root / snf-pithos-backend / pithos / backends / lib / sqlalchemy / node.py @ 9f5c8386
History | View | Annotate | Download (45.4 kB)
1 |
# Copyright 2011-2012 GRNET S.A. All rights reserved.
|
---|---|
2 |
#
|
3 |
# Redistribution and use in source and binary forms, with or
|
4 |
# without modification, are permitted provided that the following
|
5 |
# conditions are met:
|
6 |
#
|
7 |
# 1. Redistributions of source code must retain the above
|
8 |
# copyright notice, this list of conditions and the following
|
9 |
# disclaimer.
|
10 |
#
|
11 |
# 2. Redistributions in binary form must reproduce the above
|
12 |
# copyright notice, this list of conditions and the following
|
13 |
# disclaimer in the documentation and/or other materials
|
14 |
# provided with the distribution.
|
15 |
#
|
16 |
# THIS SOFTWARE IS PROVIDED BY GRNET S.A. ``AS IS'' AND ANY EXPRESS
|
17 |
# OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
|
18 |
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
19 |
# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL GRNET S.A OR
|
20 |
# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
21 |
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
22 |
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
|
23 |
# USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
|
24 |
# AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
25 |
# LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
|
26 |
# ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
27 |
# POSSIBILITY OF SUCH DAMAGE.
|
28 |
#
|
29 |
# The views and conclusions contained in the software and
|
30 |
# documentation are those of the authors and should not be
|
31 |
# interpreted as representing official policies, either expressed
|
32 |
# or implied, of GRNET S.A.
|
33 |
|
34 |
from time import time |
35 |
from operator import itemgetter |
36 |
from itertools import groupby |
37 |
|
38 |
from sqlalchemy import (Table, Integer, BigInteger, DECIMAL, Boolean, |
39 |
Column, String, MetaData, ForeignKey) |
40 |
from sqlalchemy.types import Text |
41 |
from sqlalchemy.schema import Index, Sequence |
42 |
from sqlalchemy.sql import func, and_, or_, not_, null, select, bindparam, text, exists |
43 |
from sqlalchemy.ext.compiler import compiles |
44 |
from sqlalchemy.engine.reflection import Inspector |
45 |
from sqlalchemy.exc import NoSuchTableError |
46 |
|
47 |
from dbworker import DBWorker, ESCAPE_CHAR |
48 |
|
49 |
from pithos.backends.filter import parse_filters |
50 |
|
51 |
|
52 |
ROOTNODE = 0
|
53 |
|
54 |
(SERIAL, NODE, HASH, SIZE, TYPE, SOURCE, MTIME, MUSER, UUID, CHECKSUM, |
55 |
CLUSTER) = range(11) |
56 |
|
57 |
(MATCH_PREFIX, MATCH_EXACT) = range(2) |
58 |
|
59 |
inf = float('inf') |
60 |
|
61 |
|
62 |
def strnextling(prefix): |
63 |
"""Return the first unicode string
|
64 |
greater than but not starting with given prefix.
|
65 |
strnextling('hello') -> 'hellp'
|
66 |
"""
|
67 |
if not prefix: |
68 |
## all strings start with the null string,
|
69 |
## therefore we have to approximate strnextling('')
|
70 |
## with the last unicode character supported by python
|
71 |
## 0x10ffff for wide (32-bit unicode) python builds
|
72 |
## 0x00ffff for narrow (16-bit unicode) python builds
|
73 |
## We will not autodetect. 0xffff is safe enough.
|
74 |
return unichr(0xffff) |
75 |
s = prefix[:-1]
|
76 |
c = ord(prefix[-1]) |
77 |
if c >= 0xffff: |
78 |
raise RuntimeError |
79 |
s += unichr(c + 1) |
80 |
return s
|
81 |
|
82 |
|
83 |
def strprevling(prefix): |
84 |
"""Return an approximation of the last unicode string
|
85 |
less than but not starting with given prefix.
|
86 |
strprevling(u'hello') -> u'helln\\xffff'
|
87 |
"""
|
88 |
if not prefix: |
89 |
## There is no prevling for the null string
|
90 |
return prefix
|
91 |
s = prefix[:-1]
|
92 |
c = ord(prefix[-1]) |
93 |
if c > 0: |
94 |
s += unichr(c - 1) + unichr(0xffff) |
95 |
return s
|
96 |
|
97 |
_propnames = { |
98 |
'serial': 0, |
99 |
'node': 1, |
100 |
'hash': 2, |
101 |
'size': 3, |
102 |
'type': 4, |
103 |
'source': 5, |
104 |
'mtime': 6, |
105 |
'muser': 7, |
106 |
'uuid': 8, |
107 |
'checksum': 9, |
108 |
'cluster': 10 |
109 |
} |
110 |
|
111 |
|
112 |
def create_tables(engine): |
113 |
metadata = MetaData() |
114 |
|
115 |
#create nodes table
|
116 |
columns = [] |
117 |
columns.append(Column('node', Integer, primary_key=True)) |
118 |
columns.append(Column('parent', Integer,
|
119 |
ForeignKey('nodes.node',
|
120 |
ondelete='CASCADE',
|
121 |
onupdate='CASCADE'),
|
122 |
autoincrement=False))
|
123 |
columns.append(Column('latest_version', Integer))
|
124 |
columns.append(Column('path', String(2048), default='', nullable=False)) |
125 |
nodes = Table('nodes', metadata, *columns, mysql_engine='InnoDB') |
126 |
Index('idx_nodes_path', nodes.c.path, unique=True) |
127 |
Index('idx_nodes_parent', nodes.c.parent)
|
128 |
|
129 |
#create policy table
|
130 |
columns = [] |
131 |
columns.append(Column('node', Integer,
|
132 |
ForeignKey('nodes.node',
|
133 |
ondelete='CASCADE',
|
134 |
onupdate='CASCADE'),
|
135 |
primary_key=True))
|
136 |
columns.append(Column('key', String(128), primary_key=True)) |
137 |
columns.append(Column('value', String(256))) |
138 |
policy = Table('policy', metadata, *columns, mysql_engine='InnoDB') |
139 |
|
140 |
#create statistics table
|
141 |
columns = [] |
142 |
columns.append(Column('node', Integer,
|
143 |
ForeignKey('nodes.node',
|
144 |
ondelete='CASCADE',
|
145 |
onupdate='CASCADE'),
|
146 |
primary_key=True))
|
147 |
columns.append(Column('population', Integer, nullable=False, default=0)) |
148 |
columns.append(Column('size', BigInteger, nullable=False, default=0)) |
149 |
columns.append(Column('mtime', DECIMAL(precision=16, scale=6))) |
