Coverage for python/lsst/daf/butler/registry/queries/expressions/parser/parserYacc.py: 21%

162 statements  

« prev     ^ index     » next       coverage.py v7.3.2, created at 2023-12-06 10:53 +0000

1# This file is part of daf_butler. 

2# 

3# Developed for the LSST Data Management System. 

4# This product includes software developed by the LSST Project 

5# (https://www.lsst.org). 

6# See the COPYRIGHT file at the top-level directory of this distribution 

7# for details of code ownership. 

8# 

9# This software is dual licensed under the GNU General Public License and also 

10# under a 3-clause BSD license. Recipients may choose which of these licenses 

11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt, 

12# respectively. If you choose the GPL option then the following text applies 

13# (but note that there is still no warranty even if you opt for BSD instead): 

14# 

15# This program is free software: you can redistribute it and/or modify 

16# it under the terms of the GNU General Public License as published by 

17# the Free Software Foundation, either version 3 of the License, or 

18# (at your option) any later version. 

19# 

20# This program is distributed in the hope that it will be useful, 

21# but WITHOUT ANY WARRANTY; without even the implied warranty of 

22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 

23# GNU General Public License for more details. 

24# 

25# You should have received a copy of the GNU General Public License 

26# along with this program. If not, see <https://www.gnu.org/licenses/>. 

27 

28# type: ignore 

29 

30"""Syntax definition for user expression parser. 

31""" 

32 

33__all__ = ["ParserYacc", "ParserYaccError", "ParseError", "ParserEOFError"] 

34 

35# ------------------------------- 

36# Imports of standard modules -- 

37# ------------------------------- 

38import re 

39import warnings 

40 

41# ----------------------------- 

42# Imports for other modules -- 

43# ----------------------------- 

44import astropy.time 

45 

46# As of astropy 4.2, the erfa interface is shipped independently and 

47# ErfaWarning is no longer an AstropyWarning 

48try: 

49 import erfa 

50except ImportError: 

51 erfa = None 

52 

53from .exprTree import ( 

54 BinaryOp, 

55 Identifier, 

56 IsIn, 

57 NumericLiteral, 

58 Parens, 

59 RangeLiteral, 

60 StringLiteral, 

61 TimeLiteral, 

62 TupleNode, 

63 UnaryOp, 

64 function_call, 

65) 

66from .parserLex import ParserLex 

67from .ply import yacc 

68 

69# ---------------------------------- 

70# Local non-exported definitions -- 

71# ---------------------------------- 

72 

73# The purpose of this regex is to guess time format if it is not explicitly 

74# provided in the string itself 

75_re_time_str = re.compile( 

76 r""" 

77 ((?P<format>\w+)/)? # optionally prefixed by "format/" 

78 (?P<value> 

79 (?P<number>-?(\d+(\.\d*)|(\.\d+))) # floating point number 

80 | 

81 (?P<iso>\d+-\d+-\d+([ T]\d+:\d+(:\d+([.]\d*)?)?)?) # iso(t) [no timezone] 

82 | 

83 (?P<fits>[+]\d+-\d+-\d+(T\d+:\d+:\d+([.]\d*)?)?) # fits 

84 | 

85 (?P<yday>\d+:\d+(:\d+:\d+(:\d+([.]\d*)?)?)?) # yday 

86 ) 

87 (/(?P<scale>\w+))? # optionally followed by "/scale" 

88 $ 

89""", 

90 re.VERBOSE | re.IGNORECASE, 

91) 

92 

93 

94def _parseTimeString(time_str): 

95 """Try to convert time string into astropy.Time. 

96 

97 Parameters 

98 ---------- 

99 time_str : `str` 

100 Input string. 

101 

102 Returns 

103 ------- 

104 time : `astropy.time.Time` 

105 

106 Raises 

107 ------ 

108 ValueError 

109 Raised if input string has unexpected format 

110 """ 

111 # Check for time zone. Python datetime objects can be timezone-aware 

112 # and if one has been stringified then there will be a +00:00 on the end. 

113 # Special case UTC. Fail for other timezones. 

