File size: 4,367 Bytes
9c6b97d | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 | import json
import os
# import PyPDF2
import pandas as pd
from application import *
'''
following functions are for file manipulation
'''
# read pdf file and return text
def read_pdf(file_path):
# open the pdf file
try:
filename = file_path
pdfFileObj = open(file_path, 'rb')
except TypeError:
filename = file_path.name
pdfFileObj = open(file_path.name, 'rb')
# create a pdf reader object
pdfReader = PyPDF2.PdfReader(pdfFileObj)
# get the number of pages in the pdf file
num_pages = len(pdfReader.pages)
# create an empty string
text = ''
# iterate through all the pages
for page_num in range(num_pages):
page_obj = pdfReader.pages[page_num]
text += page_obj.extract_text ()
# close the pdf file object
pdfFileObj.close()
text = remove_symbols(text)
with open(f"{filename.split('.')[0]}.txt", "w") as f:
f.write(text)
# return the string of text
return text, pdfReader.metadata
'''
following functions are for format standard response
'''
# format standard response for status code and data
def format_response(code,data):
return {
"statusCode":code,
"headers":{
"Access-Control-Allow-Origin": "*",
"Content-Type": "application/json"
},
"body":json.dumps(data),
"isBase64Encoded": False
}
'''
following functions are for string manipulation
'''
# format text output by removing excessive characters
def format_text(text,remove_char_ls = ["\\n--\\n","\\n\\n","\n"]):
for c in remove_char_ls:
text = text.replace(c,"")
return text
# function to remove symbols that are not in unicode
def remove_symbols(text):
# remove symbols that are not in unicode
text = text.encode("ascii", "ignore").decode()
# remove the break word new line return
text = text.replace('-\n', '')
return text
def str_to_tuple(s):
return tuple(s.replace("(","").replace(")","").split(","))
'''
following functions are for dynamodb data manipulation
'''
# convert dynamodb map to python dictionary
def db_map_to_py_dict(db_map):
py_dict = {}
for k,i in db_map.items():
for l,v in i.items():
if l == "M":
py_dict[k] = db_map_to_py_dict(v)
elif l == "S":
py_dict[k] = v
elif l == "N":
py_dict[k] = int(v) if float(v)%1 ==0 else float(v)
elif l == "L":
py_dict[k] = db_list_to_py_list(v)
else:
py_dict[k] = v
return py_dict
def db_map_to_pd_dataframe(db_map):
py_dict = db_map_to_py_dict(db_map)
return pd.DataFrame(py_dict)
# convert python dictionary to dynamodb map
def py_dict_to_db_map(py_dict):
db_map = {}
for key,value in py_dict.items():
key = str(key)
if type(value) is str:
db_map[key] = {"S":value}
elif type(value) is int or type(value) is float:
db_map[key] = {"N":value}
elif type(value) is dict:
db_map[key] = {"M":py_dict_to_db_map(value)}
elif type(value) is list:
db_map[key] = {"L":py_list_to_db_list(value)}
return db_map
# convert dynamodb list to python list
def db_list_to_py_list(db_list):
py_list = []
for d in db_list:
for t,v in d.items():
if t == "M":
py_list.append(db_map_to_py_dict(v))
elif t == "L":
py_list.append(db_list_to_py_list(v))
else:
py_list.append(v)
return py_list
# convert python list to dynamodb list
def py_list_to_db_list(py_list):
db_list = []
for value in py_list:
if type(value) is str:
item = {"S":value}
elif type(value) is int or float:
item = {"N":value}
elif type(value) is dict:
item = {"M":py_dict_to_db_map(value)}
elif type(value) is list:
item = {"L":py_list_to_db_list(value)}
db_list.append(item)
return db_list
'''
following functions are used for business logic. (to be moved to business logic layer)
'''
# function to calculate the estimated cost of the translation
def est_cost(n_tokens,rate):
return round(rate*n_tokens/1000,4)
|