150 |
columns.append(Column('cluster', Integer, nullable=False, default=0, |
151 |
primary_key=True, autoincrement=False)) |
152 |
statistics = Table('statistics', metadata, *columns, mysql_engine='InnoDB') |
153 |
|
154 |
#create versions table
|
155 |
columns = [] |
156 |
columns.append(Column('serial', Integer, primary_key=True)) |
157 |
columns.append(Column('node', Integer,
|
158 |
ForeignKey('nodes.node',
|
159 |
ondelete='CASCADE',
|
160 |
onupdate='CASCADE')))
|
161 |
columns.append(Column('hash', String(256))) |
162 |
columns.append(Column('size', BigInteger, nullable=False, default=0)) |
163 |
columns.append(Column('type', String(256), nullable=False, default='')) |
164 |
columns.append(Column('source', Integer))
|
165 |
columns.append(Column('mtime', DECIMAL(precision=16, scale=6))) |
166 |
columns.append(Column('muser', String(256), nullable=False, default='')) |
167 |
columns.append(Column('uuid', String(64), nullable=False, default='')) |
168 |
columns.append(Column('checksum', String(256), nullable=False, default='')) |
169 |
columns.append(Column('cluster', Integer, nullable=False, default=0)) |
170 |
versions = Table('versions', metadata, *columns, mysql_engine='InnoDB') |
171 |
Index('idx_versions_node_mtime', versions.c.node, versions.c.mtime)
|
172 |
Index('idx_versions_node_uuid', versions.c.uuid)
|
173 |
|
174 |
#create attributes table
|
175 |
columns = [] |
176 |
columns.append(Column('serial', Integer,
|
177 |
ForeignKey('versions.serial',
|
178 |
ondelete='CASCADE',
|
179 |
onupdate='CASCADE'),
|
180 |
primary_key=True))
|
181 |
columns.append(Column('domain', String(256), primary_key=True)) |
182 |
columns.append(Column('key', String(128), primary_key=True)) |
183 |
columns.append(Column('value', String(256))) |
184 |
columns.append(Column('node', Integer, nullable=False, default=0)) |
185 |
columns.append(Column('is_latest', Boolean, nullable=False, default=True)) |
186 |
attributes = Table('attributes', metadata, *columns, mysql_engine='InnoDB') |
187 |
Index('idx_attributes_domain', attributes.c.domain)
|
188 |
Index('idx_attributes_serial_node', attributes.c.serial, attributes.c.node)
|
189 |
|
190 |
metadata.create_all(engine) |
191 |
return metadata.sorted_tables
|
192 |
|
193 |
|
194 |
class Node(DBWorker): |
195 |
"""Nodes store path organization and have multiple versions.
|
196 |
Versions store object history and have multiple attributes.
|
197 |
Attributes store metadata.
|
198 |
"""
|
199 |
|
200 |
# TODO: Provide an interface for included and excluded clusters.
|
201 |
|
202 |
def __init__(self, **params): |
203 |
DBWorker.__init__(self, **params)
|
204 |
try:
|
205 |
metadata = MetaData(self.engine)
|
206 |
self.nodes = Table('nodes', metadata, autoload=True) |
207 |
self.policy = Table('policy', metadata, autoload=True) |
208 |
self.statistics = Table('statistics', metadata, autoload=True) |
209 |
self.versions = Table('versions', metadata, autoload=True) |
210 |
self.attributes = Table('attributes', metadata, autoload=True) |
211 |
except NoSuchTableError:
|
212 |
tables = create_tables(self.engine)
|
213 |
map(lambda t: self.__setattr__(t.name, t), tables) |
214 |
|
215 |
s = self.nodes.select().where(and_(self.nodes.c.node == ROOTNODE, |
216 |
self.nodes.c.parent == ROOTNODE))
|
217 |
wrapper = self.wrapper
|
218 |
wrapper.execute() |
219 |
try:
|
220 |
rp = self.conn.execute(s)
|
221 |
r = rp.fetchone() |
222 |
rp.close() |
223 |
if not r: |
224 |
s = self.nodes.insert(
|
225 |
).values(node=ROOTNODE, parent=ROOTNODE, path='')
|
226 |
self.conn.execute(s)
|
227 |
finally:
|
228 |
wrapper.commit() |
229 |
|
230 |
def node_create(self, parent, path): |
231 |
"""Create a new node from the given properties.
|
232 |
Return the node identifier of the new node.
|
233 |
"""
|
234 |
#TODO catch IntegrityError?
|
235 |
s = self.nodes.insert().values(parent=parent, path=path)
|
236 |
r = self.conn.execute(s)
|
237 |
inserted_primary_key = r.inserted_primary_key[0]
|
238 |
r.close() |
239 |
return inserted_primary_key
|
240 |
|
241 |
def node_lookup(self, path, for_update=False): |
242 |
"""Lookup the current node of the given path.
|
243 |
Return None if the path is not found.
|
244 |
"""
|
245 |
|
246 |
# Use LIKE for comparison to avoid MySQL problems with trailing spaces.
|
247 |
s = select([self.nodes.c.node], self.nodes.c.path.like( |
248 |
self.escape_like(path), escape=ESCAPE_CHAR), for_update=for_update)
|
249 |
r = self.conn.execute(s)
|
250 |
row = r.fetchone() |
251 |
r.close() |
252 |
if row:
|
253 |
return row[0] |
254 |
return None |
255 |
|
256 |
def node_lookup_bulk(self, paths): |
257 |
"""Lookup the current nodes for the given paths.
|
258 |
Return () if the path is not found.
|
259 |
"""
|
260 |
|
261 |
if not paths: |
262 |
return ()
|
263 |
# Use LIKE for comparison to avoid MySQL problems with trailing spaces.
|
264 |
s = select([self.nodes.c.node], self.nodes.c.path.in_(paths)) |
265 |
r = self.conn.execute(s)
|
266 |
rows = r.fetchall() |
267 |
r.close() |
268 |
return [row[0] for row in rows] |
269 |
|
270 |
def node_get_properties(self, node): |
271 |
"""Return the node's (parent, path).
|
272 |
Return None if the node is not found.
|
273 |
"""
|
274 |
|
275 |
s = select([self.nodes.c.parent, self.nodes.c.path]) |
276 |
s = s.where(self.nodes.c.node == node)
|
277 |
r = self.conn.execute(s)
|
278 |
l = r.fetchone() |
279 |
r.close() |
280 |
return l
|
281 |
|
282 |
def node_get_versions(self, node, keys=(), propnames=_propnames): |
283 |
"""Return the properties of all versions at node.
|
284 |
If keys is empty, return all properties in the order
|
285 |
(serial, node, hash, size, type, source, mtime, muser, uuid, checksum, cluster).