114 time_str = time_str.replace("+00:00", "") 

115 

116 match = _re_time_str.match(time_str) 

117 if not match: 

118 raise ValueError(f'Time string "{time_str}" does not match known formats') 

119 

120 value, fmt, scale = match.group("value", "format", "scale") 

121 if fmt is not None: 

122 fmt = fmt.lower() 

123 if fmt not in astropy.time.Time.FORMATS: 

124 raise ValueError(f'Time string "{time_str}" specifies unknown time format "{fmt}"') 

125 if scale is not None: 

126 scale = scale.lower() 

127 if scale not in astropy.time.Time.SCALES: 

128 raise ValueError(f'Time string "{time_str}" specifies unknown time scale "{scale}"') 

129 

130 # convert number string to floating point 

131 if match.group("number") is not None: 

132 value = float(value) 

133 

134 # guess format if not given 

135 if fmt is None: 

136 if match.group("number") is not None: 

137 fmt = "mjd" 

138 elif match.group("iso") is not None: 

139 if "T" in value or "t" in value: 

140 fmt = "isot" 

141 else: 

142 fmt = "iso" 

143 elif match.group("fits") is not None: 

144 fmt = "fits" 

145 elif match.group("yday") is not None: 

146 fmt = "yday" 

147 assert fmt is not None 

148 

149 # guess scale if not given 

150 if scale is None: 

151 if fmt in ("iso", "isot", "fits", "yday", "unix"): 

152 scale = "utc" 

153 elif fmt == "cxcsec": 

154 scale = "tt" 

155 else: 

156 scale = "tai" 

157 

158 try: 

159 # Hide warnings about future dates 

160 with warnings.catch_warnings(): 

161 warnings.simplefilter("ignore", category=astropy.utils.exceptions.AstropyWarning) 

162 if erfa is not None: 

163 warnings.simplefilter("ignore", category=erfa.ErfaWarning) 

164 value = astropy.time.Time(value, format=fmt, scale=scale) 

165 except ValueError: 

166 # astropy makes very verbose exception that is not super-useful in 

167 # many context, just say we don't like it. 

168 raise ValueError(f'Time string "{time_str}" does not match format "{fmt}"') from None 

169 

170 return value 

171 

172 

173# ------------------------ 

174# Exported definitions -- 

175# ------------------------ 

176 

177 

178class ParserYaccError(Exception): 

179 """Base class for exceptions generated by parser.""" 

180 

181 pass 

182 

183 

184class ParseError(ParserYaccError): 

185 """Exception raised for parsing errors. 

186 

187 Attributes 

188 ---------- 

189 expression : str 

190 Full initial expression being parsed 

191 token : str 

192 Current token at parsing position 

193 pos : int 

194 Current parsing position, offset from beginning of expression in 

195 characters 

196 lineno : int 

197 Current line number in the expression 

198 posInLine : int 

199 Parsing position in current line, 0-based 

200 """ 

201 

202 def __init__(self, expression, token, pos, lineno): 

203 self.expression = expression 

204 self.token = token 

205 self.pos = pos 

206 self.lineno = lineno 

207 self.posInLine = self._posInLine() 

208 msg = "Syntax error at or near '{0}' (line: {1}, pos: {2})" 

209 msg = msg.format(token, lineno, self.posInLine + 1) 

210 ParserYaccError.__init__(self, msg) 

211 

212 def _posInLine(self): 

213 """Return position in current line""" 

214 lines = self.expression.split("\n") 

215 pos = self.pos 

216 for line in lines[: self.lineno - 1]: 

217 # +1 for newline 

218 pos -= len(line) + 1 

219 return pos 

220 

221 

222class ParserEOFError(ParserYaccError): 

223 """Exception raised for EOF-during-parser.""" 

224 

225 def __init__(self): 

226 Exception.__init__(self, "End of input reached while expecting further input") 

227 

228 

229class ParserYacc: 

230 """Class which defines PLY grammar. 

231 

232 Based on MySQL grammar for expressions 

233 (https://dev.mysql.com/doc/refman/5.7/en/expressions.html). 

