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
|
#!/usr/bin/env python
"""Generate the errorcodes module starting from PostgreSQL documentation.
The script can be run at a new PostgreSQL release to refresh the module.
"""
# Copyright (C) 2010 Daniele Varrazzo <daniele.varrazzo@gmail.com>
#
# psycopg2 is free software: you can redistribute it and/or modify it
# under the terms of the GNU Lesser General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# psycopg2 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 Lesser General Public
# License for more details.
import re
import sys
import urllib2
from collections import defaultdict
from BeautifulSoup import BeautifulSoup as BS
def main():
if len(sys.argv) != 2:
print >>sys.stderr, "usage: %s /path/to/errorcodes.py" % sys.argv[0]
return 2
filename = sys.argv[1]
file_start = read_base_file(filename)
classes, errors = fetch_errors(
['8.1', '8.2', '8.3', '8.4', '9.0', '9.1', '9.2', '9.3'])
f = open(filename, "w")
for line in file_start:
print >>f, line
for line in generate_module_data(classes, errors):
print >>f, line
def read_base_file(filename):
rv = []
for line in open(filename):
rv.append(line.rstrip("\n"))
if line.startswith("# autogenerated"):
return rv
raise ValueError("can't find the separator. Is this the right file?")
def parse_errors_txt(url):
classes = {}
errors = defaultdict(dict)
page = urllib2.urlopen(url)
for line in page:
# Strip comments and skip blanks
line = line.split('#')[0].strip()
if not line:
continue
# Parse a section
m = re.match(r"Section: (Class (..) - .+)", line)
if m:
label, class_ = m.groups()
classes[class_] = label
continue
# Parse an error
m = re.match(r"(.....)\s+(?:E|W|S)\s+ERRCODE_(\S+)(?:\s+(\S+))?$", line)
if m:
errcode, macro, spec = m.groups()
# error 22008 has 2 macros and 1 def: give priority to the def
# as it's the one we used to parse from sgml
if not spec:
if errcode in errors[class_]:
continue
errlabel = macro.upper()
else:
errlabel = spec.upper()
errors[class_][errcode] = errlabel
continue
# We don't expect anything else
raise ValueError("unexpected line:\n%s" % line)
return classes, errors
def parse_errors_sgml(url):
page = BS(urllib2.urlopen(url))
table = page('table')[1]('tbody')[0]
classes = {}
errors = defaultdict(dict)
for tr in table('tr'):
if tr.td.get('colspan'): # it's a class
label = ' '.join(' '.join(tr(text=True)).split()) \
.replace(u'\u2014', '-').encode('ascii')
assert label.startswith('Class')
class_ = label.split()[1]
assert len(class_) == 2
classes[class_] = label
else: # it's an error
errcode = tr.tt.string.encode("ascii")
assert len(errcode) == 5
tds = tr('td')
if len(tds) == 3:
errlabel = '_'.join(tds[1].string.split()).encode('ascii')
# double check the columns are equal
cond_name = tds[2].string.strip().upper().encode("ascii")
assert errlabel == cond_name, tr
elif len(tds) == 2:
# found in PG 9.1 docs
errlabel = tds[1].tt.string.upper().encode("ascii")
else:
assert False, tr
errors[class_][errcode] = errlabel
return classes, errors
errors_sgml_url = \
"http://www.postgresql.org/docs/%s/static/errcodes-appendix.html"
errors_txt_url = \
"http://git.postgresql.org/gitweb/?p=postgresql.git;a=blob_plain;" \
"f=src/backend/utils/errcodes.txt;hb=REL%s_STABLE"
def fetch_errors(versions):
classes = {}
errors = defaultdict(dict)
for version in versions:
print >> sys.stderr, version
tver = tuple(map(int, version.split('.')))
if tver < (9, 1):
c1, e1 = parse_errors_sgml(errors_sgml_url % version)
else:
c1, e1 = parse_errors_txt(
errors_txt_url % version.replace('.', '_'))
classes.update(c1)
for c, cerrs in e1.iteritems():
errors[c].update(cerrs)
return classes, errors
def generate_module_data(classes, errors):
yield ""
yield "# Error classes"
for clscode, clslabel in sorted(classes.items()):
err = clslabel.split(" - ")[1].split("(")[0] \
.strip().replace(" ", "_").replace('/', "_").upper()
yield "CLASS_%s = %r" % (err, clscode)
for clscode, clslabel in sorted(classes.items()):
yield ""
yield "# %s" % clslabel
for errcode, errlabel in sorted(errors[clscode].items()):
yield "%s = %r" % (errlabel, errcode)
if __name__ == '__main__':
sys.exit(main())
|
