summaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools/importers/wayback_static.py
blob: 5b23347fa0a405d00abec59f32a911987dd1fc23 (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
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
#!/usr/bin/env python3

"""
Helpers to create Web Capture entities from extracted wayback content.

Works as a stand-alone script (for debugging) or as library routines.
"""

import sys
import json
import hashlib
import requests
import datetime
import argparse
import subprocess
from bs4 import BeautifulSoup

from fatcat_openapi_client import *
from .common import b32_hex

CDX_API_BASE = "https://web.archive.org/cdx/search/cdx"
GWB_URL_BASE = "https://web.archive.org/web"
REQ_SESSION = requests.Session()


def parse_wbm_url(url):
    """Takes a wayback machine URL, and returns a tuple:

        (timestamp, datetime, original_url)
    """
    chunks = url.split('/')
    assert len(chunks) >= 6
    assert chunks[2] == 'web.archive.org'
    assert chunks[3] == 'web'
    return (chunks[4],
            parse_wbm_timestamp(chunks[4]),
            '/'.join(chunks[5:]))

def test_parse_wbm_url():
    u = "http://web.archive.org/web/20010712114837/http://www.dlib.org/dlib/june01/reich/06reich.html"
    assert parse_wbm_url(u) == (
        "20010712114837",
        datetime.datetime(2001, 7, 12, 11, 48, 37),
        "http://www.dlib.org/dlib/june01/reich/06reich.html")

def parse_wbm_timestamp(timestamp):
    """
    Takes a complete WBM timestamp string (like "20020327115625") and returns a
    python datetime object (UTC)
    """
    # strip any "im_" or "id_" suffix
    if timestamp.endswith('_'):
        timestamp = timestamp[:-3]
    # inflexible; require the full second-precision timestamp
    assert len(timestamp) == 14
    return datetime.datetime(
        year=int(timestamp[0:4]),
        month=int(timestamp[4:6]),
        day=int(timestamp[6:8]),
        hour=int(timestamp[8:10]),
        minute=int(timestamp[10:12]),
        second=int(timestamp[12:14]))

def test_parse_wbm_timestamp():
    assert parse_wbm_timestamp("20010712114837") == \
        datetime.datetime(2001, 7, 12, 11, 48, 37)

def fetch_wbm(url):
    resp = REQ_SESSION.get(url)
    resp.raise_for_status()
    assert resp.content
    return resp.content

def lookup_cdx(embed_url, verify_hashes=True, cdx_output=None):
    sys.stderr.write(embed_url + "\n")
    assert embed_url.startswith('/web/')
    embed_url = embed_url.split('/')
    timestamp = embed_url[2]
    if timestamp.endswith('_'):
        timestamp = timestamp[:-3]
    url = '/'.join(embed_url[3:])
    #print((timestamp, url))
    resp = REQ_SESSION.get(CDX_API_BASE, params=dict(
        url=url,
        closest=timestamp,
        sort="closest",
        resolveRevisits="true",
        matchType="exact",
        limit=1,
    ))
    resp.raise_for_status()
    #print(resp.url)
    if resp.content:
        hit = resp.content.decode('utf-8').split('\n')[0]
        if cdx_output:
            cdx_output.write(hit + "\n")
        cdx = hit.split(' ')
        cdx = [x if (x and x != '-') else None for x in cdx]
        webcapture_cdx = WebcaptureCdxLine(
            surt=cdx[0],
            timestamp=parse_wbm_timestamp(cdx[1]).isoformat() + "Z",
            url=cdx[2],
            mimetype=cdx[3],
            status_code=(cdx[4] and int(cdx[4])) or None,
            sha1=b32_hex(cdx[5]),
            sha256=None,
        )
        if verify_hashes:
            resp = REQ_SESSION.get(GWB_URL_BASE + "/{}id_/{}".format(
                cdx[1], # raw timestamp
                webcapture_cdx.url))
            resp.raise_for_status()
            assert webcapture_cdx.sha1 == hashlib.sha1(resp.content).digest().hex()
            webcapture_cdx.sha256 = hashlib.sha256(resp.content).digest().hex()
            webcapture_cdx.size = len(resp.content)
        return webcapture_cdx
    else:
        return None

def wayback_url_to_relative(url):
    """
    Wayback URLs can be relative or absolute in rewritten documents. This
    function converts any form of rewritten URL to a relative (to
    web.archive.org) one, or returns None if it isn't a rewritten URL at all.
    """
    if url.startswith('https://web.archive.org/'):
        url = url[23:]
    elif url.startswith('http://web.archive.org/'):
        url = url[22:]

    if url.startswith('/web/'):
        return url
    else:
        return None

def extract_embeds(soup):

    embeds = set()

