File size: 4,661 Bytes
fa82d94 6c8312a fa82d94 6c8312a b80e2f3 6c8312a b80e2f3 6c8312a |
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 |
#
# Copyright 2025 The InfiniFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import re
import csv
from copy import deepcopy
from deepdoc.parser.utils import get_text
from rag.app.qa import Excel
from rag.nlp import rag_tokenizer
def beAdoc(d, q, a, eng, row_num=-1):
d["content_with_weight"] = q
d["content_ltks"] = rag_tokenizer.tokenize(q)
d["content_sm_ltks"] = rag_tokenizer.fine_grained_tokenize(d["content_ltks"])
d["tag_kwd"] = [t.strip() for t in a.split(",") if t.strip()]
if row_num >= 0:
d["top_int"] = [row_num]
return d
def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs):
"""
Excel and csv(txt) format files are supported.
If the file is in excel format, there should be 2 column content and tags without header.
And content column is ahead of tags column.
And it's O.K if it has multiple sheets as long as the columns are rightly composed.
If it's in csv format, it should be UTF-8 encoded. Use TAB as delimiter to separate content and tags.
All the deformed lines will be ignored.
Every pair will be treated as a chunk.
"""
eng = lang.lower() == "english"
res = []
doc = {
"docnm_kwd": filename,
"title_tks": rag_tokenizer.tokenize(re.sub(r"\.[a-zA-Z]+$", "", filename))
}
if re.search(r"\.xlsx?$", filename, re.IGNORECASE):
callback(0.1, "Start to parse.")
excel_parser = Excel()
for ii, (q, a) in enumerate(excel_parser(filename, binary, callback)):
res.append(beAdoc(deepcopy(doc), q, a, eng, ii))
return res
elif re.search(r"\.(txt)$", filename, re.IGNORECASE):
callback(0.1, "Start to parse.")
txt = get_text(filename, binary)
lines = txt.split("\n")
comma, tab = 0, 0
for line in lines:
if len(line.split(",")) == 2:
comma += 1
if len(line.split("\t")) == 2:
tab += 1
delimiter = "\t" if tab >= comma else ","
fails = []
content = ""
i = 0
while i < len(lines):
arr = lines[i].split(delimiter)
if len(arr) != 2:
content += "\n" + lines[i]
elif len(arr) == 2:
content += "\n" + arr[0]
res.append(beAdoc(deepcopy(doc), content, arr[1], eng, i))
content = ""
i += 1
if len(res) % 999 == 0:
callback(len(res) * 0.6 / len(lines), ("Extract TAG: {}".format(len(res)) + (
f"{len(fails)} failure, line: %s..." % (",".join(fails[:3])) if fails else "")))
callback(0.6, ("Extract TAG: {}".format(len(res)) + (
f"{len(fails)} failure, line: %s..." % (",".join(fails[:3])) if fails else "")))
return res
elif re.search(r"\.(csv)$", filename, re.IGNORECASE):
callback(0.1, "Start to parse.")
txt = get_text(filename, binary)
lines = txt.split("\n")
fails = []
content = ""
res = []
reader = csv.reader(lines)
for i, row in enumerate(reader):
row = [r.strip() for r in row if r.strip()]
if len(row) != 2:
content += "\n" + lines[i]
elif len(row) == 2:
content += "\n" + row[0]
res.append(beAdoc(deepcopy(doc), content, row[1], eng, i))
content = ""
if len(res) % 999 == 0:
callback(len(res) * 0.6 / len(lines), ("Extract Tags: {}".format(len(res)) + (
f"{len(fails)} failure, line: %s..." % (",".join(fails[:3])) if fails else "")))
callback(0.6, ("Extract TAG : {}".format(len(res)) + (
f"{len(fails)} failure, line: %s..." % (",".join(fails[:3])) if fails else "")))
return res
raise NotImplementedError(
"Excel, csv(txt) format files are supported.")
if __name__ == "__main__":
import sys
def dummy(prog=None, msg=""):
pass
chunk(sys.argv[1], from_page=0, to_page=10, callback=dummy) |