|
286 |
"""
|
287 |
|
288 |
s = select([self.versions.c.serial,
|
289 |
self.versions.c.node,
|
290 |
self.versions.c.hash,
|
291 |
self.versions.c.size,
|
292 |
self.versions.c.type,
|
293 |
self.versions.c.source,
|
294 |
self.versions.c.mtime,
|
295 |
self.versions.c.muser,
|
296 |
self.versions.c.uuid,
|
297 |
self.versions.c.checksum,
|
298 |
self.versions.c.cluster], self.versions.c.node == node) |
299 |
s = s.order_by(self.versions.c.serial)
|
300 |
r = self.conn.execute(s)
|
301 |
rows = r.fetchall() |
302 |
r.close() |
303 |
if not rows: |
304 |
return rows
|
305 |
|
306 |
if not keys: |
307 |
return rows
|
308 |
|
309 |
return [[p[propnames[k]] for k in keys if k in propnames] for p in rows] |
310 |
|
311 |
def node_count_children(self, node): |
312 |
"""Return node's child count."""
|
313 |
|
314 |
s = select([func.count(self.nodes.c.node)])
|
315 |
s = s.where(and_(self.nodes.c.parent == node,
|
316 |
self.nodes.c.node != ROOTNODE))
|
317 |
r = self.conn.execute(s)
|
318 |
row = r.fetchone() |
319 |
r.close() |
320 |
return row[0] |
321 |
|
322 |
def node_purge_children(self, parent, before=inf, cluster=0, |
323 |
update_statistics_ancestors_depth=None):
|
324 |
"""Delete all versions with the specified
|
325 |
parent and cluster, and return
|
326 |
the hashes, the total size and the serials of versions deleted.
|
327 |
Clears out nodes with no remaining versions.
|
328 |
"""
|
329 |
#update statistics
|
330 |
c1 = select([self.nodes.c.node],
|
331 |
self.nodes.c.parent == parent)
|
332 |
where_clause = and_(self.versions.c.node.in_(c1),
|
333 |
self.versions.c.cluster == cluster)
|
334 |
if before != inf:
|
335 |
where_clause = and_(where_clause, |
336 |
self.versions.c.mtime <= before)
|
337 |
s = select([func.count(self.versions.c.serial),
|
338 |
func.sum(self.versions.c.size)])
|
339 |
s = s.where(where_clause) |
340 |
r = self.conn.execute(s)
|
341 |
row = r.fetchone() |
342 |
r.close() |
343 |
if not row: |
344 |
return (), 0, () |
345 |
nr, size = row[0], row[1] if row[1] else 0 |
346 |
mtime = time() |
347 |
self.statistics_update(parent, -nr, -size, mtime, cluster)
|
348 |
self.statistics_update_ancestors(parent, -nr, -size, mtime, cluster,
|
349 |
update_statistics_ancestors_depth) |
350 |
|
351 |
s = select([self.versions.c.hash, self.versions.c.serial]) |
352 |
s = s.where(where_clause) |
353 |
r = self.conn.execute(s)
|
354 |
hashes = [] |
355 |
serials = [] |
356 |
for row in r.fetchall(): |
357 |
hashes += [row[0]]
|
358 |
serials += [row[1]]
|
359 |
r.close() |
360 |
|
361 |
#delete versions
|
362 |
s = self.versions.delete().where(where_clause)
|
363 |
r = self.conn.execute(s)
|
364 |
r.close() |
365 |
|
366 |
#delete nodes
|
367 |
s = select([self.nodes.c.node],
|
368 |
and_(self.nodes.c.parent == parent,
|
369 |
select([func.count(self.versions.c.serial)],
|
370 |
self.versions.c.node == self.nodes.c.node).as_scalar() == 0)) |
371 |
rp = self.conn.execute(s)
|
372 |
nodes = [r[0] for r in rp.fetchall()] |
373 |
rp.close() |
374 |
if nodes:
|
375 |
s = self.nodes.delete().where(self.nodes.c.node.in_(nodes)) |
376 |
self.conn.execute(s).close()
|
377 |
|
378 |
return hashes, size, serials
|
379 |
|
380 |
def node_purge(self, node, before=inf, cluster=0, |
381 |
update_statistics_ancestors_depth=None):
|
382 |
"""Delete all versions with the specified
|
383 |
node and cluster, and return
|
384 |
the hashes and size of versions deleted.
|
385 |
Clears out the node if it has no remaining versions.
|
386 |
"""
|
387 |
|
388 |
#update statistics
|
389 |
s = select([func.count(self.versions.c.serial),
|
390 |
func.sum(self.versions.c.size)])
|
391 |
where_clause = and_(self.versions.c.node == node,
|
392 |
self.versions.c.cluster == cluster)
|
393 |
if before != inf:
|
394 |
where_clause = and_(where_clause, |
395 |
self.versions.c.mtime <= before)
|
396 |
s = s.where(where_clause) |
397 |
r = self.conn.execute(s)
|
398 |
row = r.fetchone() |
399 |
nr, size = row[0], row[1] |
400 |
r.close() |
401 |
if not nr: |
402 |
return (), 0, () |
403 |
mtime = time() |
404 |
self.statistics_update_ancestors(node, -nr, -size, mtime, cluster,
|
405 |
update_statistics_ancestors_depth) |
406 |
|
407 |
s = select([self.versions.c.hash, self.versions.c.serial]) |
408 |
s = s.where(where_clause) |
409 |
r = self.conn.execute(s)
|
410 |
hashes = [] |
411 |
serials = [] |
412 |
for row in r.fetchall(): |
413 |
hashes += [row[0]]
|
414 |
serials += [row[1]]
|
415 |
r.close() |
416 |
|
417 |
#delete versions
|
418 |
s = self.versions.delete().where(where_clause)
|
419 |
r = self.conn.execute(s)
|
420 |
r.close() |
421 |
|
422 |
#delete nodes
|
423 |
s = select([self.nodes.c.node],
|
424 |
and_(self.nodes.c.node == node,
|
425 |
select([func.count(self.versions.c.serial)],
|
426 |
self.versions.c.node == self.nodes.c.node).as_scalar() == 0)) |
427 |
rp= self.conn.execute(s)
|
428 |
nodes = [r[0] for r in rp.fetchall()] |
429 |
rp.close() |
430 |
if nodes:
|
431 |
s = self.nodes.delete().where(self.nodes.c.node.in_(nodes)) |
432 |
self.conn.execute(s).close()
|
433 |
|
434 |
return hashes, size, serials
|
435 |
|
436 |
def node_remove(self, node, update_statistics_ancestors_depth=None): |
437 |
"""Remove the node specified.
|
438 |
Return false if the node has children or is not found.
