aboutsummaryrefslogtreecommitdiffstats
path: root/python/sandcrawler/pdftrio.py
blob: 161dc9c3116b8faa2a8bb3c4d4b502b9c4712162 (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

import time
import requests

from .workers import SandcrawlerWorker, SandcrawlerFetchWorker
from .misc import gen_file_metadata, requests_retry_session


class PdfTrioClient(object):

    def __init__(self, host_url="http://pdftrio.qa.fatcat.wiki", **kwargs):
        self.host_url = host_url
        self.http_session = requests_retry_session(retries=3, backoff_factor=3)

    def classify_pdf(self, blob, mode="auto"):
        """
        Returns a dict with at least:

            - status_code (int, always set)
            - status (success, or error-*)

        On success, the other remote API JSON response keys are also included.

        On HTTP-level failures, the status_code and status field are set
        appropriately; an optional `error_msg` may also be set. For some other
        errors, like connection failure, an exception is raised.
        """
        assert blob

        try:
            pdftrio_response = requests.post(
                self.host_url + "/classify/research-pub/" + mode,
                files={
                    'pdf_content': blob,
                },
                timeout=60.0,
            )
        except requests.Timeout:
            return {
                'status': 'error-timeout',
                'status_code': -4,  # heritrix3 "HTTP timeout" code
                'error_msg': 'pdftrio request (HTTP POST) timeout',
            }
        except requests.exceptions.ConnectionError:
            # crude back-off
            time.sleep(2.0)
            return {
                'status': 'error-connect',
                'status_code': -2,  # heritrix3 "HTTP connect" code
                'error_msg': 'pdftrio request connection timout',
            }

        info = dict(
            status_code=pdftrio_response.status_code,
        )
        if pdftrio_response.status_code == 200:
            resp_json = pdftrio_response.json()
            assert 'ensemble_score' in resp_json
            assert 'status' in resp_json
            assert 'versions' in resp_json
            info.update(resp_json)
        else:
            info['status'] = 'error'
            # TODO: might return JSON with some info?

        info['_total_sec'] = pdftrio_response.elapsed.total_seconds()
        return info


class PdfTrioWorker(SandcrawlerFetchWorker):
    """
    This class is basically copied directly from GrobidWorker
    """

    def __init__(self, pdftrio_client, wayback_client=None, sink=None, **kwargs):
        super().__init__(wayback_client=wayback_client)
        self.pdftrio_client = pdftrio_client
        self.sink = sink

    def process(self, record, key=None):
        start_process = time.time()
        default_key = record['sha1hex']
        fetch_sec = None

        start = time.time()
        fetch_result = self.fetch_blob(record)
        fetch_sec = time.time() - start
        if fetch_result['status'] != 'success':
            return fetch_result
        blob = fetch_result['blob']

        result = dict()
        result['file_meta'] = gen_file_metadata(blob)
        result['key'] = result['file_meta']['sha1hex']
        result['pdf_trio'] = self.pdftrio_client.classify_pdf(blob)
        result['source'] = record
        result['timing'] = dict(
            pdftrio_sec=result['pdf_trio'].pop('_total_sec', None),
            total_sec=time.time() - start_process,
        )
        if fetch_sec:
            result['timing']['fetch_sec'] = fetch_sec
        return result

class PdfTrioBlobWorker(SandcrawlerWorker):
    """
    This is sort of like PdfTrioWorker, except it receives blobs directly,
    instead of fetching blobs from some remote store.
    """

    def __init__(self, pdftrio_client, sink=None, mode="auto", **kwargs):
        super().__init__()
        self.pdftrio_client = pdftrio_client
        self.sink = sink
        self.mode = mode

    def process(self, blob, key=None):
        start_process = time.time()
        if not blob:
            return None
        result = dict()
        result['file_meta'] = gen_file_metadata(blob)
        result['key'] = result['file_meta']['sha1hex']
        result['pdf_trio'] = self.pdftrio_client.classify_pdf(blob, mode=self.mode)
        result['timing'] = dict(
            pdftrio_sec=result['pdf_trio'].pop('_total_sec', None),
            total_sec=time.time() - start_process,
        )
        return result