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
|
#!/usr/bin/env python3
"""
Enrich videobot’s europarl streamdumps with metadata from europarltv, including
topic, speaker and timestamps.
Expects a config file videobot_enricheuroparl.ini in the working directory.
TODO:
Additional speaker info: http://www.europarl.europa.eu/ep-live/en/speaker/1911?date=27-04-2016
"""
import json, math
from datetime import datetime, timedelta
from pytz import timezone
from dateutil.parser import parse
import requests
brussels = timezone('Europe/Brussels')
def fromTimestamp (ts):
"""
Convert timestamp to datetime. Dates are local time (i.e. utc+1) and in
milliseconds.
"""
return datetime.fromtimestamp (ts/1000, tz=brussels)
def overlap (a, b):
"""
Do intervals (a0, a1) and (b0, b1) overlap?
"""
def contains (a, b):
return a[0] <= b[0] <= a[1] or a[0] <= b[1] <= a[1]
return contains (a, b) or contains (b, a)
def convertTimestamps (event):
"""
Convert timestamps to datetime
"""
chapters = event['chapters']
for c in chapters:
c['startTimeCode'] = fromTimestamp (c['startTimeCode'])
c['endTimeCode'] = fromTimestamp (c['endTimeCode'])
for s in c['speeches']:
s['startTimeCode'] = fromTimestamp (s['startTimeCode'])
s['endTimeCode'] = fromTimestamp (s['endTimeCode'])
return event
def filterRecording (event, recstart, recend):
"""
Remove event data not within recording range
"""
def inRecording (o):
return overlap ((recstart, recend), (o['startTimeCode'], o['endTimeCode']))
event['chapters'] = sorted (filter (inRecording, event['chapters']), key = lambda x: x['startTimeCode'])
for c in event['chapters']:
c['speeches'] = sorted (filter (inRecording, c['speeches']), key = lambda x: x['startTimeCode'])
return event
def formatTimedelta (d, precision=0, minlen=3):
"""
Convert timedelta to string
"""
sec = round (d.total_seconds ()*1000)
div = [(1000, '{:03d}', ''), (60, '{:02d}', '.'), (60, '{:02d}', ':'), (24, '{:02d}', ':')]
res = ""
c = 1
for factor, fmt, sep in div:
sec, r = divmod (sec, factor)
if c > precision:
# no separator if nothing is there yet
if not res:
sep = ''
res = fmt.format (r) + sep + res
if sec < 1 and c >= minlen:
break
c += 1
return res
def formatDescription (event, recstart, fd):
"""
Create IA description
"""
chapters = ['Recording of European Parliament plenary session on {humandate}.<br><a href="http://www.europarl.europa.eu/sides/getDoc.do?type=AGENDA&reference={date}&secondRef=SIT&language=EN">Agenda</a>, <a href="http://www.europarl.europa.eu/sides/getDoc.do?type=PV&reference={date}&secondRef=TOC&language=EN">Minutes</a>'.format (date=recstart.strftime ('%Y%m%d'), humandate=recstart.strftime ('%Y-%m-%d %H:%M %Z'))]
for c in event['chapters']:
start = c['startTimeCode']
end = c['endTimeCode']
if start < recstart:
prefix = '{} (continued)'.format (c['title']['en'])
else:
prefix = '{}'.format (c['title']['en'])
speeches = []
for s in c['speeches']:
start = s['startTimeCode']
end = s['endTimeCode']
continued = False
if start < recstart:
continued = True
line = '{}'.format (formatTimedelta (timedelta (0), 1))
else:
line = '{}'.format (formatTimedelta (start-recstart, 1))
line += ' <a href="http://www.europarl.europa.eu/meps/en/{}/mep.html">{} {}</a>'.format (s['externalId'], s['firstName'], s['lastName'])
if continued:
line += ' (continued)'
speeches.append (line)
chapters.append (prefix + '<br>' + '<br>'.join (speeches))
print ('<br><br>'.join (chapters), file=fd)
def formatWebVTT (event, recstart, fd):
"""
Create WebVTT subtitles with speaker and topic
"""
def formatEvent (start, end, data):
return '{} --> {}\n{}\n\n'.format (formatTimedelta (start), formatTimedelta (end), data)
print ('WEBVTT\n', file=fd)
for c in event['chapters']:
# XXX: assuming the player can show multiple subtitles at the same time
speeches = c['speeches']
start = c['startTimeCode']
if start < recstart:
start = recstart
relstart = start-recstart
relend = c['endTimeCode']-recstart
print (formatEvent (relstart, relend, 'Topic: {}'.format (c['title']['en'])), file=fd)
for s in c['speeches']:
start = s['startTimeCode']
if start < recstart:
start = recstart
relstart = start-recstart
relend = s['endTimeCode']-recstart
print (formatEvent (relstart, relend, 'Speaker: {} {}'.format (s['firstName'], s['lastName'])), file=fd)
if __name__ == '__main__':
import argparse, sys, configparser
config = configparser.ConfigParser()
config.read ('videobot_enricheuroparl.ini')
parser = argparse.ArgumentParser(description='Enrich videobot’s europarl items with metadata.')
parser.add_argument ('identifier', help='Internet Archive identifier')
args = parser.parse_args()
iaitem = requests.get ('https://archive.org/metadata/{}'.format (args.identifier)).json ()
if 'error' in iaitem:
print (iaitem['error'], file=sys.stderr)
sys.exit (1)
recordingtz = timezone (config['recording']['tz'])
recordingstart = recordingtz.localize (parse (iaitem['metadata']['date'])).astimezone (brussels)
recordingend = recordingstart+timedelta (seconds = int (config['recording']['length']))
o = requests.get ('http://www.europarl.europa.eu/ep-live/en/json/plenary/video?date={}'.format (recordingstart.strftime ('%d-%m-%Y'))).json ()
# save raw data
with open ('{}.json'.format (args.identifier), 'w') as fd:
json.dump (o, fd)
event = filterRecording (convertTimestamps (o['event']), recordingstart, recordingend)
#event = convertTimestamps (o['event'])
with open ('{}.html'.format (args.identifier), 'w') as fd:
formatDescription (event, recordingstart, fd)
with open ('{}.vtt'.format (args.identifier), 'w') as fd:
formatWebVTT (event, recordingstart, fd)
|