|
439 |
"""
|
440 |
|
441 |
if self.node_count_children(node): |
442 |
return False |
443 |
|
444 |
mtime = time() |
445 |
s = select([func.count(self.versions.c.serial),
|
446 |
func.sum(self.versions.c.size),
|
447 |
self.versions.c.cluster])
|
448 |
s = s.where(self.versions.c.node == node)
|
449 |
s = s.group_by(self.versions.c.cluster)
|
450 |
r = self.conn.execute(s)
|
451 |
for population, size, cluster in r.fetchall(): |
452 |
self.statistics_update_ancestors(
|
453 |
node, -population, -size, mtime, cluster, |
454 |
update_statistics_ancestors_depth) |
455 |
r.close() |
456 |
|
457 |
s = self.nodes.delete().where(self.nodes.c.node == node) |
458 |
self.conn.execute(s).close()
|
459 |
return True |
460 |
|
461 |
def node_accounts(self, accounts=()): |
462 |
s = select([self.nodes.c.path, self.nodes.c.node]) |
463 |
s = s.where(and_(self.nodes.c.node != 0, |
464 |
self.nodes.c.parent == 0)) |
465 |
if accounts:
|
466 |
s = s.where(self.nodes.c.path.in_(accounts))
|
467 |
r = self.conn.execute(s)
|
468 |
rows = r.fetchall() |
469 |
r.close() |
470 |
return rows
|
471 |
|
472 |
def node_account_quotas(self): |
473 |
s = select([self.nodes.c.path, self.policy.c.value]) |
474 |
s = s.where(and_(self.nodes.c.node != 0, |
475 |
self.nodes.c.parent == 0)) |
476 |
s = s.where(self.nodes.c.node == self.policy.c.node) |
477 |
s = s.where(self.policy.c.key == 'quota') |
478 |
r = self.conn.execute(s)
|
479 |
rows = r.fetchall() |
480 |
r.close() |
481 |
return dict(rows) |
482 |
|
483 |
def node_account_usage(self, account_node, cluster): |
484 |
select_children = select( |
485 |
[self.nodes.c.node]).where(self.nodes.c.parent == account_node) |
486 |
select_descendants = select([self.nodes.c.node]).where(
|
487 |
or_(self.nodes.c.parent.in_(select_children),
|
488 |
self.nodes.c.node.in_(select_children)))
|
489 |
s = select([func.sum(self.versions.c.size)])
|
490 |
s = s.where(self.nodes.c.node == self.versions.c.node) |
491 |
s = s.where(self.nodes.c.node.in_(select_descendants))
|
492 |
s = s.where(self.versions.c.cluster == cluster)
|
493 |
r = self.conn.execute(s)
|
494 |
usage = r.fetchone()[0]
|
495 |
r.close() |
496 |
return usage
|
497 |
|
498 |
def policy_get(self, node): |
499 |
s = select([self.policy.c.key, self.policy.c.value], |
500 |
self.policy.c.node == node)
|
501 |
r = self.conn.execute(s)
|
502 |
d = dict(r.fetchall())
|
503 |
r.close() |
504 |
return d
|
505 |
|
506 |
def policy_set(self, node, policy): |
507 |
#insert or replace
|
508 |
for k, v in policy.iteritems(): |
509 |
s = self.policy.update().where(and_(self.policy.c.node == node, |
510 |
self.policy.c.key == k))
|
511 |
s = s.values(value=v) |
512 |
rp = self.conn.execute(s)
|
513 |
rp.close() |
514 |
if rp.rowcount == 0: |
515 |
s = self.policy.insert()
|
516 |
values = {'node': node, 'key': k, 'value': v} |
517 |
r = self.conn.execute(s, values)
|
518 |
r.close() |
519 |
|
520 |
def statistics_get(self, node, cluster=0): |
521 |
"""Return population, total size and last mtime
|
522 |
for all versions under node that belong to the cluster.
|
523 |
"""
|
524 |
|
525 |
s = select([self.statistics.c.population,
|
526 |
self.statistics.c.size,
|
527 |
self.statistics.c.mtime])
|
528 |
s = s.where(and_(self.statistics.c.node == node,
|
529 |
self.statistics.c.cluster == cluster))
|
530 |
r = self.conn.execute(s)
|
531 |
row = r.fetchone() |
532 |
r.close() |
533 |
return row
|
534 |
|
535 |
def statistics_update(self, node, population, size, mtime, cluster=0): |
536 |
"""Update the statistics of the given node.
|
537 |
Statistics keep track the population, total
|
538 |
size of objects and mtime in the node's namespace.
|
539 |
May be zero or positive or negative numbers.
|
540 |
"""
|
541 |
s = select([self.statistics.c.population, self.statistics.c.size], |
542 |
and_(self.statistics.c.node == node,
|
543 |
self.statistics.c.cluster == cluster))
|
544 |
rp = self.conn.execute(s)
|
545 |
r = rp.fetchone() |
546 |
rp.close() |
547 |
if not r: |
548 |
prepopulation, presize = (0, 0) |
549 |
else:
|
550 |
prepopulation, presize = r |
551 |
population += prepopulation |
552 |
population = max(population, 0) |
553 |
size += presize |
554 |
|
555 |
#insert or replace
|
556 |
#TODO better upsert
|
557 |
u = self.statistics.update().where(and_(self.statistics.c.node == node, |
558 |
self.statistics.c.cluster == cluster))
|
559 |
u = u.values(population=population, size=size, mtime=mtime) |
560 |
rp = self.conn.execute(u)
|
561 |
rp.close() |
562 |
if rp.rowcount == 0: |
563 |
ins = self.statistics.insert()
|
564 |
ins = ins.values(node=node, population=population, size=size, |
565 |
mtime=mtime, cluster=cluster) |
566 |
self.conn.execute(ins).close()
|
567 |
|
568 |
def statistics_update_ancestors(self, node, population, size, mtime, |
569 |
cluster=0, recursion_depth=None): |
570 |
"""Update the statistics of the given node's parent.
|
571 |
Then recursively update all parents up to the root
|
572 |
or up to the ``recursion_depth`` (if not None).
|
573 |
Population is not recursive.
|
574 |
"""
|
575 |
|
576 |
i = 0
|
577 |
while True: |
578 |
if node == ROOTNODE:
|
579 |
break
|
580 |
if recursion_depth and recursion_depth <= i: |
581 |
break
|
582 |
props = self.node_get_properties(node)
|
583 |
if props is None: |
584 |
break
|
585 |
parent, path = props |
586 |
self.statistics_update(parent, population, size, mtime, cluster)
|
587 |
node = parent |
588 |
population = 0 # Population isn't recursive |
589 |
i += 1
|
590 |
|
591 |
def statistics_latest(self, node, before=inf, except_cluster=0): |
592 |
"""Return population, total size and last mtime
|
593 |
for all latest versions under node that
|
594 |
do not belong to the cluster.
