/
changes.py
302 lines (262 loc) · 11.3 KB
/
changes.py
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
# This file is part of Buildbot. Buildbot is free software: you can
# redistribute it and/or modify it under the terms of the GNU General Public
# License as published by the Free Software Foundation, version 2.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
# details.
#
# You should have received a copy of the GNU General Public License along with
# this program; if not, write to the Free Software Foundation, Inc., 51
# Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Copyright Buildbot Team Members
"""
Support for changes in the database
"""
import sqlalchemy as sa
from buildbot.db import base
from buildbot.util import datetime2epoch
from buildbot.util import epoch2datetime
from buildbot.util import json
from twisted.internet import defer
from twisted.internet import reactor
from twisted.python import log
class ChDict(dict):
pass
class ChangesConnectorComponent(base.DBConnectorComponent):
# Documentation is in developer/db.rst
@defer.inlineCallbacks
def addChange(self, author=None, files=None, comments=None, is_dir=None,
revision=None, when_timestamp=None, branch=None,
category=None, revlink='', properties={}, repository='', codebase='',
project='', uid=None, _reactor=reactor):
assert project is not None, "project must be a string, not None"
assert repository is not None, "repository must be a string, not None"
if is_dir is not None:
log.msg("WARNING: change source is providing deprecated "
"value is_dir (ignored)")
if when_timestamp is None:
when_timestamp = epoch2datetime(_reactor.seconds())
# verify that source is 'Change' for each property
for pv in properties.values():
assert pv[1] == 'Change', ("properties must be qualified with"
"source 'Change'")
ch_tbl = self.db.model.changes
self.checkLength(ch_tbl.c.author, author)
self.checkLength(ch_tbl.c.branch, branch)
self.checkLength(ch_tbl.c.revision, revision)
self.checkLength(ch_tbl.c.revlink, revlink)
self.checkLength(ch_tbl.c.category, category)
self.checkLength(ch_tbl.c.repository, repository)
self.checkLength(ch_tbl.c.project, project)
# calculate the sourcestamp first, before adding it
ssid = yield self.db.sourcestamps.findSourceStampId(
revision=revision, branch=branch, repository=repository,
codebase=codebase, project=project, _reactor=_reactor)
def thd(conn):
# note that in a read-uncommitted database like SQLite this
# transaction does not buy atomicity - other database users may
# still come across a change without its files, properties,
# etc. That's OK, since we don't announce the change until it's
# all in the database, but beware.
transaction = conn.begin()
r = conn.execute(ch_tbl.insert(), dict(
author=author,
comments=comments,
branch=branch,
revision=revision,
revlink=revlink,
when_timestamp=datetime2epoch(when_timestamp),
category=category,
repository=repository,
codebase=codebase,
project=project,
sourcestampid=ssid))
changeid = r.inserted_primary_key[0]
if files:
tbl = self.db.model.change_files
for f in files:
self.checkLength(tbl.c.filename, f)
conn.execute(tbl.insert(), [
dict(changeid=changeid, filename=f)
for f in files
])
if properties:
tbl = self.db.model.change_properties
inserts = [
dict(changeid=changeid,
property_name=k,
property_value=json.dumps(v))
for k, v in properties.iteritems()
]
for i in inserts:
self.checkLength(tbl.c.property_name,
i['property_name'])
self.checkLength(tbl.c.property_value,
i['property_value'])
conn.execute(tbl.insert(), inserts)
if uid:
ins = self.db.model.change_users.insert()
conn.execute(ins, dict(changeid=changeid, uid=uid))
transaction.commit()
return changeid
defer.returnValue((yield self.db.pool.do(thd)))
@base.cached("chdicts")
def getChange(self, changeid):
assert changeid >= 0
def thd(conn):
# get the row from the 'changes' table
changes_tbl = self.db.model.changes
q = changes_tbl.select(whereclause=(changes_tbl.c.changeid == changeid))
rp = conn.execute(q)
row = rp.fetchone()
if not row:
return None
# and fetch the ancillary data (files, properties)
return self._chdict_from_change_row_thd(conn, row)
d = self.db.pool.do(thd)
return d
def getChangeUids(self, changeid):
assert changeid >= 0
def thd(conn):
cu_tbl = self.db.model.change_users
