/usr/share/pyshared/sqlparse/engine/grouping.py is in python-sqlparse 0.1.10-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 | # -*- coding: utf-8 -*-
import itertools
from sqlparse import sql
from sqlparse import tokens as T
try:
next
except NameError: # Python < 2.6
next = lambda i: i.next()
def _group_left_right(tlist, ttype, value, cls,
check_right=lambda t: True,
check_left=lambda t: True,
include_semicolon=False):
[_group_left_right(sgroup, ttype, value, cls, check_right, check_left,
include_semicolon) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, cls)]
idx = 0
token = tlist.token_next_match(idx, ttype, value)
while token:
right = tlist.token_next(tlist.token_index(token))
left = tlist.token_prev(tlist.token_index(token))
if right is None or not check_right(right):
token = tlist.token_next_match(tlist.token_index(token) + 1,
ttype, value)
elif left is None or not check_left(left):
token = tlist.token_next_match(tlist.token_index(token) + 1,
ttype, value)
else:
if include_semicolon:
sright = tlist.token_next_match(tlist.token_index(right),
T.Punctuation, ';')
if sright is not None:
# only overwrite "right" if a semicolon is actually
# present.
right = sright
tokens = tlist.tokens_between(left, right)[1:]
if not isinstance(left, cls):
new = cls([left])
new_idx = tlist.token_index(left)
tlist.tokens.remove(left)
tlist.tokens.insert(new_idx, new)
left = new
left.tokens.extend(tokens)
for t in tokens:
tlist.tokens.remove(t)
token = tlist.token_next_match(tlist.token_index(left) + 1,
ttype, value)
def _group_matching(tlist, start_ttype, start_value, end_ttype, end_value,
cls, include_semicolon=False, recurse=False):
def _find_matching(i, tl, stt, sva, ett, eva):
depth = 1
for n in xrange(i, len(tl.tokens)):
t = tl.tokens[n]
if t.match(stt, sva):
depth += 1
elif t.match(ett, eva):
depth -= 1
if depth == 1:
return t
return None
[_group_matching(sgroup, start_ttype, start_value, end_ttype, end_value,
cls, include_semicolon) for sgroup in tlist.get_sublists()
if recurse]
if isinstance(tlist, cls):
idx = 1
else:
idx = 0
token = tlist.token_next_match(idx, start_ttype, start_value)
while token:
tidx = tlist.token_index(token)
end = _find_matching(tidx, tlist, start_ttype, start_value,
end_ttype, end_value)
if end is None:
idx = tidx + 1
else:
if include_semicolon:
next_ = tlist.token_next(tlist.token_index(end))
if next_ and next_.match(T.Punctuation, ';'):
end = next_
group = tlist.group_tokens(cls, tlist.tokens_between(token, end))
_group_matching(group, start_ttype, start_value,
end_ttype, end_value, cls, include_semicolon)
idx = tlist.token_index(group) + 1
token = tlist.token_next_match(idx, start_ttype, start_value)
def group_if(tlist):
_group_matching(tlist, T.Keyword, 'IF', T.Keyword, 'END IF', sql.If, True)
def group_for(tlist):
_group_matching(tlist, T.Keyword, 'FOR', T.Keyword, 'END LOOP',
sql.For, True)
def group_as(tlist):
def _right_valid(token):