|
595 |
"""
|
596 |
|
597 |
# The node.
|
598 |
props = self.node_get_properties(node)
|
599 |
if props is None: |
600 |
return None |
601 |
parent, path = props |
602 |
|
603 |
# The latest version.
|
604 |
s = select([self.versions.c.serial,
|
605 |
self.versions.c.node,
|
606 |
self.versions.c.hash,
|
607 |
self.versions.c.size,
|
608 |
self.versions.c.type,
|
609 |
self.versions.c.source,
|
610 |
self.versions.c.mtime,
|
611 |
self.versions.c.muser,
|
612 |
self.versions.c.uuid,
|
613 |
self.versions.c.checksum,
|
614 |
self.versions.c.cluster])
|
615 |
if before != inf:
|
616 |
filtered = select([func.max(self.versions.c.serial)],
|
617 |
self.versions.c.node == node)
|
618 |
filtered = filtered.where(self.versions.c.mtime < before)
|
619 |
else:
|
620 |
filtered = select([self.nodes.c.latest_version],
|
621 |
self.nodes.c.node == node)
|
622 |
s = s.where(and_(self.versions.c.cluster != except_cluster,
|
623 |
self.versions.c.serial == filtered))
|
624 |
r = self.conn.execute(s)
|
625 |
props = r.fetchone() |
626 |
r.close() |
627 |
if not props: |
628 |
return None |
629 |
mtime = props[MTIME] |
630 |
|
631 |
# First level, just under node (get population).
|
632 |
v = self.versions.alias('v') |
633 |
s = select([func.count(v.c.serial), |
634 |
func.sum(v.c.size), |
635 |
func.max(v.c.mtime)]) |
636 |
if before != inf:
|
637 |
c1 = select([func.max(self.versions.c.serial)])
|
638 |
c1 = c1.where(self.versions.c.mtime < before)
|
639 |
c1.where(self.versions.c.node == v.c.node)
|
640 |
else:
|
641 |
c1 = select([self.nodes.c.latest_version])
|
642 |
c1 = c1.where(self.nodes.c.node == v.c.node)
|
643 |
c2 = select([self.nodes.c.node], self.nodes.c.parent == node) |
644 |
s = s.where(and_(v.c.serial == c1, |
645 |
v.c.cluster != except_cluster, |
646 |
v.c.node.in_(c2))) |
647 |
rp = self.conn.execute(s)
|
648 |
r = rp.fetchone() |
649 |
rp.close() |
650 |
if not r: |
651 |
return None |
652 |
count = r[0]
|
653 |
mtime = max(mtime, r[2]) |
654 |
if count == 0: |
655 |
return (0, 0, mtime) |
656 |
|
657 |
# All children (get size and mtime).
|
658 |
# This is why the full path is stored.
|
659 |
s = select([func.count(v.c.serial), |
660 |
func.sum(v.c.size), |
661 |
func.max(v.c.mtime)]) |
662 |
if before != inf:
|
663 |
c1 = select([func.max(self.versions.c.serial)],
|
664 |
self.versions.c.node == v.c.node)
|
665 |
c1 = c1.where(self.versions.c.mtime < before)
|
666 |
else:
|
667 |
c1 = select([self.nodes.c.latest_version],
|
668 |
self.nodes.c.node == v.c.node)
|
669 |
c2 = select([self.nodes.c.node], self.nodes.c.path.like( |
670 |
self.escape_like(path) + '%', escape=ESCAPE_CHAR)) |
671 |
s = s.where(and_(v.c.serial == c1, |
672 |
v.c.cluster != except_cluster, |
673 |
v.c.node.in_(c2))) |
674 |
rp = self.conn.execute(s)
|
675 |
r = rp.fetchone() |
676 |
rp.close() |
677 |
if not r: |
678 |
return None |
679 |
size = r[1] - props[SIZE]
|
680 |
mtime = max(mtime, r[2]) |
681 |
return (count, size, mtime)
|
682 |
|
683 |
def nodes_set_latest_version(self, node, serial): |
684 |
s = self.nodes.update().where(self.nodes.c.node == node) |
685 |
s = s.values(latest_version=serial) |
686 |
self.conn.execute(s).close()
|
687 |
|
688 |
def version_create(self, node, hash, size, type, source, muser, uuid, |
689 |
checksum, cluster=0,
|
690 |
update_statistics_ancestors_depth=None):
|
691 |
"""Create a new version from the given properties.
|
692 |
Return the (serial, mtime) of the new version.
|
693 |
"""
|
694 |
|
695 |
mtime = time() |
696 |
s = self.versions.insert(
|
697 |
).values(node=node, hash=hash, size=size, type=type, source=source, |
698 |
mtime=mtime, muser=muser, uuid=uuid, checksum=checksum, cluster=cluster) |
699 |
serial = self.conn.execute(s).inserted_primary_key[0] |
700 |
self.statistics_update_ancestors(node, 1, size, mtime, cluster, |
701 |
update_statistics_ancestors_depth) |
702 |
|
703 |
self.nodes_set_latest_version(node, serial)
|
704 |
|
705 |
return serial, mtime
|
706 |
|
707 |
def version_lookup(self, node, before=inf, cluster=0, all_props=True): |
708 |
"""Lookup the current version of the given node.
|
709 |
Return a list with its properties:
|
710 |
(serial, node, hash, size, type, source, mtime,
|
711 |
muser, uuid, checksum, cluster)
|
712 |
or None if the current version is not found in the given cluster.
|
713 |
"""
|
714 |
|
715 |
v = self.versions.alias('v') |
716 |
if not all_props: |
717 |
s = select([v.c.serial]) |
718 |
else:
|
719 |
s = select([v.c.serial, v.c.node, v.c.hash, |
720 |
v.c.size, v.c.type, v.c.source, |
721 |
v.c.mtime, v.c.muser, v.c.uuid, |
722 |
v.c.checksum, v.c.cluster]) |
723 |
if before != inf:
|
724 |
c = select([func.max(self.versions.c.serial)],
|
725 |
self.versions.c.node == node)
|
726 |
c = c.where(self.versions.c.mtime < before)
|
727 |
else:
|
728 |
c = select([self.nodes.c.latest_version],
|
729 |
self.nodes.c.node == node)
|
730 |
s = s.where(and_(v.c.serial == c, |
731 |
v.c.cluster == cluster)) |
732 |
r = self.conn.execute(s)
|
733 |
props = r.fetchone() |
734 |
r.close() |
735 |
if props:
|
736 |
return props
|
737 |
return None |
738 |
|
739 |
def version_lookup_bulk(self, nodes, before=inf, cluster=0, all_props=True): |
740 |
"""Lookup the current versions of the given nodes.
|
741 |
Return a list with their properties:
|
742 |
(serial, node, hash, size, type, source, mtime, muser, uuid, checksum, cluster).
