 'save_heading', 'LBLOCK', 'data_name', 'END', 'data_heading']:
276            raise yappsrt.SyntaxError(charpos=self._scanner.get_prev_char_pos(), context=_context, msg='Need one of ' + ', '.join(['save_end', 'save_heading', 'LBLOCK', 'data_name', 'END', 'data_heading']))
277        save_end = self._scan('save_end')
278        return monitor('save_frame',savebc)
279
280
281def parse(rule, text):
282    P = StarParser(StarParserScanner(text))
283    return yappsrt.wrap_error_reporter(P, rule)
284
285# End -- grammar generated by Yapps
286
287
288
Note: See TracBrowser for help on using the repository browser.