234 

235 Parameters 

236 ---------- 

237 idMap : `collections.abc.Mapping` [ `str`, `Node` ], optional 

238 Mapping that provides substitutions for identifiers in the expression. 

239 The key in the map is the identifier name, the value is the 

240 `exprTree.Node` instance that will replace identifier in the full 

241 expression. If identifier does not exist in the mapping then 

242 `Identifier` is inserted into parse tree. 

243 **kwargs 

244 optional keyword arguments that are passed to `yacc.yacc` constructor. 

245 """ 

246 

247 def __init__(self, idMap=None, **kwargs): 

248 kw = dict(write_tables=0, debug=False) 

249 kw.update(kwargs) 

250 

251 self.parser = yacc.yacc(module=self, **kw) 

252 self._idMap = idMap or {} 

253 

254 def parse(self, input, lexer=None, debug=False, tracking=False): 

255 """Parse input expression ad return parsed tree object. 

256 

257 This is a trivial wrapper for yacc.LRParser.parse method which 

258 provides lexer if not given in arguments. 

259 

260 Parameters 

261 ---------- 

262 input : str 

263 Expression to parse 

264 lexer : object, optional 

265 Lexer instance, if not given then ParserLex.make_lexer() is 

266 called to create one. 

267 debug : bool, optional 

268 Set to True for debugging output. 

269 tracking : bool, optional 

270 Set to True for tracking line numbers in parser. 

271 """ 

272 # make lexer 

273 if lexer is None: 

274 lexer = ParserLex.make_lexer() 

275 tree = self.parser.parse(input=input, lexer=lexer, debug=debug, tracking=tracking) 

276 return tree 

277 

278 tokens = ParserLex.tokens[:] 

279 

280 precedence = ( 

281 ("left", "OR"), 

282 ("left", "AND"), 

283 ("nonassoc", "OVERLAPS"), # Nonassociative operators 

284 ("nonassoc", "EQ", "NE"), # Nonassociative operators 

285 ("nonassoc", "LT", "LE", "GT", "GE"), # Nonassociative operators 

286 ("left", "ADD", "SUB"), 

287 ("left", "MUL", "DIV", "MOD"), 

288 ("right", "UPLUS", "UMINUS", "NOT"), # unary plus and minus 

289 ) 

290 

291 # this is the starting rule 

292 def p_input(self, p): 

293 """input : expr 

294 | empty 

295 """ 

296 p[0] = p[1] 

297 

298 def p_empty(self, p): 

299 """empty :""" 

300 p[0] = None 

301 

302 def p_expr(self, p): 

303 """expr : expr OR expr 

304 | expr AND expr 

305 | NOT expr 

306 | bool_primary 

307 """ 

308 if len(p) == 4: 

309 p[0] = BinaryOp(lhs=p[1], op=p[2].upper(), rhs=p[3]) 

310 elif len(p) == 3: 

311 p[0] = UnaryOp(op=p[1].upper(), operand=p[2]) 

312 else: 

313 p[0] = p[1] 

314 

315 def p_bool_primary(self, p): 

316 """bool_primary : bool_primary EQ predicate 

317 | bool_primary NE predicate 

318 | bool_primary LT predicate 

319 | bool_primary LE predicate 

320 | bool_primary GE predicate 

321 | bool_primary GT predicate 

322 | bool_primary OVERLAPS predicate 

323 | predicate 

324 """ 

325 if len(p) == 2: 

326 p[0] = p[1] 

327 else: 

328 p[0] = BinaryOp(lhs=p[1], op=p[2], rhs=p[3]) 

329 

330 def p_predicate(self, p): 

331 """predicate : bit_expr IN LPAREN literal_or_id_list RPAREN 

332 | bit_expr NOT IN LPAREN literal_or_id_list RPAREN 

333 | bit_expr 

334 """ 

335 if len(p) == 6: 

336 p[0] = IsIn(lhs=p[1], values=p[4]) 

337 elif len(p) == 7: 

338 p[0] = IsIn(lhs=p[1], values=p[5], not_in=True) 

339 else: 

340 p[0] = p[1] 

341 

342 def p_identifier(self, p): 

343 """identifier : SIMPLE_IDENTIFIER 

344 | QUALIFIED_IDENTIFIER 

345 """ 

346 node = self._idMap.get(p[1]) 

