File size: 4,205 Bytes
072f9dd
 
 
 
 
51482f3
 
072f9dd
96a1a44
072f9dd
51482f3
072f9dd
 
96a1a44
072f9dd
 
 
 
 
 
 
51482f3
072f9dd
 
51482f3
 
 
 
 
 
 
 
 
 
 
072f9dd
 
 
 
 
 
 
 
 
 
 
e6acaf6
072f9dd
 
 
 
e6acaf6
51482f3
 
072f9dd
e6acaf6
072f9dd
51482f3
072f9dd
 
51482f3
072f9dd
 
 
 
 
 
 
 
e6acaf6
072f9dd
 
e6acaf6
6224edc
072f9dd
 
 
 
6224edc
e6acaf6
072f9dd
 
 
 
 
 
 
 
407b252
 
072f9dd
e6acaf6
6224edc
072f9dd
 
96a1a44
072f9dd
51482f3
072f9dd
51482f3
072f9dd
51482f3
 
072f9dd
 
 
 
 
 
 
 
 
 
96a1a44
072f9dd
 
 
 
 
 
51482f3
 
 
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
import copy
import re
from io import BytesIO
from docx import Document
import numpy as np
from rag.parser import bullets_category, is_english, tokenize, remove_contents_table, hierarchical_merge, \
    make_colon_as_title
from rag.nlp import huqie
from rag.parser.docx_parser import HuDocxParser
from rag.parser.pdf_parser import HuParser
from rag.settings import cron_logger


class Docx(HuDocxParser):
    def __init__(self):
        pass

    def __clean(self, line):
        line = re.sub(r"\u3000", " ", line).strip()
        return line

    def __call__(self, filename, binary=None, from_page=0, to_page=100000):
        self.doc = Document(
            filename) if not binary else Document(BytesIO(binary))
        pn = 0
        lines = []
        for p in self.doc.paragraphs:
            if pn > to_page:break
            if from_page <= pn < to_page and p.text.strip(): lines.append(self.__clean(p.text))
            for run in p.runs:
                if 'lastRenderedPageBreak' in run._element.xml:
                    pn += 1
                    continue
                if 'w:br' in run._element.xml and 'type="page"' in run._element.xml:
                    pn += 1
        return [l for l in lines if l]


class Pdf(HuParser):
    def __call__(self, filename, binary=None, from_page=0,

                 to_page=100000, zoomin=3, callback=None):
        self.__images__(
            filename if not binary else binary,
            zoomin,
            from_page,
            to_page)
        callback(0.1, "OCR finished")

        from timeit import default_timer as timer
        start = timer()
        self._layouts_paddle(zoomin)
        callback(0.77, "Layout analysis finished")
        cron_logger.info("paddle layouts:".format((timer()-start)/(self.total_page+0.1)))
        self._naive_vertical_merge()

        callback(0.8, "Text extraction finished")

        return [b["text"] + self._line_tag(b, zoomin) for b in self.boxes]


def chunk(filename, binary=None, from_page=0, to_page=100000, callback=None, **kwargs):
    doc = {
        "docnm_kwd": filename,
        "title_tks": huqie.qie(re.sub(r"\.[a-zA-Z]+$", "", filename))
    }
    doc["title_sm_tks"] = huqie.qieqie(doc["title_tks"])
    pdf_parser = None
    sections = []
    if re.search(r"\.docx?$", filename, re.IGNORECASE):
        callback(0.1, "Start to parse.")
        for txt in Docx()(filename, binary):
            sections.append(txt)
        callback(0.8, "Finish parsing.")
    elif re.search(r"\.pdf$", filename, re.IGNORECASE):
        pdf_parser = Pdf()
        for txt in pdf_parser(filename if not binary else binary,
                         from_page=from_page, to_page=to_page, callback=callback):
            sections.append(txt)
    elif re.search(r"\.txt$", filename, re.IGNORECASE):
        callback(0.1, "Start to parse.")
        txt = ""
        if binary:txt = binary.decode("utf-8")
        else:
            with open(filename, "r") as f:
                while True:
                    l = f.readline()
                    if not l:break
                    txt += l
        sections = txt.split("\n")
        sections = txt.split("\n")
        sections = [l for l in sections if l]
        callback(0.8, "Finish parsing.")
    else: raise NotImplementedError("file type not supported yet(docx, pdf, txt supported)")

    # is it English
    eng = is_english(sections)
    # Remove 'Contents' part
    remove_contents_table(sections, eng)

    make_colon_as_title(sections)
    bull = bullets_category(sections)
    cks = hierarchical_merge(bull, sections, 3)
    if not cks: callback(0.99, "No chunk parsed out.")

    res = []
    # wrap up to es documents
    for ck in cks:
        print("\n-".join(ck))
        ck = "\n".join(ck)
        d = copy.deepcopy(doc)
        if pdf_parser:
            d["image"] = pdf_parser.crop(ck)
            ck = pdf_parser.remove_tag(ck)
        tokenize(d, ck, eng)
        res.append(d)
    return res


if __name__ == "__main__":
    import sys
    def dummy(a, b):
        pass
    chunk(sys.argv[1], callback=dummy)