-
Notifications
You must be signed in to change notification settings - Fork 15
/
bot_log_parser.py
269 lines (219 loc) · 9.87 KB
/
bot_log_parser.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
import glob
import os
import webbrowser
from ast import literal_eval
from functools import lru_cache
from json import JSONDecodeError
import click
import pandas as pd
import json
import sys
import re
import requests
from dateutil.parser import parse as dateutil_parse
from jinja2 import Template
pd.options.display.max_colwidth = 100
@lru_cache()
def get_prop_formatter(pid):
if not (pid.startswith("P") and isint(pid[1:])):
return None
try:
d = requests.get("https://www.wikidata.org/wiki/Special:EntityData/{}.json".format(pid)).json()
return d['entities'][pid]['claims']['P1630'][0]['mainsnak']['datavalue']['value']
except Exception:
return None
def parse_log(file_path):
# todo: Actually parse the header and col names
# note, missing Rev ID in the old logs will just be NaN and won't throw an error
df = pd.read_csv(file_path, sep=",",
names=['Level', 'Timestamp', 'External ID', 'Prop', 'QID', 'Message', 'Msg Type', 'Rev ID'],
skiprows=2, dtype={'External ID': str, 'Rev ID': str},
comment='#', quotechar='"', skipinitialspace=True, delimiter=';')
df.fillna('', inplace=True)
df.replace("None", "", inplace=True)
df = df.apply(lambda x: x.str.strip())
df.Timestamp = pd.to_datetime(df.Timestamp, format='%m/%d/%Y %H:%M:%S')
return df
def gen_ext_id_links(df: pd.DataFrame):
# given the columns "Prop" and "External ID", if prop is a wikidata property, get the formatter URL
# and create links to the external ID
df['External ID'] = df.apply(lambda row: get_ext_id_link(row['Prop'], row['External ID']), axis=1)
return df
def get_ext_id_link(pid, ext_id):
formatter = get_prop_formatter(pid)
if formatter:
url = formatter.replace("$1", ext_id)
return '<a href="{}">{}</a>'.format(url, ext_id)
else:
return ext_id
def process_log(file_path):
"""
Expects header as first line in log file. Header begins with comment character '#'. The line is a json string dump of
a dictionary that contains the following keys:
name: str, Task name
maintainer: str, Name of person
tags: list of tags associated with the task. can be empty
properties: list of properties associated with the task. can be empty
run_id: str, a run ID for the task run
timestamp: str, timestamp for the task run
:param file_path:
:return:
"""
# read header
if isinstance(file_path, str):
with open(file_path) as f:
line = f.readline()
else:
line = file_path.readline()
if not line.startswith("#"):
raise ValueError("Expecting header in log file")
try:
metadata = json.loads(line[1:])
if 'timestamp' in metadata:
metadata['timestamp'] = dateutil_parse(metadata['timestamp'])
except JSONDecodeError as e:
metadata = {"name": "", "timestamp": "", "run_id": ""}
df = parse_log(file_path)
return df, metadata
def generate_summary(df):
level_counts = df.Level.value_counts().to_dict()
zlist = list(zip(*[('<a href="#info">Items Processed Succesfully</a>', level_counts.get('INFO', 0)),
('<a href="#warning">Items Skipped Due to a Warning</a>', level_counts.get('WARNING', 0)),
('<a href="#error">Items Skipped Due to an Error</a>', level_counts.get('ERROR', 0))]))
level_counts = pd.Series(zlist[1], index=zlist[0])
level_counts.name = "Count"
info_counts = df.query("Level == 'INFO'").Message.value_counts().to_dict()
zlist = list(zip(*[('No Action', info_counts.get('SKIP', 0)),
('Update', info_counts.get('UPDATE', 0)),
('Create', info_counts.get('CREATE', 0))]))
info_counts = pd.Series(zlist[1], index=zlist[0])
info_counts.name = "Count"
warning_counts = df.query("Level == 'WARNING'")['Msg Type'].value_counts()
warning_counts.name = "Count"
error_counts = df.query("Level == 'ERROR'")['Msg Type'].value_counts()
error_counts.name = "Count"
return level_counts, info_counts, warning_counts, error_counts
def isint(s):
try:
int(s)
except Exception:
return False
return True
def url_qid(df, col):
href = "<a href=https://www.wikidata.org/wiki/{}>{}</a>"
f = lambda x: href.format(x, x) if x.startswith("Q") and isint(x[1:]) else x
# df.is_copy = False
df.loc[:, col] = df[col].apply(f)
return df
def escape_html_chars(s):
return s.replace("&", r'&').replace("<", r'<').replace(">", r'>')
def try_json(s):
try:
d = json.loads(s.replace("'", '"'))
return d
except Exception:
return s
def wiki_links_to_html(s):
# in a string, convert things like "[[Q27826279|Q27826279]]" into
# '<a href="https://www.wikidata.org/wiki/Q27826279>Q27826279</a>'
for match in re.findall('\[\[(Q\d{1,12})\|Q\d{1,12}\]\]', s):