|
743 |
"""
|
744 |
if not nodes: |
745 |
return ()
|
746 |
v = self.versions.alias('v') |
747 |
if not all_props: |
748 |
s = select([v.c.serial]) |
749 |
else:
|
750 |
s = select([v.c.serial, v.c.node, v.c.hash, |
751 |
v.c.size, v.c.type, v.c.source, |
752 |
v.c.mtime, v.c.muser, v.c.uuid, |
753 |
v.c.checksum, v.c.cluster]) |
754 |
if before != inf:
|
755 |
c = select([func.max(self.versions.c.serial)],
|
756 |
self.versions.c.node.in_(nodes))
|
757 |
c = c.where(self.versions.c.mtime < before)
|
758 |
c = c.group_by(self.versions.c.node)
|
759 |
else:
|
760 |
c = select([self.nodes.c.latest_version],
|
761 |
self.nodes.c.node.in_(nodes))
|
762 |
s = s.where(and_(v.c.serial.in_(c), |
763 |
v.c.cluster == cluster)) |
764 |
s = s.order_by(v.c.node) |
765 |
r = self.conn.execute(s)
|
766 |
rproxy = r.fetchall() |
767 |
r.close() |
768 |
return (tuple(row.values()) for row in rproxy) |
769 |
|
770 |
def version_get_properties(self, serial, keys=(), propnames=_propnames): |
771 |
"""Return a sequence of values for the properties of
|
772 |
the version specified by serial and the keys, in the order given.
|
773 |
If keys is empty, return all properties in the order
|
774 |
(serial, node, hash, size, type, source, mtime, muser, uuid, checksum, cluster).
|
775 |
"""
|
776 |
|
777 |
v = self.versions.alias()
|
778 |
s = select([v.c.serial, v.c.node, v.c.hash, |
779 |
v.c.size, v.c.type, v.c.source, |
780 |
v.c.mtime, v.c.muser, v.c.uuid, |
781 |
v.c.checksum, v.c.cluster], v.c.serial == serial) |
782 |
rp = self.conn.execute(s)
|
783 |
r = rp.fetchone() |
784 |
rp.close() |
785 |
if r is None: |
786 |
return r
|
787 |
|
788 |
if not keys: |
789 |
return r
|
790 |
return [r[propnames[k]] for k in keys if k in propnames] |
791 |
|
792 |
def version_put_property(self, serial, key, value): |
793 |
"""Set value for the property of version specified by key."""
|
794 |
|
795 |
if key not in _propnames: |
796 |
return
|
797 |
s = self.versions.update()
|
798 |
s = s.where(self.versions.c.serial == serial)
|
799 |
s = s.values(**{key: value}) |
800 |
self.conn.execute(s).close()
|
801 |
|
802 |
def version_recluster(self, serial, cluster, |
803 |
update_statistics_ancestors_depth=None):
|
804 |
"""Move the version into another cluster."""
|
805 |
|
806 |
props = self.version_get_properties(serial)
|
807 |
if not props: |
808 |
return
|
809 |
node = props[NODE] |
810 |
size = props[SIZE] |
811 |
oldcluster = props[CLUSTER] |
812 |
if cluster == oldcluster:
|
813 |
return
|
814 |
|
815 |
mtime = time() |
816 |
self.statistics_update_ancestors(node, -1, -size, mtime, oldcluster, |
817 |
update_statistics_ancestors_depth) |
818 |
self.statistics_update_ancestors(node, 1, size, mtime, cluster, |
819 |
update_statistics_ancestors_depth) |
820 |
|
821 |
s = self.versions.update()
|
822 |
s = s.where(self.versions.c.serial == serial)
|
823 |
s = s.values(cluster=cluster) |
824 |
self.conn.execute(s).close()
|
825 |
|
826 |
def version_remove(self, serial, update_statistics_ancestors_depth=None): |
827 |
"""Remove the serial specified."""
|
828 |
|
829 |
props = self.version_get_properties(serial)
|
830 |
if not props: |
831 |
return
|
832 |
node = props[NODE] |
833 |
hash = props[HASH] |
834 |
size = props[SIZE] |
835 |
cluster = props[CLUSTER] |
836 |
|
837 |
mtime = time() |
838 |
self.statistics_update_ancestors(node, -1, -size, mtime, cluster, |
839 |
update_statistics_ancestors_depth) |
840 |
|
841 |
s = self.versions.delete().where(self.versions.c.serial == serial) |
842 |
self.conn.execute(s).close()
|
843 |
|
844 |
props = self.version_lookup(node, cluster=cluster, all_props=False) |
845 |
if props:
|
846 |
self.nodes_set_latest_version(node, serial)
|
847 |
|
848 |
return hash, size |
849 |
|
850 |
def attribute_get(self, serial, domain, keys=()): |
851 |
"""Return a list of (key, value) pairs of the version specified by serial.
|
852 |
If keys is empty, return all attributes.
|
853 |
Othwerise, return only those specified.
|
854 |
"""
|
855 |
|
856 |
if keys:
|
857 |
attrs = self.attributes.alias()
|
858 |
s = select([attrs.c.key, attrs.c.value]) |
859 |
s = s.where(and_(attrs.c.key.in_(keys), |
860 |
attrs.c.serial == serial, |
861 |
attrs.c.domain == domain)) |
862 |
else:
|
863 |
attrs = self.attributes.alias()
|
864 |
s = select([attrs.c.key, attrs.c.value]) |
865 |
s = s.where(and_(attrs.c.serial == serial, |
866 |
attrs.c.domain == domain)) |
867 |
r = self.conn.execute(s)
|
868 |
l = r.fetchall() |
869 |
r.close() |
870 |
return l
|
871 |
|
872 |
def attribute_set(self, serial, domain, node, items, is_latest=True): |
873 |
"""Set the attributes of the version specified by serial.
|
874 |
Receive attributes as an iterable of (key, value) pairs.
|
875 |
"""
|
876 |
#insert or replace
|
877 |
#TODO better upsert
|
878 |
for k, v in items: |
879 |
s = self.attributes.update()
|
880 |
s = s.where(and_(self.attributes.c.serial == serial,
|
881 |
self.attributes.c.domain == domain,
|
882 |
self.attributes.c.key == k))
|
883 |
s = s.values(value=v) |
884 |
rp = self.conn.execute(s)
|
885 |
rp.close() |
886 |
if rp.rowcount == 0: |
887 |
s = self.attributes.insert()
|
888 |
s = s.values(serial=serial, domain=domain, node=node, |
889 |
is_latest=is_latest, key=k, value=v) |
890 |
self.conn.execute(s).close()
|
891 |
|
892 |
def attribute_del(self, serial, domain, keys=()): |
893 |
"""Delete attributes of the version specified by serial.
|
894 |
If keys is empty, delete all attributes.
|
895 |
Otherwise delete those specified.
|
896 |
"""
|
897 |
|
898 |
if keys:
|
899 |
#TODO more efficient way to do this?
