hgbook

view web/hgbook/reviewers.py @ 596:4d26453ef459

Fix reviewers script.
author Bryan O'Sullivan <bos@serpentine.com>
date Thu Mar 26 23:01:51 2009 -0700 (2009-03-26)
parents ad304b606163
children
line source
1 #!/usr/bin/env python
2 # -*- coding: utf-8 -*-
4 import os, sys
5 sys.path.append(os.path.dirname(__file__))
6 import dbutil
8 conn = dbutil.connect()
9 c = conn.cursor()
11 c.execute('''select submitter_name from comments_comment''')
13 reviewers = {}
15 mappings = {
16 u'alejandro "tab-lover" dubrovsky': u'Alejandro Dubrovsky',
17 u'alex hirzel <ahirzel@mtu.edu>': u'Alex Hirzel',
18 u'anonymous coward': u'Anonymous',
19 u'arthur van leeuwen': u'Arthur van Leeuwen',
20 u'augustss': u'Lennart Augustsson',
21 u'ed t': u'Anonymous',
22 u'geogre moschovitis': u'George Moschovitis',
23 u'george m': u'George Moschovitis',
24 u'haskell newb': u'Anonymous',
25 u'j. pablo fernandez': u'J. Pablo Fernández',
26 u'kamal al-marhoobi': u'Kamal Al-Marhubi',
27 u'kevin w.': u'Kevin Watters',
28 u'max cantor (#haskell - mxc)': u'Max Cantor',
29 u'michael campbell': u'Michael Campbell',
30 u'mike btauwerman': u'Mike Brauwerman',
31 u'no credit necessary': u'Anonymous',
32 u'nykänen, matti': u'Matti Nykänen',
33 u'omar antolin camarena': u'Omar Antolín Camarena',
34 u'ryan t mulligan': u'Ryan T. Mulligan',
35 u'sengan baring-gould': u'Sengan Baring-Gould',
36 u'some guy': u'Anonymous',
37 u'tomas janousek': u'Tomáš Janoušek',
38 u'william halchin': u'William N. Halchin',
39 }
41 def fixup(s):
42 try:
43 return s.encode('ascii')
44 except UnicodeEncodeError:
45 def f(c):
46 o = ord(c)
47 if o < 128:
48 return c
49 return '&#%d;' % o
50 return ''.join(map(f, s))
52 total = 0
53 for r in c.fetchall():
54 r = r[0].decode('utf-8')
55 if r in ("Bryan O'Sullivan",):
56 continue
57 total += 1
58 m = mappings.get(r.lower())
59 if m:
60 r = m
61 elif len(r) < 2 or ' ' not in r:
62 r = 'Anonymous'
63 reviewers.setdefault(r, 0)
64 reviewers[r] += 1
66 reviewers = sorted(reviewers.iteritems(), key=lambda x: x[0])
68 cohorts = [(.01,1),(.002,.01)]
70 for (lo,hi) in cohorts:
71 lo = total * lo
72 hi = total * hi
73 for r in [n for n in reviewers if lo <= n[1] < hi]:
74 if r[1] > 3:
75 print '%s,' % fixup(r[0])
76 print
78 lo = total * .002
79 for n in reviewers:
80 if n[1] < lo:
81 print '%s,' % fixup(n[0])