File size: 4,684 Bytes
2319518
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import datetime
import os
import re
from urllib.parse import unquote, urlparse

import add_qwen_libs  # NOQA
import jsonlines

from qwen_agent.log import logger
from qwen_agent.utils.doc_parser import parse_doc, parse_html_bs
from qwen_agent.utils.utils import print_traceback, save_text_to_file
from qwen_server.schema import Record


def is_local_path(path):
    if path.startswith('https://') or path.startswith('http://'):
        return False
    return True


def sanitize_chrome_file_path(file_path: str) -> str:
    # For Linux and macOS.
    if os.path.exists(file_path):
        return file_path

    # For native Windows, drop the leading '/' in '/C:/'
    win_path = file_path
    if win_path.startswith('/'):
        win_path = win_path[1:]
    if os.path.exists(win_path):
        return win_path

    # For Windows + WSL.
    if re.match(r'^[A-Za-z]:/', win_path):
        wsl_path = f'/mnt/{win_path[0].lower()}/{win_path[3:]}'
        if os.path.exists(wsl_path):
            return wsl_path

    # For native Windows, replace / with \.
    win_path = win_path.replace('/', '\\')
    if os.path.exists(win_path):
        return win_path

    return file_path


def extract_and_cache_document(data, cache_file, cache_root):
    logger.info('Starting cache pages...')
    if data['url'].split('.')[-1].lower() in ['pdf', 'docx', 'pptx']:
        date1 = datetime.datetime.now()

        # generate one processing record
        new_record = Record(url=data['url'],
                            time='',
                            type=data['type'],
                            raw=[],
                            extract='',
                            topic='',
                            checked=False,
                            session=[]).to_dict()
        with jsonlines.open(cache_file, mode='a') as writer:
            writer.write(new_record)

        if data['url'].startswith('https://') or data['url'].startswith(
                'http://'):
            pdf_path = data['url']
        else:
            parsed_url = urlparse(data['url'])
            pdf_path = unquote(parsed_url.path)
            pdf_path = sanitize_chrome_file_path(pdf_path)

        try:
            pdf_content = parse_doc(pdf_path)
        except Exception:
            print_traceback()
            # del the processing record
            lines = []
            if os.path.exists(cache_file):
                for line in jsonlines.open(cache_file):
                    if line['url'] != data['url']:
                        lines.append(line)
            with jsonlines.open(cache_file, mode='w') as writer:
                for new_line in lines:
                    writer.write(new_line)
            return 'failed'

        date2 = datetime.datetime.now()
        logger.info('Parsing pdf time: ' + str(date2 - date1))
        data['content'] = pdf_content
        data['type'] = 'pdf'
        extract = pdf_path.split('/')[-1].split('\\')[-1].split('.')[0]
    elif data['content'] and data['type'] == 'html':
        new_record = Record(url=data['url'],
                            time='',
                            type=data['type'],
                            raw=[],
                            extract='',
                            topic='',
                            checked=False,
                            session=[]).to_dict()
        with jsonlines.open(cache_file, mode='a') as writer:
            writer.write(new_record)

        try:
            tmp_html_file = os.path.join(cache_root, 'tmp.html')
            save_text_to_file(tmp_html_file, data['content'])
            data['content'] = parse_html_bs(tmp_html_file)
        except Exception:
            print_traceback()
        extract = data['content'][0]['metadata']['title']
    else:
        logger.error(
            'Only Support the Following File Types: [\'.html\', \'.pdf\', \'.docx\', \'.pptx\']'
        )
        raise NotImplementedError

    today = datetime.date.today()
    new_record = Record(url=data['url'],
                        time=str(today),
                        type=data['type'],
                        raw=data['content'],
                        extract=extract,
                        topic='',
                        checked=True,
                        session=[])
    lines = []
    if os.path.exists(cache_file):
        for line in jsonlines.open(cache_file):
            if line['url'] != data['url']:
                lines.append(line)
    lines.append(new_record.to_dict())  # cache
    with jsonlines.open(cache_file, mode='w') as writer:
        for new_line in lines:
            writer.write(new_line)

    response = 'Cached'
    return response