    # <link href="">
    for tag in soup.find_all('link', href=True):
        if tag['rel'] not in ('stylesheet',):
            continue
        url = wayback_url_to_relative(tag['href'])
        if url:
            embeds.add(url)
    # <img src="">
    for tag in soup.find_all('img', src=True):
        url = wayback_url_to_relative(tag['src'])
        if url:
            embeds.add(url)

    # <script src="">
    for tag in soup.find_all('script', src=True):
        url = wayback_url_to_relative(tag['src'])
        if url:
            embeds.add(url)

    return list(embeds)

def static_wayback_webcapture(wayback_url, cdx_output=None):
    """
    Given a complete wayback machine capture URL, like:

        http://web.archive.org/web/20010712114837/http://www.dlib.org/dlib/june01/reich/06reich.html

    Will return a new ("bare") fatcat webcapture entity python object, with all
    the CDX entries filled in.
    """

    wbm_html = fetch_wbm(wayback_url)
    raw_timestamp, timestamp, original_url = parse_wbm_url(wayback_url)
    #with open(rewritten_path, 'r') as fp:
    #    soup = BeautifulSoup(fp, "lxml")
    soup = BeautifulSoup(wbm_html, "lxml")
    embeds = extract_embeds(soup)
    cdx_obj = lookup_cdx("/web/{}/{}".format(raw_timestamp, original_url),
        cdx_output=cdx_output)
    cdx_list = [cdx_obj]
    for url in embeds:
        cdx_obj = lookup_cdx(url, cdx_output=cdx_output)
        cdx_list.append(cdx_obj)
    archive_urls = [WebcaptureUrl(
        rel="wayback",
        url="https://web.archive.org/web/",
    )]
    wc = WebcaptureEntity(
        cdx=cdx_list,
        timestamp=timestamp.isoformat() + "Z",
        original_url=original_url,
        archive_urls=archive_urls,
        release_ids=None)
    return wc

def auto_wayback_static(api, release_id, wayback_url, editgroup_id=None):
    """
    Returns a tuple: (editgroup_id, edit). If failed, both are None
    """

    raw_timestamp, timestamp, original_url = parse_wbm_url(wayback_url)
    git_rev = subprocess.check_output(
        ["git", "describe", "--always"]).strip().decode('utf-8')

    release = api.get_release(release_id, expand="webcaptures")

    # check for existing webcapture with same parameters
    for wc in release.webcaptures:
        if wc.original_url == original_url and wc.timestamp.date() == timestamp.date():
            # skipping: already existed
            print("release {} already had webcapture {} {}".format(
                release_id, raw_timestamp, original_url))
            return (None, None)

    wc = static_wayback_webcapture(wayback_url)
    assert len(wc.cdx) >= 1
    wc.release_ids = [release_id]
    if not editgroup_id:
        eg = api.create_editgroup(Editgroup(
            description="One-off import of static web content from wayback machine",
            extra=dict(
                git_rev=git_rev,
                agent="fatcat_tools.auto_wayback_static")))
        editgroup_id = eg.editgroup_id
    edit = api.create_webcapture(eg.editgroup_id, wc)
    return (editgroup_id, edit)

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument('--verbose',
        action='store_true',
        help="verbose output")
    parser.add_argument('wayback_url',
        type=str,
        help="URL of wayback capture to extract from")
    parser.add_argument('--json-output',
        type=argparse.FileType('w'), default=sys.stdout,
        help="where to write out webcapture entity (as JSON)")
    parser.add_argument('--cdx-output',
        type=argparse.FileType('w'), default=None,
        help="(optional) file to write out CDX stub")

    args = parser.parse_args()

    # entity-to-JSON code; duplicate of entity_to_dict()
    api_client = ApiClient()
    wc = static_wayback_webcapture(args.wayback_url, cdx_output=args.cdx_output)
    wc_dict = api_client.sanitize_for_serialization(wc)
    print(json.dumps(wc_dict))

if __name__ == '__main__':
    main()