# Currently limited to DML/DDL. Maybe additional more non SQL reserved
# keywords should appear here (see issue8).
return not token.ttype in (T.DML, T.DDL)
def _left_valid(token):
return token.ttype is not T.Keyword
_group_left_right(tlist, T.Keyword, 'AS', sql.Identifier,
check_right=_right_valid,
check_left=_left_valid)
def group_assignment(tlist):
_group_left_right(tlist, T.Assignment, ':=', sql.Assignment,
include_semicolon=True)
def group_comparison(tlist):
def _parts_valid(token):
return (token.ttype in (T.String.Symbol, T.Name, T.Number,
T.Number.Integer, T.Literal,
T.Literal.Number.Integer)
or isinstance(token, (sql.Identifier, sql.Parenthesis))
or (token.ttype is T.Keyword
and token.value.upper() in ['NULL', ]))
_group_left_right(tlist, T.Operator.Comparison, None, sql.Comparison,
check_left=_parts_valid, check_right=_parts_valid)
def group_case(tlist):
_group_matching(tlist, T.Keyword, 'CASE', T.Keyword, 'END', sql.Case,
include_semicolon=True, recurse=True)
def group_identifier(tlist):
def _consume_cycle(tl, i):
# TODO: Usage of Wildcard token is ambivalent here.
x = itertools.cycle((
lambda y: (y.match(T.Punctuation, '.')
or y.ttype is T.Operator
or y.ttype is T.Wildcard),
lambda y: (y.ttype in (T.String.Symbol,
T.Name,
T.Wildcard,
T.Literal.Number.Integer,
T.Literal.Number.Float)
or isinstance(y, (sql.Parenthesis, sql.Function)))))
for t in tl.tokens[i:]:
# Don't take whitespaces into account.
if t.ttype is T.Whitespace:
yield t
continue
if next(x)(t):
yield t
else:
raise StopIteration
def _next_token(tl, i):
# chooses the next token. if two tokens are found then the
# first is returned.
t1 = tl.token_next_by_type(
i, (T.String.Symbol, T.Name, T.Literal.Number.Integer,
T.Literal.Number.Float))
t2 = tl.token_next_by_instance(i, (sql.Function, sql.Parenthesis))
if t1 and t2:
i1 = tl.token_index(t1)
i2 = tl.token_index(t2)
if i1 > i2:
return t2
else:
return t1
elif t1:
return t1
else:
return t2
# bottom up approach: group subgroups first
[group_identifier(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, sql.Identifier)]
# real processing
idx = 0
token = _next_token(tlist, idx)
while token:
identifier_tokens = [token] + list(
_consume_cycle(tlist,
tlist.token_index(token) + 1))
# remove trailing whitespace
if identifier_tokens and identifier_tokens[-1].ttype is T.Whitespace:
identifier_tokens = identifier_tokens[:-1]
if not (len(identifier_tokens) == 1
and (isinstance(identifier_tokens[0], (sql.Function, sql.Parenthesis))
or identifier_tokens[0].ttype in (T.Literal.Number.Integer,
T.Literal.Number.Float))):
group = tlist.group_tokens(sql.Identifier, identifier_tokens)
idx = tlist.token_index(group) + 1
else:
idx += 1
token = _next_token(tlist, idx)
def group_identifier_list(tlist):
[group_identifier_list(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, sql.IdentifierList)]
idx = 0
# Allowed list items
fend1_funcs = [lambda t: isinstance(t, (sql.Identifier, sql.Function,
sql.Case)),
lambda t: t.is_whitespace(),
lambda t: t.ttype == T.Name,
lambda t: t.ttype == T.Wildcard,
lambda t: t.match(T.Keyword, 'null'),
lambda t: t.match(T.Keyword, 'role'),
lambda t: t.ttype == T.Number.Integer,
lambda t: t.ttype == T.String.Single,
lambda t: t.ttype == T.Name.Placeholder,
lambda t: t.ttype == T.Keyword,
lambda t: isinstance(t, sql.Comparison),
lambda t: isinstance(t, sql.Comment),
]
tcomma = tlist.token_next_match(idx, T.Punctuation, ',')
start = None
while tcomma is not None:
before = tlist.token_prev(tcomma)
after = tlist.token_next(tcomma)
# Check if the tokens around tcomma belong to a list
bpassed = apassed = False
for func in fend1_funcs:
if before is not None and func(before):
bpassed = True
