3 """ Various statistical pr0nography fun and games """
4 # Copyright (C) 2000, 2001, 2002, 2003, 2006 James Troup <james@nocrew.org>
6 # This program is free software; you can redistribute it and/or modify
7 # it under the terms of the GNU General Public License as published by
8 # the Free Software Foundation; either version 2 of the License, or
9 # (at your option) any later version.
11 # This program is distributed in the hope that it will be useful,
12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 # GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License
17 # along with this program; if not, write to the Free Software
18 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
20 ################################################################################
22 # <aj> can we change the standards instead?
24 # <aj> whatever we're not conforming to
25 # <aj> if there's no written standard, why don't we declare linux as
26 # the defacto standard
29 # [aj's attempt to avoid ABI changes for released architecture(s)]
31 ################################################################################
35 from daklib import utils
36 from daklib.dbconn import DBConn, get_suite_architectures, Suite, Architecture, \
39 ################################################################################
43 ################################################################################
45 def usage(exit_code=0):
46 print """Usage: dak stats MODE
49 -h, --help show this help and exit.
51 The following MODEs are available:
53 arch-space - displays space used by each architecture
54 pkg-nums - displays the number of packages by suite/architecture
55 daily-install - displays daily install stats suitable for graphing
59 ################################################################################
61 def per_arch_space_use():
62 session = DBConn().session()
63 q = session.execute("""
64 SELECT a.arch_string as Architecture, sum(f.size) AS sum
65 FROM files f, binaries b, architecture a
66 WHERE a.id=b.architecture AND f.id=b.file
67 GROUP BY a.arch_string ORDER BY sum""").fetchall()
69 print "%-15.15s %s" % (j[0], j[1])
71 q = session.execute("SELECT sum(size) FROM files WHERE filename ~ '.(diff.gz|tar.gz|dsc)$'").fetchall()
72 print "%-15.15s %s" % ("Source", q[0][0])
74 ################################################################################
76 def daily_install_stats():
78 f = utils.open_file("2001-11")
79 for line in f.readlines():
80 split = line.strip().split('|')
82 if program != "katie" and program != "process-accepted":
85 if action != "installing changes" and action != "installed":
88 if not stats.has_key(date):
90 stats[date]["packages"] = 0
91 stats[date]["size"] = 0.0
92 if action == "installing changes":
93 stats[date]["packages"] += 1
94 elif action == "installed":
95 stats[date]["size"] += float(split[5])
100 packages = stats[date]["packages"]
101 size = int(stats[date]["size"] / 1024.0 / 1024.0)
102 print "%s %s %s" % (date, packages, size)
104 ################################################################################
114 def suite_sort(a, b):
115 if Cnf.has_key("Suite::%s::Priority" % (a)):
116 a_priority = int(Cnf["Suite::%s::Priority" % (a)])
119 if Cnf.has_key("Suite::%s::Priority" % (b)):
120 b_priority = int(Cnf["Suite::%s::Priority" % (b)])
123 return cmp(a_priority, b_priority)
125 def output_format(suite):
127 for word in suite.split("-"):
128 output_suite.append(word[0])
129 return "-".join(output_suite)
131 # Obvious query with GROUP BY and mapped names -> 50 seconds
132 # GROUP BY but ids instead of suite/architecture names -> 28 seconds
133 # Simple query -> 14 seconds
134 # Simple query into large dictionary + processing -> 21 seconds
135 # Simple query into large pre-created dictionary + processing -> 18 seconds
137 def number_of_packages():
143 session = DBConn().session()
144 # Build up suite mapping
145 for i in session.query(Suite).all():
146 suites[i.suite_id] = i.suite_name
147 suite_ids[i.suite_name] = i.suite_id
148 # Build up architecture mapping
149 for i in session.query(Architecture).all():
150 arches[i.arch_id] = i.arch_string
151 arch_ids[i.arch_string] = i.arch_id
152 # Pre-create the dictionary
153 for suite_id in suites.keys():
155 for arch_id in arches.keys():
156 d[suite_id][arch_id] = 0
157 # Get the raw data for binaries
158 # Simultate 'GROUP by suite, architecture' with a dictionary
159 # XXX: Why don't we just get the DB to do this?
160 for i in session.query(BinAssociation):
161 suite_id = i.suite_id
162 arch_id = i.binary.arch_id
163 d[suite_id][arch_id] = d[suite_id][arch_id] + 1
164 # Get the raw data for source
165 arch_id = arch_ids["source"]
166 for i in session.execute('SELECT suite, COUNT(suite) FROM src_associations GROUP BY suite').all():
167 (suite_id, count) = i
168 d[suite_id][arch_id] = d[suite_id][arch_id] + count
171 suite_list = suites.values()
172 suite_list.sort(suite_sort)
175 for suite in suite_list:
176 suite_id = suite_ids[suite]
177 suite_arches[suite_id] = {}
178 for arch in get_suite_architectures(suite):
179 suite_arches[suite_id][arch.arch_string] = ""
180 suite_id_list.append(suite_id)
181 output_list = [ output_format(i) for i in suite_list ]
182 longest_suite = longest(output_list)
183 arch_list = arches.values()
185 longest_arch = longest(arch_list)
187 output = (" "*longest_arch) + " |"
188 for suite in output_list:
189 output = output + suite.center(longest_suite)+" |"
190 output = output + "\n"+(len(output)*"-")+"\n"
192 arch_list = arches.values()
194 longest_arch = longest(arch_list)
195 for arch in arch_list:
196 arch_id = arch_ids[arch]
197 output = output + arch.center(longest_arch)+" |"
198 for suite_id in suite_id_list:
199 if suite_arches[suite_id].has_key(arch):
200 count = repr(d[suite_id][arch_id])
203 output = output + count.rjust(longest_suite)+" |"
204 output = output + "\n"
207 ################################################################################
212 Cnf = utils.get_conf()
213 Arguments = [('h',"help","Stats::Options::Help")]
215 if not Cnf.has_key("Stats::Options::%s" % (i)):
216 Cnf["Stats::Options::%s" % (i)] = ""
218 args = apt_pkg.ParseCommandLine(Cnf, Arguments, sys.argv)
220 Options = Cnf.SubTree("Stats::Options")
225 utils.warn("dak stats requires a MODE argument")
228 utils.warn("dak stats accepts only one MODE argument")
230 mode = args[0].lower()
232 if mode == "arch-space":
234 elif mode == "pkg-nums":
236 elif mode == "daily-install":
237 daily_install_stats()
239 utils.warn("unknown mode '%s'" % (mode))
242 ################################################################################
244 if __name__ == '__main__':