|
900 |
for key in keys: |
901 |
s = self.attributes.delete()
|
902 |
s = s.where(and_(self.attributes.c.serial == serial,
|
903 |
self.attributes.c.domain == domain,
|
904 |
self.attributes.c.key == key))
|
905 |
self.conn.execute(s).close()
|
906 |
else:
|
907 |
s = self.attributes.delete()
|
908 |
s = s.where(and_(self.attributes.c.serial == serial,
|
909 |
self.attributes.c.domain == domain))
|
910 |
self.conn.execute(s).close()
|
911 |
|
912 |
def attribute_copy(self, source, dest): |
913 |
s = select( |
914 |
[dest, self.attributes.c.domain, self.attributes.c.node, |
915 |
self.attributes.c.key, self.attributes.c.value], |
916 |
self.attributes.c.serial == source)
|
917 |
rp = self.conn.execute(s)
|
918 |
attributes = rp.fetchall() |
919 |
rp.close() |
920 |
for dest, domain, node, k, v in attributes: |
921 |
select_src_node = select( |
922 |
[self.versions.c.node],
|
923 |
self.versions.c.serial == dest)
|
924 |
# insert or replace
|
925 |
s = self.attributes.update().where(and_(
|
926 |
self.attributes.c.serial == dest,
|
927 |
self.attributes.c.domain == domain,
|
928 |
self.attributes.c.key == k))
|
929 |
s = s.values(node = select_src_node, value=v) |
930 |
rp = self.conn.execute(s)
|
931 |
rp.close() |
932 |
if rp.rowcount == 0: |
933 |
s = self.attributes.insert()
|
934 |
s = s.values(serial=dest, domain=domain, node=select_src_node, |
935 |
is_latest=True, key=k, value=v)
|
936 |
self.conn.execute(s).close()
|
937 |
|
938 |
def attribute_unset_is_latest(self, node, exclude): |
939 |
u = self.attributes.update().where(and_(
|
940 |
self.attributes.c.node == node,
|
941 |
self.attributes.c.serial != exclude)).values(
|
942 |
{'is_latest': False}) |
943 |
self.conn.execute(u)
|
944 |
|
945 |
def latest_attribute_keys(self, parent, domain, before=inf, except_cluster=0, pathq=None): |
946 |
"""Return a list with all keys pairs defined
|
947 |
for all latest versions under parent that
|
948 |
do not belong to the cluster.
|
949 |
"""
|
950 |
|
951 |
pathq = pathq or []
|
952 |
|
953 |
# TODO: Use another table to store before=inf results.
|
954 |
a = self.attributes.alias('a') |
955 |
v = self.versions.alias('v') |
956 |
n = self.nodes.alias('n') |
957 |
s = select([a.c.key]).distinct() |
958 |
if before != inf:
|
959 |
filtered = select([func.max(self.versions.c.serial)])
|
960 |
filtered = filtered.where(self.versions.c.mtime < before)
|
961 |
filtered = filtered.where(self.versions.c.node == v.c.node)
|
962 |
else:
|
963 |
filtered = select([self.nodes.c.latest_version])
|
964 |
filtered = filtered.where(self.nodes.c.node == v.c.node)
|
965 |
s = s.where(v.c.serial == filtered) |
966 |
s = s.where(v.c.cluster != except_cluster) |
967 |
s = s.where(v.c.node.in_(select([self.nodes.c.node],
|
968 |
self.nodes.c.parent == parent)))
|
969 |
s = s.where(a.c.serial == v.c.serial) |
970 |
s = s.where(a.c.domain == domain) |
971 |
s = s.where(n.c.node == v.c.node) |
972 |
conj = [] |
973 |
for path, match in pathq: |
974 |
if match == MATCH_PREFIX:
|
975 |
conj.append( |
976 |
n.c.path.like( |
977 |
self.escape_like(path) + '%', |
978 |
escape=ESCAPE_CHAR |
979 |
) |
980 |
) |
981 |
elif match == MATCH_EXACT:
|
982 |
conj.append(n.c.path == path) |
983 |
if conj:
|
984 |
s = s.where(or_(*conj)) |
985 |
rp = self.conn.execute(s)
|
986 |
rows = rp.fetchall() |
987 |
rp.close() |
988 |
return [r[0] for r in rows] |
989 |
|
990 |
def latest_version_list(self, parent, prefix='', delimiter=None, |
991 |
start='', limit=10000, before=inf, |
992 |
except_cluster=0, pathq=[], domain=None, |
993 |
filterq=[], sizeq=None, all_props=False): |
994 |
"""Return a (list of (path, serial) tuples, list of common prefixes)
|
995 |
for the current versions of the paths with the given parent,
|
996 |
matching the following criteria.
|
997 |
|
998 |
The property tuple for a version is returned if all
|
999 |
of these conditions are true:
|
1000 |
|
1001 |
a. parent matches
|
1002 |
|
1003 |
b. path > start
|
1004 |
|
1005 |
c. path starts with prefix (and paths in pathq)
|
1006 |
|
1007 |
d. version is the max up to before
|
1008 |
|
1009 |
e. version is not in cluster
|
1010 |
|
1011 |
f. the path does not have the delimiter occuring
|
1012 |
after the prefix, or ends with the delimiter
|
1013 |
|
1014 |
g. serial matches the attribute filter query.
|
1015 |
|
1016 |
A filter query is a comma-separated list of
|
1017 |
terms in one of these three forms:
|
1018 |
|
1019 |
key
|
1020 |
an attribute with this key must exist
|
1021 |
|
1022 |
!key
|
1023 |
an attribute with this key must not exist
|
1024 |
|
1025 |
key ?op value
|
1026 |
the attribute with this key satisfies the value
|
1027 |
where ?op is one of ==, != <=, >=, <, >.
|
1028 |
|
1029 |
h. the size is in the range set by sizeq
|
1030 |
|
1031 |
The list of common prefixes includes the prefixes
|
1032 |
matching up to the first delimiter after prefix,
|
1033 |
and are reported only once, as "virtual directories".
|
1034 |
The delimiter is included in the prefixes.
|
1035 |
|
1036 |
If arguments are None, then the corresponding matching rule
|
1037 |
will always match.
|
1038 |
|
1039 |
Limit applies to the first list of tuples returned.
|
1040 |
|
1041 |
If all_props is True, return all properties after path, not just serial.