347 if node is None: 

348 node = Identifier(p[1]) 

349 p[0] = node 

350 

351 def p_literal_or_id_list(self, p): 

352 """literal_or_id_list : literal_or_id_list COMMA literal 

353 | literal_or_id_list COMMA identifier 

354 | literal 

355 | identifier 

356 """ 

357 if len(p) == 2: 

358 p[0] = [p[1]] 

359 else: 

360 p[0] = p[1] + [p[3]] 

361 

362 def p_bit_expr(self, p): 

363 """bit_expr : bit_expr ADD bit_expr 

364 | bit_expr SUB bit_expr 

365 | bit_expr MUL bit_expr 

366 | bit_expr DIV bit_expr 

367 | bit_expr MOD bit_expr 

368 | simple_expr 

369 """ 

370 if len(p) == 2: 

371 p[0] = p[1] 

372 else: 

373 p[0] = BinaryOp(lhs=p[1], op=p[2], rhs=p[3]) 

374 

375 def p_simple_expr_lit(self, p): 

376 """simple_expr : literal""" 

377 p[0] = p[1] 

378 

379 def p_simple_expr_id(self, p): 

380 """simple_expr : identifier""" 

381 p[0] = p[1] 

382 

383 def p_simple_expr_function_call(self, p): 

384 """simple_expr : function_call""" 

385 p[0] = p[1] 

386 

387 def p_simple_expr_unary(self, p): 

388 """simple_expr : ADD simple_expr %prec UPLUS 

389 | SUB simple_expr %prec UMINUS 

390 """ 

391 p[0] = UnaryOp(op=p[1], operand=p[2]) 

392 

393 def p_simple_expr_paren(self, p): 

394 """simple_expr : LPAREN expr RPAREN""" 

395 p[0] = Parens(p[2]) 

396 

397 def p_simple_expr_tuple(self, p): 

398 """simple_expr : LPAREN expr COMMA expr RPAREN""" 

399 # For now we only support tuples with two items, 

400 # these are used for time ranges. 

401 p[0] = TupleNode((p[2], p[4])) 

402 

403 def p_literal_num(self, p): 

404 """literal : NUMERIC_LITERAL""" 

405 p[0] = NumericLiteral(p[1]) 

406 

407 def p_literal_num_signed(self, p): 

408 """literal : ADD NUMERIC_LITERAL %prec UPLUS 

409 | SUB NUMERIC_LITERAL %prec UMINUS 

410 """ 

411 p[0] = NumericLiteral(p[1] + p[2]) 

412 

413 def p_literal_str(self, p): 

414 """literal : STRING_LITERAL""" 

415 p[0] = StringLiteral(p[1]) 

416 

417 def p_literal_time(self, p): 

418 """literal : TIME_LITERAL""" 

419 try: 

420 value = _parseTimeString(p[1]) 

421 except ValueError as e: 

422 raise ParseError(p.lexer.lexdata, p[1], p.lexpos(1), p.lineno(1)) from e 

423 p[0] = TimeLiteral(value) 

424 

425 def p_literal_range(self, p): 

426 """literal : RANGE_LITERAL""" 

427 # RANGE_LITERAL value is tuple of three numbers 

428 start, stop, stride = p[1] 

429 p[0] = RangeLiteral(start, stop, stride) 

430 

431 def p_function_call(self, p): 

432 """function_call : SIMPLE_IDENTIFIER LPAREN expr_list RPAREN""" 

433 p[0] = function_call(p[1], p[3]) 

434 

435 def p_expr_list(self, p): 

436 """expr_list : expr_list COMMA expr 

437 | expr 

438 | empty 

439 """ 

440 if len(p) == 2: 

441 if p[1] is None: 

442 p[0] = [] 

443 else: 

444 p[0] = [p[1]] 

445 else: 

446 p[0] = p[1] + [p[3]] 

447 

448 # ---------- end of all grammar rules ---------- 

449 

450 # Error rule for syntax errors 

451 def p_error(self, p): 

452 if p is None: 

453 raise ParserEOFError() 

454 else: 

455 raise ParseError(p.lexer.lexdata, p.value, p.lexpos, p.lineno)