aboutsummaryrefslogtreecommitdiffstats
path: root/check_issn_urls.py
blob: 71593f5c500686cc203d633a0304081bd3ed8180 (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
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
#!/usr/bin/env python3
"""
Check journal homepage status (live web and wayback)

Takes a tsv filepath or lines on stdin and dumps to stdout. The stdin thing means you can:

    # be sure to randomize order if you are going to use high parallelism so no
    # individual domain gets swamped. also remember this hits CDX API multiple
    # times.
    parallel -j10 --bar --pipepart -a urls_to_crawl.shuf.tsv ./check_issn_urls.py > url_status.json

Input columns (no header):

    ISSN-L, URL

For each URL, do a request and record, as JSON:

    issnl: passthrough
    url: passthrough
    status_code: initial HTTP crawl status
    terminal_url: final URL (or original if no redirects)
    terminal_status_code: final URL (or original if no redirects)
    terminal_content_type: content type (mimetype)
    platform_software: slug of hosting platform, if detected
    issnl_in_body: whether raw issnl appears in body text
    blocked: whether we think crawler was "blocked"
    gwb_url_success_dt: latest wayback datetime that an HTTP 200 exists
    gwb_terminal_url_success_dt: latest wayback datetime that an HTTP 200 exists

HTTP status will be -1 if domain does not even resolve.
"""

import sys
import json
import time
import requests


def sniff_platform(resp):
    """
    This function would try to figure out what software platform (eg, OJS) the
    site is running.
    TODO: unimplemented
    """
    # these are mostly here to filter out huge platforms and stop sniffing
    domain_map = {
        "jstor.org/": "jstor",
        "springer.com/": "springer",
        "springerlink.com/": "springer",
        "tandfonline.com/": "t_and_f",
        "elsevier.com/": "elsevier",
        "wiley.com/": "wiley",
        "sciencedirect.com/": "elsevier",
        "sagepub.com/": "sage",
        "hypotheses.org/": "hypothesis",
        "tandf.co.uk/": "t_and_f",
        "scielo": "scielo",
    }
    for domain, platform in domain_map.items():
        if domain in resp.url:
            return platform
    if '<meta name="generator" content="Open Journal Systems' in resp.text:
        return "ojs"
    return None


def sniff_blocked(resp):
    """
    This function would try to figure out if we got blocked: soft-block, hard
    block, etc.
    TODO: unimplemented
    """
    if resp.status_code in (403, 420, 429):
        return True
    # JSTOR does this
    if (
        "Our systems have detected unusual traffic activity from your network. Please complete this reCAPTCHA"
        in resp.text
    ):
        return True
    if (
        resp.status_code == 416
        and "something about your browser made us think you were a bot" in resp.text
    ):
        return True
    return None


def check_gwb(url, match_type="exact"):
    if "//web.archive.org/" in url:
        return None
    # crude/bad retry loop to work around CDX API throttling
    for i in range(5):
        resp = requests.get(
            "https://web.archive.org/cdx/search/cdx",
            params={
                "url": url,
                "matchType": match_type,
                "limit": -1,
                "filter": "statuscode:200",
            },
        )
        if resp.status_code == 200:
            break
        time.sleep(5)
    if not resp.status_code == 200:
        sys.stderr.write("CDX ERR {}: {}\n".format(resp.status_code, url))
        # TODO: this isn't really correct, but not sure what to return/record
        # if we failed through all timeouts
        return None
    line = resp.text.strip().split("\n")[0]
    if line:
        dt = line.split()[1]
        int(dt)
        return dt
    else:
        return None


def check_url(issnl, url):
    # print("Fetching: %s" % url)
    info = dict(issnl=issnl, url=url)
    try:
        resp = requests.get(
            url,
            timeout=15.0,
            headers={
                "User-Agent": "ia_bot/0.0 (python requests) journal-live-check; contact:info@archive.org"
            },
        )
    except requests.exceptions.TooManyRedirects:
        info["error"] = "TooManyRedirects"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.SSLError:
        info["error"] = "SSLError"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.ReadTimeout:
        info["error"] = "ReadTimeout"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.ConnectionError:
        info["error"] = "ConnectionError"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.ChunkedEncodingError:
        info["error"] = "ChunkedEncodingError"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.ContentDecodingError:
        info["error"] = "ContentDecodingError"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except requests.exceptions.InvalidSchema:
        info["error"] = "InvalidSchema"
        info["terminal_status_code"] = info["status_code"] = -1
        return info
    except UnicodeDecodeError:
        info["error"] = "UnicodeDecodeError"
        info["terminal_status_code"] = info["status_code"] = -1
        return info

    if resp.history:
        info["status_code"] = resp.history[0].status_code
    else:
        info["status_code"] = resp.status_code

    info["terminal_status_code"] = resp.status_code
    info["terminal_url"] = resp.url
    content_type = resp.headers.get("Content-Type")
    if content_type:
        info["terminal_content_type"] = content_type.split(";")[0]
    info["issnl_in_body"] = bool(issnl in resp.text)
    info["gwb_url_success_dt"] = check_gwb(url, match_type="exact")
    info["gwb_terminal_url_success_dt"] = check_gwb(
        info["terminal_url"], match_type="exact"
    )
    info["blocked"] = sniff_blocked(resp)
    info["software_platform"] = sniff_platform(resp)
    # info['gwb_host_success_dt'] = check_gwb(url, match_type='host')
    return info


def run(tsvfile):
    for line in tsvfile:
        records = line.split("\t")
        issnl = records[0]
        url = records[1].strip()
        print(json.dumps(check_url(issnl, url)))


if __name__ == "__main__":
    if len(sys.argv) != 2:
        f = sys.stdin
    else:
        f = open(sys.argv[1], "r")
    run(f)