|
1042 |
"""
|
1043 |
|
1044 |
if not start or start < prefix: |
1045 |
start = strprevling(prefix) |
1046 |
nextling = strnextling(prefix) |
1047 |
|
1048 |
v = self.versions.alias('v') |
1049 |
n = self.nodes.alias('n') |
1050 |
if not all_props: |
1051 |
s = select([n.c.path, v.c.serial]).distinct() |
1052 |
else:
|
1053 |
s = select([n.c.path, |
1054 |
v.c.serial, v.c.node, v.c.hash, |
1055 |
v.c.size, v.c.type, v.c.source, |
1056 |
v.c.mtime, v.c.muser, v.c.uuid, |
1057 |
v.c.checksum, v.c.cluster]).distinct() |
1058 |
if before != inf:
|
1059 |
filtered = select([func.max(self.versions.c.serial)])
|
1060 |
filtered = filtered.where(self.versions.c.mtime < before)
|
1061 |
else:
|
1062 |
filtered = select([self.nodes.c.latest_version])
|
1063 |
s = s.where( |
1064 |
v.c.serial == filtered.where(self.nodes.c.node == v.c.node))
|
1065 |
s = s.where(v.c.cluster != except_cluster) |
1066 |
s = s.where(v.c.node.in_(select([self.nodes.c.node],
|
1067 |
self.nodes.c.parent == parent)))
|
1068 |
|
1069 |
s = s.where(n.c.node == v.c.node) |
1070 |
s = s.where(and_(n.c.path > bindparam('start'), n.c.path < nextling))
|
1071 |
conj = [] |
1072 |
for path, match in pathq: |
1073 |
if match == MATCH_PREFIX:
|
1074 |
conj.append( |
1075 |
n.c.path.like( |
1076 |
self.escape_like(path) + '%', |
1077 |
escape=ESCAPE_CHAR |
1078 |
) |
1079 |
) |
1080 |
elif match == MATCH_EXACT:
|
1081 |
conj.append(n.c.path == path) |
1082 |
if conj:
|
1083 |
s = s.where(or_(*conj)) |
1084 |
|
1085 |
if sizeq and len(sizeq) == 2: |
1086 |
if sizeq[0]: |
1087 |
s = s.where(v.c.size >= sizeq[0])
|
1088 |
if sizeq[1]: |
1089 |
s = s.where(v.c.size < sizeq[1])
|
1090 |
|
1091 |
if domain and filterq: |
1092 |
a = self.attributes.alias('a') |
1093 |
included, excluded, opers = parse_filters(filterq) |
1094 |
if included:
|
1095 |
subs = select([1])
|
1096 |
subs = subs.where(a.c.serial == v.c.serial).correlate(v) |
1097 |
subs = subs.where(a.c.domain == domain) |
1098 |
subs = subs.where(or_(*[a.c.key.op('=')(x) for x in included])) |
1099 |
s = s.where(exists(subs)) |
1100 |
if excluded:
|
1101 |
subs = select([1])
|
1102 |
subs = subs.where(a.c.serial == v.c.serial).correlate(v) |
1103 |
subs = subs.where(a.c.domain == domain) |
1104 |
subs = subs.where(or_(*[a.c.key.op('=')(x) for x in excluded])) |
1105 |
s = s.where(not_(exists(subs))) |
1106 |
if opers:
|
1107 |
for k, o, val in opers: |
1108 |
subs = select([1])
|
1109 |
subs = subs.where(a.c.serial == v.c.serial).correlate(v) |
1110 |
subs = subs.where(a.c.domain == domain) |
1111 |
subs = subs.where( |
1112 |
and_(a.c.key.op('=')(k), a.c.value.op(o)(val)))
|
1113 |
s = s.where(exists(subs)) |
1114 |
|
1115 |
s = s.order_by(n.c.path) |
1116 |
|
1117 |
if not delimiter: |
1118 |
s = s.limit(limit) |
1119 |
rp = self.conn.execute(s, start=start)
|
1120 |
r = rp.fetchall() |
1121 |
rp.close() |
1122 |
return r, ()
|
1123 |
|
1124 |
pfz = len(prefix)
|
1125 |
dz = len(delimiter)
|
1126 |
count = 0
|
1127 |
prefixes = [] |
1128 |
pappend = prefixes.append |
1129 |
matches = [] |
1130 |
mappend = matches.append |
1131 |
|
1132 |
rp = self.conn.execute(s, start=start)
|
1133 |
while True: |
1134 |
props = rp.fetchone() |
1135 |
if props is None: |
1136 |
break
|
1137 |
path = props[0]
|
1138 |
serial = props[1]
|
1139 |
idx = path.find(delimiter, pfz) |
1140 |
|
1141 |
if idx < 0: |
1142 |
mappend(props) |
1143 |
count += 1
|
1144 |
if count >= limit:
|
1145 |
break
|
1146 |
continue
|
1147 |
|
1148 |
if idx + dz == len(path): |
1149 |
mappend(props) |
1150 |
count += 1
|
1151 |
continue # Get one more, in case there is a path. |
1152 |
pf = path[:idx + dz] |
1153 |
pappend(pf) |
1154 |
if count >= limit:
|
1155 |
break
|
1156 |
|
1157 |
rp = self.conn.execute(s, start=strnextling(pf)) # New start. |
1158 |
rp.close() |
1159 |
|
1160 |
return matches, prefixes
|
1161 |
|
1162 |
def latest_uuid(self, uuid, cluster): |
1163 |
"""Return the latest version of the given uuid and cluster.
|
1164 |
|
1165 |
Return a (path, serial) tuple.
|
1166 |
If cluster is None, all clusters are considered.
|
1167 |
|
1168 |
"""
|
1169 |
|
1170 |
v = self.versions.alias('v') |
1171 |
n = self.nodes.alias('n') |
1172 |
s = select([n.c.path, v.c.serial]) |
1173 |
filtered = select([func.max(self.versions.c.serial)])
|
1174 |
filtered = filtered.where(self.versions.c.uuid == uuid)
|
1175 |
if cluster is not None: |
1176 |
filtered = filtered.where(self.versions.c.cluster == cluster)
|
1177 |
s = s.where(v.c.serial == filtered) |
1178 |
s = s.where(n.c.node == v.c.node) |
1179 |
|
1180 |
r = self.conn.execute(s)
|
1181 |
l = r.fetchone() |
1182 |
r.close() |
1183 |
return l
|
1184 |
|
1185 |
def domain_object_list(self, domain, paths, cluster=None): |
1186 |
"""Return a list of (path, property list, attribute dictionary)
|
1187 |
for the objects in the specific domain and cluster.
|
1188 |
"""
|
1189 |
|
1190 |
v = self.versions.alias('v') |
1191 |
n = self.nodes.alias('n') |
1192 |
a = self.attributes.alias('a') |
1193 |
|
1194 |
s = select([n.c.path, v.c.serial, v.c.node, v.c.hash, v.c.size, |
1195 |
v.c.type, v.c.source, v.c.mtime, v.c.muser, v.c.uuid, |
1196 |
v.c.checksum, v.c.cluster, a.c.key, a.c.value]) |
1197 |
if cluster:
|
1198 |
s = s.where(v.c.cluster == cluster) |
1199 |
s = s.where(v.c.serial == a.c.serial) |
1200 |
s = s.where(a.c.domain == domain) |
1201 |
s = s.where(a.c.node == n.c.node) |
1202 |
s = s.where(a.c.is_latest == True)
|
1203 |
if paths:
|
1204 |
s = s.where(n.c.path.in_(paths)) |
1205 |
|
1206 |
r = self.conn.execute(s)
|
1207 |
rows = r.fetchall() |
1208 |
r.close() |
1209 |
|
1210 |
group_by = itemgetter(slice(12)) |
1211 |
rows.sort(key = group_by) |
1212 |
groups = groupby(rows, group_by) |
1213 |
return [(k[0], k[1:], dict([i[12:] for i in data])) \ |
1214 |
for (k, data) in groups] |