q = cu_tbl.select(whereclause=(cu_tbl.c.changeid == changeid))
res = conn.execute(q)
rows = res.fetchall()
row_uids = [row.uid for row in rows]
return row_uids
d = self.db.pool.do(thd)
return d
def getRecentChanges(self, count):
def thd(conn):
# get the changeids from the 'changes' table
changes_tbl = self.db.model.changes
q = sa.select([changes_tbl.c.changeid],
order_by=[sa.desc(changes_tbl.c.changeid)],
limit=count)
rp = conn.execute(q)
changeids = [row.changeid for row in rp]
rp.close()
return list(reversed(changeids))
d = self.db.pool.do(thd)
# then turn those into changes, using the cache
def get_changes(changeids):
return defer.gatherResults([self.getChange(changeid)
for changeid in changeids])
d.addCallback(get_changes)
return d
def getChanges(self):
def thd(conn):
# get the changeids from the 'changes' table
changes_tbl = self.db.model.changes
q = sa.select([changes_tbl.c.changeid])
rp = conn.execute(q)
changeids = [row.changeid for row in rp]
rp.close()
return list(changeids)
d = self.db.pool.do(thd)
# then turn those into changes, using the cache
def get_changes(changeids):
return defer.gatherResults([self.getChange(changeid)
for changeid in changeids])
d.addCallback(get_changes)
return d
def getChangesCount(self):
def thd(conn):
changes_tbl = self.db.model.changes
q = sa.select([sa.func.count()]).select_from(changes_tbl)
rp = conn.execute(q)
r = 0
for row in rp:
r = row[0]
rp.close()
return int(r)
d = self.db.pool.do(thd)
return d
def getLatestChangeid(self):
def thd(conn):
changes_tbl = self.db.model.changes
q = sa.select([changes_tbl.c.changeid],
order_by=sa.desc(changes_tbl.c.changeid),
limit=1)
return conn.scalar(q)
d = self.db.pool.do(thd)
return d
# utility methods
def pruneChanges(self, changeHorizon):
"""
Called periodically by DBConnector, this method deletes changes older
than C{changeHorizon}.
"""
if not changeHorizon:
return defer.succeed(None)
def thd(conn):
changes_tbl = self.db.model.changes
# First, get the list of changes to delete. This could be written
# as a subquery but then that subquery would be run for every
# table, which is very inefficient; also, MySQL's subquery support
# leaves much to be desired, and doesn't support this particular
# form.
q = sa.select([changes_tbl.c.changeid],
order_by=[sa.desc(changes_tbl.c.changeid)],
offset=changeHorizon)
res = conn.execute(q)
ids_to_delete = [r.changeid for r in res]
# and delete from all relevant tables, in dependency order
for table_name in ('scheduler_changes', 'change_files',
'change_properties', 'changes', 'change_users'):
remaining = ids_to_delete[:]
while remaining:
batch, remaining = remaining[:100], remaining[100:]
table = self.db.model.metadata.tables[table_name]
conn.execute(
table.delete(table.c.changeid.in_(batch)))
return self.db.pool.do(thd)
def _chdict_from_change_row_thd(self, conn, ch_row):
# This method must be run in a db.pool thread, and returns a chdict
# given a row from the 'changes' table
change_files_tbl = self.db.model.change_files
change_properties_tbl = self.db.model.change_properties
chdict = ChDict(
changeid=ch_row.changeid,
author=ch_row.author,
files=[], # see below
comments=ch_row.comments,
revision=ch_row.revision,
when_timestamp=epoch2datetime(ch_row.when_timestamp),
branch=ch_row.branch,
category=ch_row.category,
revlink=ch_row.revlink,
properties={}, # see below
repository=ch_row.repository,
codebase=ch_row.codebase,
project=ch_row.project,
sourcestampid=int(ch_row.sourcestampid))
query = change_files_tbl.select(
whereclause=(change_files_tbl.c.changeid == ch_row.changeid))
rows = conn.execute(query)
for r in rows:
chdict['files'].append(r.filename)
# and properties must be given without a source, so strip that, but
# be flexible in case users have used a development version where the
# change properties were recorded incorrectly
def split_vs(vs):
try:
v, s = vs
if s != "Change":
v, s = vs, "Change"
except (ValueError, TypeError):
v, s = vs, "Change"
return v, s
query = change_properties_tbl.select(
whereclause=(change_properties_tbl.c.changeid == ch_row.changeid))
rows = conn.execute(query)
for r in rows:
try:
v, s = split_vs(json.loads(r.property_value))
chdict['properties'][r.property_name] = (v, s)
except ValueError:
pass
return chdict