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 | class PreprocessReader(BaseReader):
def __init__(self, api_key: str, *args, **kwargs):
if api_key is None or api_key == "":
raise ValueError(
"Please provide an api key to be used while doing the auth with the system."
)
try:
from pypreprocess import Preprocess
except ImportError:
raise ImportError(
"`pypreprocess` package not found, please run `pip install"
" pypreprocess`"
)
_info = {}
self._preprocess = Preprocess(api_key)
self._filepath = None
self._process_id = None
for key, value in kwargs.items():
if key == "filepath":
self._filepath = value
self._preprocess.set_filepath(value)
if key == "process_id":
self._process_id = value
self._preprocess.set_process_id(value)
elif key in ["table_output_format", "table_output"]:
_info["table_output_format"] = value
elif key in ["repeat_table_header", "table_header"]:
_info["repeat_table_header"] = value
elif key in [
"merge",
"repeat_title",
"keep_header",
"keep_footer",
"smart_header",
"image_text",
]:
_info[key] = value
if _info != {}:
self._preprocess.set_info(_info)
if self._filepath is None and self._process_id is None:
raise ValueError(
"Please provide either filepath or process_id to handle the resutls."
)
self._chunks = None
def load_data(self, return_whole_document=False) -> List[Document]:
if self._chunks is None:
if self._process_id is not None:
self._get_data_by_process()
elif self._filepath is not None:
self._get_data_by_filepath()
if self._chunks is not None:
if return_whole_document is True:
return [
Document(
text=" ".join(self._chunks),
metadata={"filename": os.path.basename(self._filepath)},
)
]
else:
return [
Document(
text=chunk,
metadata={"filename": os.path.basename(self._filepath)},
)
for chunk in self._chunks
]
else:
raise Exception(
"There is error happened during handling your file, please try again."
)
else:
if return_whole_document is True:
return [
Document(
text=" ".join(self._chunks),
metadata={"filename": os.path.basename(self._filepath)},
)
]
else:
return [
Document(
text=chunk,
metadata={"filename": os.path.basename(self._filepath)},
)
for chunk in self._chunks
]
def get_process_id(self):
return self._process_id
def get_nodes(self) -> List[TextNode]:
if self._chunks is None:
self.load_data()
nodes = []
for chunk in self._chunks:
text = str(chunk)
id = hashlib.md5(text.encode()).hexdigest()
nodes.append(TextNode(text=text, id_=id))
if len(nodes) > 1:
nodes[0].relationships[NodeRelationship.NEXT] = RelatedNodeInfo(
node_id=nodes[1].node_id,
metadata={"filename": os.path.basename(self._filepath)},
)
for i in range(1, len(nodes) - 1):
nodes[i].relationships[NodeRelationship.NEXT] = RelatedNodeInfo(
node_id=nodes[i + 1].node_id,
metadata={"filename": os.path.basename(self._filepath)},
)
nodes[i].relationships[NodeRelationship.PREVIOUS] = RelatedNodeInfo(
node_id=nodes[i - 1].node_id,
metadata={"filename": os.path.basename(self._filepath)},
)
nodes[-1].relationships[NodeRelationship.PREVIOUS] = RelatedNodeInfo(
node_id=nodes[-2].node_id,
metadata={"filename": os.path.basename(self._filepath)},
)
return nodes
def _get_data_by_filepath(self) -> None:
pp_response = self._preprocess.chunk()
if pp_response.status == "OK" and pp_response.success is True:
self._process_id = pp_response.data["process"]["id"]
response = self._preprocess.wait()
if response.status == "OK" and response.success is True:
# self._filepath = response.data['info']['file']['name']
self._chunks = response.data["chunks"]
def _get_data_by_process(self) -> None:
response = self._preprocess.wait()
if response.status == "OK" and response.success is True:
self._filepath = response.data["info"]["file"]["name"]
self._chunks = response.data["chunks"]
|