blob: cf9626f2d20af51d7a7b41bf9ce0c46e16e37d72 (
plain)
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
|
Need beautiful soup; we aren't using pipenv here:
sudo apt install python3-bs4 python3-lxml
## CNKI List
Base URL: <http://en.gzbd.cnki.net/GZBT/brief/Default.aspx>
2020-03-29: "Found 1914 articles"
2020-04-06: "Found 2224 articles"
Uses JS to fetch tables, URLs look like:
http://en.gzbd.cnki.net/gzbt/request/otherhandler.ashx?action=gzbdFlag&contentID=0&orderStr=1&page=1&grouptype=undefined&groupvalue=undefined
Fetch a bunch:
# bump this seq number based on number of articles divided by 30, round up
seq 0 77 | parallel http get "http://en.gzbd.cnki.net/gzbt/request/otherhandler.ashx?action=gzbdFlag\&contentID=0\&orderStr=1\&page={}\&grouptype=undefined\&groupvalue=undefined" > metadata/cnki_tables.`date -I`.html
Parse HTML snippets to JSON:
./extra/scrape/parse_cnki_tables.py metadata/cnki_tables.`date -I`.html > metadata/cnki_metadata.`date -I`.json
The `info_url` seems to work, but the direct PDF download links don't naively.
Maybe need to set a referer, something like that?
## Wanfang Data
mark=32 指南与共识 Guidelines and consensus
mark=34 文献速递 Literature Express
mark=38 中医药防治 Prevention and treatment of traditional Chinese medicine
wget 'http://subject.med.wanfangdata.com.cn/Channel/7?mark=32' -O metadata/wanfang_guidance.`date -I`.html
wget 'http://subject.med.wanfangdata.com.cn/Channel/7?mark=34' -O metadata/wanfang_papers.`date -I`.html
./extra/scrape/parse_wanfang_html.py metadata/wanfang_papers.`date -I`.html > metadata/wanfang_papers.`date -I`.json
./extra/scrape/parse_wanfang_html.py metadata/wanfang_guidance.`date -I`.html > metadata/wanfang_guidance.`date -I`.json
Download PDFs (without clobbering existing):
cat metadata/wanfang_papers.`date -I`.json metadata/wanfang_guidance.`date -I`.json | jq .url -r | shuf | parallel wget -P fulltext_wanfang --no-clobber {}
Rename based on mimetype:
./bin/fix_extensions.sh fulltext_wanfang
What did we get?
file fulltext_wanfang/* | awk '{print $2}' | sort | uniq -c
144 HTML
609 PDF
2020-04-06:
wc -l metadata/wanfang_*.json
200 metadata/wanfang_guidance.2020-04-06.json
739 metadata/wanfang_papers.2020-04-06.json
939 total
144 HTML
679 PDF
ls fulltext_wanfang | wc -l
823
|