s = s.replace("[[{}|{}]]".format(match, match),
'<a href="https://www.wikidata.org/wiki/{}">{}</a>'.format(match, match))
return s
def format_error(error_type, s):
# attempts to format an error message, depending on the type of error
if "WDApiError" in error_type or "NonUniqueLabelDescriptionPairError" in error_type:
return format_wdapierror(try_json(s))
if "ManualInterventionReqException" in error_type:
return format_ManualInterventionReqException(s)
else:
return s
def format_ManualInterventionReqException(s):
# More than one WD item has the same property value Property: P1748, items affected: ['186020', '18557906']
if "More than one WD item has the same property value" in s:
pid = s.split("Property: ")[1].split(",")[0].strip()
qids = literal_eval(s.split("items affected: ")[1])
urls = ['<a href="https://www.wikidata.org/wiki/Q{}#{}">Q{}</a>'.format(x, pid, x) for x in qids]
return "More than one WD item has the same property value: {}".format(", ".join(urls))
elif "does not match provided core ID" in s:
qid = s.split("Retrieved item (")[1].split(")")[0]
s = s.replace(qid, '<a href="https://www.wikidata.org/wiki/{}">{}</a>'.format(qid, qid))
return s
else:
return s
def format_wdapierror(d):
# if the message type is a WDApiError, attempt to clean it up a little
try:
return {'code': d['error']['code'],
'info': wiki_links_to_html(d['error']['info']),
'messages': d['error']['messages']}
except Exception:
return d
@click.command()
@click.argument('log-path')
@click.option('--show-browser', default=False, is_flag=True, help='show log in browser')
def main(log_path, show_browser=False):
if os.path.isdir(log_path):
# run on all files in dir (that don't end in html), and ignore show_browser
files = [x for x in glob.glob(os.path.join(log_path, "*")) if
not x.endswith(".html") and os.stat(x).st_size != 0]
for file in files:
try:
_main(file)
except Exception as e:
print("Parsing log failed: {}".format(file))
else:
_main(log_path, show_browser)
def _main(log_path, show_browser=False):
print(log_path)
df, metadata = process_log(log_path)
del df['Timestamp']
df['Msg Type'] = df['Msg Type'].apply(escape_html_chars)
df['Message'] = df['Message'].apply(escape_html_chars)
# df['Message'] = df['Message'].apply(try_json)
df['Message'] = df.apply(lambda row: format_error(row['Msg Type'], row['Message']), 1)
df['Rev ID'] = df['Rev ID'].apply(lambda x: '<a href="https://www.wikidata.org/w/index.php?oldid={}&diff=prev">{}</a>'.format(x,x) if x else x)
level_counts, info_counts, warning_counts, error_counts = generate_summary(df)
warnings_df = df.query("Level == 'WARNING'")
warnings_df.is_copy = False
del warnings_df['Level']
if not warnings_df.empty:
warnings_df = gen_ext_id_links(warnings_df)
warnings_df = url_qid(warnings_df, "QID")
errors_df = df.query("Level == 'ERROR'")
errors_df.is_copy = False
del errors_df['Level']
if not errors_df.empty:
errors_df = gen_ext_id_links(errors_df)
errors_df = url_qid(errors_df, "QID")
# errors_df['Message'] = errors_df['Message'].apply(try_format_error)
info_df = df.query("Level == 'INFO'")
info_df.is_copy = False
del info_df['Level']
if not info_df.empty:
info_df = gen_ext_id_links(info_df)
info_df = url_qid(info_df, "QID")
info_df.Message = info_df.Message.str.replace("SKIP", "No Action")
with pd.option_context('display.max_colwidth', -1):
# this class nonsense is an ugly hack: https://stackoverflow.com/questions/15079118/js-datatables-from-pandas/41536906
level_counts = level_counts.to_frame().to_html(escape=False)
info_counts = info_counts.to_frame().to_html(escape=False)
warning_counts = warning_counts.to_frame().to_html(escape=False)
error_counts = error_counts.to_frame().to_html(escape=False)
info_df = info_df.to_html(escape=False, classes='df" id = "info_df')
warnings_df = warnings_df.to_html(escape=False, classes='df" id = "warning_df')
errors_df = errors_df.to_html(escape=False, classes='df" id = "error_df')
template = Template(open(os.path.join(sys.path[0], "template.html")).read())
s = template.render(name=metadata['name'], run_id=metadata['run_id'],
level_counts=level_counts,
info_counts=info_counts,
warning_counts=warning_counts,
error_counts=error_counts,
warnings_df=warnings_df, errors_df=errors_df, info_df=info_df)
out_path = log_path.rsplit(".", 1)[0] + ".html"
with open(out_path, 'w') as f:
f.write(s)
if show_browser:
webbrowser.open(out_path)
if __name__ == "__main__":
main()