/
normalize.py
231 lines (188 loc) · 6.45 KB
/
normalize.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
#!/usr/bin/env python
"""
Script for extracting relevant data points from raw data.
Invocation:
$ python normalize.py data/raw/*.xml -o data/normalized
"""
import argparse
import sys
import os
from bs4 import BeautifulSoup
from common import write_json_file, progress, read_records
GND_PREFIX = "http://d-nb.info/gnd"
# Metadata extraction functions
def find_tags_in_id_range(soup, start, end):
tags = [tag for tag in soup.find_all("varfield")
if tag['id'] and tag['id'].isnumeric() and
int(tag['id']) >= start and int(tag['id']) < end]
return tags
def persons(soup):
persons = []
tags = find_tags_in_id_range(soup, 100, 200)
for tag in tags:
person = {}
# name
name = tag.find(label="p")
if not name:
name = tag.find(label="a")
ext = ''
if ' ' not in name.string:
ext_tag = tag.find(label="n")
if ext_tag:
ext = ' ' + tag.find(label="n").string
if name:
person['name'] = name.string + ext
# lifetime
lifetime = tag.find(label="d")
if lifetime:
person['lifetime'] = lifetime.string
# role
role = tag.find(label="b")
if role:
role = role.string.replace("[", "").replace("]", "")
person['role'] = role
# gnd_link
gnd_link = tag.find(label="9")
save = True
if gnd_link != '':
save = False
if gnd_link:
gnd_link = GND_PREFIX + "/" + gnd_link.string[8:]
person['sameas'] = [gnd_link]
if save:
persons.append(person)
return persons
def content(soup):
contents = []
for tag in soup.find_all("varfield", id="655"):
uri = tag.find("subfield", label="u")
note = tag.find("subfield", label="z")
if note:
note = note.string
if uri:
content = {
'uri': uri.string,
'note': note
}
contents.append(content)
return contents
def title(soup):
tag = soup.find("varfield", id="303")
if not tag:
return None
else:
title = tag.find("subfield", label="t")
if not title:
return None
else:
return title.string
def subtitles(soup):
subtitles = []
tags = find_tags_in_id_range(soup, 304, 400)
for tag in tags:
subfields = tag.find_all("subfield")
for subfield in subfields:
contents = subfield.string.replace("[", "").replace("]", "")
subtitles.append(contents)
return subtitles
def dates(soup):
dates = []
date_tags = soup.find_all("varfield", id="425")
for tag in date_tags:
dates.append(tag.subfield.string)
return dates
def gnd_link(soup):
tag = soup.find("varfield", id="303")
if not tag:
return None
else:
gnd_link = tag.find("subfield", label="9")
if not gnd_link:
return None
else:
return GND_PREFIX + "/" + gnd_link.string[8:]
def notes(soup):
notes = []
tags = find_tags_in_id_range(soup, 400, 600)
for tag in tags:
if tag.id != "425":
for subfield in tag.find_all("subfield", label="a"):
notes.append(subfield.string)
return notes
def terms(soup):
terms = []
tags = find_tags_in_id_range(soup, 900, 1000)
for tag in tags:
term = {}
labels = []
for subfield in tag.find_all("subfield"):
if subfield["label"] != "9":
labels.append(subfield.string)
if(len(labels) > 0):
term['labels'] = labels
gnd_link = tag.find("subfield", label="9")
if gnd_link:
gnd_link = GND_PREFIX + "/" + gnd_link.string[8:]
term['sameas'] = [gnd_link]
terms.append(term)
return terms
def doc_id(soup):
return soup.doc_number.string
def aleph_id(soup):
return soup.find("varfield", id="001").find("subfield", label="a").string
# Main normalization routine
def normalize(raw_record):
soup = BeautifulSoup(raw_record, "html.parser")
normalized_record = {}
# mandatory fields
normalized_record['aleph_id'] = aleph_id(soup)
normalized_record['doc_id'] = doc_id(soup)
# optional fields
if title(soup):
normalized_record['title'] = title(soup)
if subtitles(soup):
normalized_record['subtitles'] = subtitles(soup)
if persons(soup):
normalized_record['persons'] = persons(soup)
if content(soup):
normalized_record['content'] = content(soup)
if dates(soup):
normalized_record['dates'] = dates(soup)
if notes(soup):
#normalized_notes = notes(soup)
#print 'normalized_notes: ', normalized_notes
#new_notes = [unicode(elem).encode('utf-8') for elem in normalized_notes]
#normalized_record['notes'] = new_notes
#print 'new_notes: ', new_notes
# normalized_record['notes'] = notes(soup)
normalized_record['notes'] = notes(soup)
if terms(soup):
normalized_record['terms'] = terms(soup)
if gnd_link(soup):
normalized_record['sameas'] = [gnd_link(soup)]
return normalized_record
def normalize_records(inputfiles, outputdir):
print("Normalizing", len(inputfiles), "records to", outputdir)
for index, (filename, record) in enumerate(read_records(inputfiles)):
progress((index+1)/len(inputfiles))
normalized_record = normalize(record)
print 'normalized record: ', normalized_record
#new_normalized_record = [unicode(elem).encode('utf-8') for elem in normalized_record]
#print 'new normalized record: ', new_normalized_record
out_file = os.path.basename(filename).replace("xml", "json")
write_json_file(outputdir, out_file, normalized_record)
# write_json_file(outputdir, out_file, new_normalized_record)
# Command line parsing
if __name__ == '__main__':
parser = argparse.ArgumentParser(
description="Extract relevant fields and convert to JSON.")
parser.add_argument('inputfiles', type=str, nargs='+',
help="Input files to be processed")
parser.add_argument('-o', '--outputdir', type=str, nargs='?',
default="data/normalized",
help="Output directory")
if len(sys.argv) < 2:
parser.print_help()
sys.exit(1)
args = parser.parse_args()
normalize_records(args.inputfiles, args.outputdir)