178 lines
5.5 KiB
Python
178 lines
5.5 KiB
Python
import os
|
|
import codecs
|
|
from datetime import datetime
|
|
from langchain_community.document_loaders.csv_loader import CSVLoader
|
|
from langchain_community.document_loaders.text import TextLoader
|
|
from docx import Document
|
|
from ebooklib import epub
|
|
from bs4 import BeautifulSoup
|
|
from pptx import Presentation
|
|
from openpyxl import load_workbook
|
|
import mobi
|
|
import PyPDF2
|
|
import html2text
|
|
|
|
class MyMobiLoader:
|
|
def __init__(self, file_path):
|
|
self.file_path = file_path
|
|
|
|
def load(self):
|
|
tempdir, filepath = mobi.extract(self.file_path)
|
|
with codecs.open(filepath, "r", "utf-8") as f:
|
|
content=f.read()
|
|
return html2text.html2text(content)
|
|
|
|
class MyChmLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
self.chm_file = None
|
|
|
|
def load(self):
|
|
"""Reads the CHM file and returns concatenated text from all text/html entries."""
|
|
self.chm_file = CHM.CHMFile(self.filepath)
|
|
content = []
|
|
|
|
def callback(chm_item):
|
|
if chm_item[0].endswith(('.html', '.htm', '.txt')):
|
|
try:
|
|
data = self.chm_file.read_file(chm_item[0])
|
|
if data:
|
|
content.append(data.decode('utf-8', errors='ignore'))
|
|
except Exception:
|
|
pass
|
|
return True
|
|
|
|
self.chm_file.walk(callback)
|
|
return '\n'.join(content)
|
|
|
|
class MyPdfLoader:
|
|
def __init__(self, file_path, **kw):
|
|
self.filepath = file_path
|
|
self.reader = None
|
|
|
|
def load(self):
|
|
"""Reads the PDF file and returns all text as a single string."""
|
|
text = ''
|
|
with open(self.filepath, 'rb') as file:
|
|
self.reader = PyPDF2.PdfReader(file)
|
|
for page in self.reader.pages:
|
|
text += page.extract_text() or ''
|
|
return ' '.join(text.split('\t'))
|
|
|
|
class MyDocLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
|
|
def load(self):
|
|
"""Extract plain text from a .doc file using python-doc."""
|
|
with open(self.filepath, 'rb') as f:
|
|
document = python_doc.Document(f)
|
|
return document.text()
|
|
|
|
class MyDocxLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
self.document = None
|
|
|
|
def load(self):
|
|
"""Reads the .docx file and returns the full text as a single string."""
|
|
self.document = Document(self.filepath)
|
|
text = '\n'.join([para.text for para in self.document.paragraphs])
|
|
return text
|
|
|
|
class MyPptLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
|
|
def load(self):
|
|
prs = Presentation(self.filepath)
|
|
text = []
|
|
for slide in prs.slides:
|
|
for shape in slide.shapes:
|
|
if hasattr(shape, "text"):
|
|
text.append(shape.text)
|
|
return "\n".join(text)
|
|
|
|
class MyCsvLoader(CSVLoader):
|
|
def load(self):
|
|
docs = super().load()
|
|
return ' '.join([d.page_content for d in docs])
|
|
|
|
class MyExcelLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
self.workbook = None
|
|
|
|
def load(self):
|
|
"""Reads all sheets in the Excel file and returns the content as a string."""
|
|
self.workbook = load_workbook(filename=self.filepath, data_only=True)
|
|
content = []
|
|
|
|
for sheet in self.workbook.worksheets:
|
|
content.append(f"--- Sheet: {sheet.title} ---")
|
|
for row in sheet.iter_rows(values_only=True):
|
|
row_text = '\t'.join(str(cell) if cell is not None else '' for cell in row)
|
|
content.append(row_text)
|
|
|
|
return '\n'.join(content)
|
|
|
|
class MyEpubLoader:
|
|
def __init__(self, file_path):
|
|
self.filepath = file_path
|
|
self.book = None
|
|
|
|
def load(self):
|
|
"""Reads the EPUB file and returns all text content as a string."""
|
|
self.book = epub.read_epub(self.filepath)
|
|
content = []
|
|
|
|
for item in self.book.get_items():
|
|
if isinstance(item, epub.EpubHtml):
|
|
soup = BeautifulSoup(item.get_content(), 'html.parser')
|
|
text = soup.get_text()
|
|
content.append(text.strip())
|
|
|
|
return '\n\n'.join(content)
|
|
|
|
class MyTextLoader(TextLoader):
|
|
def load(self):
|
|
docs = super().load()
|
|
return '\n'.join([d.page_content for d in docs])
|
|
|
|
def fileloader(file_path):
|
|
# Load the PDF file and split the data into chunks
|
|
data = None
|
|
if file_path.lower().endswith('.pdf'):
|
|
loader = MyPdfLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.docx'):
|
|
loader = MyDocxLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.doc'):
|
|
raise Exception(f'not supported file({file_path}')
|
|
# loader = MyDocLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.pptx') or file_path.lower().endswith('.pptx'):
|
|
loader = MyPptLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.xlsx') or file_path.lower().endswith('.xls'):
|
|
loader = MyExcelLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.csv'):
|
|
loader = MyCsvLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.epub'):
|
|
loader = MyEpubLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.chm'):
|
|
raise Exception(f'not supported file({file_path}')
|
|
# loader = MyChmLoader(file_path=file_path)
|
|
elif file_path.lower().endswith('.mobi'):
|
|
loader = MyMobiLoader(file_path=file_path)
|
|
else:
|
|
loader = MyTextLoader(file_path=file_path)
|
|
data = loader.load()
|
|
return data
|
|
|
|
if __name__ == '__main__':
|
|
import sys
|
|
if len(sys.argv) < 2:
|
|
print(f'{sys.argv[0]} file\nload a file and get its text')
|
|
sys.exit(1)
|
|
text = fileloader(sys.argv[1])
|
|
print(f'{text}')
|
|
|