aboutsummaryrefslogtreecommitdiffstats
path: root/mapreduce/tests/test_extraction_cdx_grobid.py
blob: 02d2b4121c2a3920844a0e68ffd66e4917036473 (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

import io
import json
import mrjob
import pytest
import struct
import responses
import happybase_mock
from unittest import mock
from extraction_cdx_grobid import MRExtractCdxGrobid


FAKE_PDF_BYTES = b"%PDF SOME JUNK" + struct.pack("!q", 112853843)

@pytest.fixture
def job():
    """
    Note: this mock only seems to work with job.run_mapper(), not job.run();
    the later results in a separate instantiation without the mock?
    """
    job = MRExtractCdxGrobid(['--no-conf', '-'])

    conn = happybase_mock.Connection()
    conn.create_table('wbgrp-journal-extract-test',
        {'file': {}, 'grobid0': {}, 'f': {}})
    job.hb_table = conn.table('wbgrp-journal-extract-test')

    return job


@mock.patch('extraction_cdx_grobid.MRExtractCdxGrobid.fetch_warc_content', return_value=(FAKE_PDF_BYTES, None))
@responses.activate
def test_mapper_lines(mock_fetch, job):

    with open('tests/files/23b29ea36382680716be08fc71aa81bd226e8a85.xml', 'r') as f:
        real_tei_xml = f.read()
    responses.add(responses.POST, 'http://localhost:8070/api/processFulltextDocument', status=200,
        body=real_tei_xml, content_type='application/json')

    raw = io.BytesIO(b"""
com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 301 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz
eu,eui,cadmus)/bitstream/handle/1814/36635/rscas_2015_03.pdf;jsessionid=761393014319a39f40d32ae3eb3a853f?sequence=1 20170705062202 http://cadmus.eui.eu/bitstream/handle/1814/36635/RSCAS_2015_03.pdf%3Bjsessionid%3D761393014319A39F40D32AE3EB3A853F?sequence%3D1 application/PDF 200 MPCXVWMUTRUGFP36SLPHKDLY6NGU4S3J - - 854156 328850624 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz
com,pbworks,educ333b)/robots.txt 20170705063311 http://educ333b.pbworks.com/robots.txt text/plain 200 6VAUYENMOU2SK2OWNRPDD6WTQTECGZAD - - 638 398190140 CITESEERX-CRAWL-2017-06-20-20170705062707827-00049-00058-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705063158203-00053-31209~wbgrp-svc284.us.archive.org~8443.warc.gz
""")

    output = io.BytesIO()
    job.sandbox(stdin=raw, stdout=output)

    job.run_mapper()

    # for debugging tests
    #print(output.getvalue().decode('utf-8'))
    #print(list(job.hb_table.scan()))

    # wayback gets FETCH 1x times
    # TODO:

    # grobid gets POST 3x times
    # TODO:

    # hbase 
    # TODO:


    assert job.hb_table.row(b'1') == {}
    # HTTP 301
    assert job.hb_table.row(b'sha1:3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ') == {}
    # valid
    assert job.hb_table.row(b'sha1:MPCXVWMUTRUGFP36SLPHKDLY6NGU4S3J') != {}
    # text/plain
    assert job.hb_table.row(b'sha1:6VAUYENMOU2SK2OWNRPDD6WTQTECGZAD') == {}

    row = job.hb_table.row(b'sha1:MPCXVWMUTRUGFP36SLPHKDLY6NGU4S3J')

    assert struct.unpack("!q", row[b'file:size'])[0] == len(FAKE_PDF_BYTES)
    assert row[b'file:mime'] == b"application/pdf"
    assert struct.unpack("!q", row[b'grobid0:status_code'])[0] == 200
    # TODO: assert row[b'grobid0:quality'] == None
    status = json.loads(row[b'grobid0:status'].decode('utf-8'))
    assert type(status) == type(dict())
    assert row[b'grobid0:tei_xml'].decode('utf-8') == real_tei_xml
    tei_json = json.loads(row[b'grobid0:tei_json'].decode('utf-8'))
    metadata = json.loads(row[b'grobid0:metadata'].decode('utf-8'))
    assert tei_json['title'] == metadata['title']
    assert 'body' in tei_json
    assert 'body' not in metadata

def test_parse_cdx_invalid(job):

    print("valid")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert status is None

    print("space-prefixed line")
    raw = " com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert info is None
    assert status['status'] == "invalid"
    assert 'prefix' in status['reason']

    print("commented line")
    raw = "#com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert info is None
    assert status['status'] == "invalid"
    assert 'prefix' in status['reason']

    print("wrong column count")
    raw = "a b c d"
    info, status = job.parse_line(raw)
    assert info is None
    assert status['status'] == "invalid"
    assert 'parse' in status['reason']

    print("missing mimetype")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf - 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert info is None
    print(status)
    assert status['status'] == "invalid"
    assert 'parse' in status['reason']

    print("HTTP status")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 501 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert info is None
    assert status['status'] == "invalid"

    print("datetime")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 501 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.parse_line(raw)
    assert info is None
    assert status['status'] == "invalid"


def test_parse_cdx_skip(job):

    job.mapper_init()

    print("warc format")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf application/pdf 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.mapper(None, raw).__next__()
    assert info is None
    assert status['status'] == "skip"
    assert 'WARC' in status['reason']

    print("mimetype")
    raw = "com,sagepub,cep)/content/28/9/960.full.pdf 20170705062200 http://cep.sagepub.com/content/28/9/960.full.pdf text/html 200 3I42H3S6NNFQ2MSVX7XZKYAYSCX5QBYJ - - 401 313356621 CITESEERX-CRAWL-2017-06-20-20170705061647307-00039-00048-wbgrp-svc284/CITESEERX-CRAWL-2017-06-20-20170705062052659-00043-31209~wbgrp-svc284.us.archive.org~8443.warc.gz"
    info, status = job.mapper(None, raw).__next__()
    assert info is None
    assert status['status'] == "skip"
    assert 'mimetype' in status['reason']

def test_tei_json_convert():
    # TODO: load xml test vector, run it through
    with open('tests/files/23b29ea36382680716be08fc71aa81bd226e8a85.xml', 'r') as f:
        xml_content = f.read()
    pass

def test_tei_json_convert_invalid():
    # TODO: pass in junk
    pass