if after is not None and func(after):
apassed = True
if not bpassed or not apassed:
# Something's wrong here, skip ahead to next ","
start = None
tcomma = tlist.token_next_match(tlist.token_index(tcomma) + 1,
T.Punctuation, ',')
else:
if start is None:
start = before
next_ = tlist.token_next(after)
if next_ is None or not next_.match(T.Punctuation, ','):
# Reached the end of the list
tokens = tlist.tokens_between(start, after)
group = tlist.group_tokens(sql.IdentifierList, tokens)
start = None
tcomma = tlist.token_next_match(tlist.token_index(group) + 1,
T.Punctuation, ',')
else:
tcomma = next_
def group_parenthesis(tlist):
_group_matching(tlist, T.Punctuation, '(', T.Punctuation, ')',
sql.Parenthesis)
def group_comments(tlist):
[group_comments(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, sql.Comment)]
idx = 0
token = tlist.token_next_by_type(idx, T.Comment)
while token:
tidx = tlist.token_index(token)
end = tlist.token_not_matching(tidx + 1,
[lambda t: t.ttype in T.Comment,
lambda t: t.is_whitespace()])
if end is None:
idx = tidx + 1
else:
eidx = tlist.token_index(end)
grp_tokens = tlist.tokens_between(token,
tlist.token_prev(eidx, False))
group = tlist.group_tokens(sql.Comment, grp_tokens)
idx = tlist.token_index(group)
token = tlist.token_next_by_type(idx, T.Comment)
def group_where(tlist):
[group_where(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, sql.Where)]
idx = 0
token = tlist.token_next_match(idx, T.Keyword, 'WHERE')
stopwords = ('ORDER', 'GROUP', 'LIMIT', 'UNION')
while token:
tidx = tlist.token_index(token)
end = tlist.token_next_match(tidx + 1, T.Keyword, stopwords)
if end is None:
end = tlist._groupable_tokens[-1]
else:
end = tlist.tokens[tlist.token_index(end) - 1]
group = tlist.group_tokens(sql.Where,
tlist.tokens_between(token, end),
ignore_ws=True)
idx = tlist.token_index(group)
token = tlist.token_next_match(idx, T.Keyword, 'WHERE')
def group_aliased(tlist):
clss = (sql.Identifier, sql.Function, sql.Case)
[group_aliased(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, clss)]
idx = 0
token = tlist.token_next_by_instance(idx, clss)
while token:
next_ = tlist.token_next(tlist.token_index(token))
if next_ is not None and isinstance(next_, clss):
if not next_.value.upper().startswith('VARCHAR'):
grp = tlist.tokens_between(token, next_)[1:]
token.tokens.extend(grp)
for t in grp:
tlist.tokens.remove(t)
idx = tlist.token_index(token) + 1
token = tlist.token_next_by_instance(idx, clss)
def group_typecasts(tlist):
_group_left_right(tlist, T.Punctuation, '::', sql.Identifier)
def group_functions(tlist):
[group_functions(sgroup) for sgroup in tlist.get_sublists()
if not isinstance(sgroup, sql.Function)]
idx = 0
token = tlist.token_next_by_type(idx, T.Name)
while token:
next_ = tlist.token_next(token)
if not isinstance(next_, sql.Parenthesis):
idx = tlist.token_index(token) + 1
else:
func = tlist.group_tokens(sql.Function,
tlist.tokens_between(token, next_))
idx = tlist.token_index(func) + 1
token = tlist.token_next_by_type(idx, T.Name)
def group_order(tlist):
idx = 0
token = tlist.token_next_by_type(idx, T.Keyword.Order)
while token:
prev = tlist.token_prev(token)
if isinstance(prev, sql.Identifier):
ido = tlist.group_tokens(sql.Identifier,
tlist.tokens_between(prev, token))
idx = tlist.token_index(ido) + 1
else:
idx = tlist.token_index(token) + 1
token = tlist.token_next_by_type(idx, T.Keyword.Order)
def group(tlist):
for func in [
group_comments,
group_parenthesis,
group_functions,
group_where,
group_case,
group_identifier,
group_order,
group_typecasts,
group_as,
group_aliased,
group_assignment,
group_comparison,
group_identifier_list,
group_if,
group_for]:
func(tlist)
|