neo4j_api/pdf_txt.py

# import fitz  # PyMuPDF
# import re
# import os
#
#
# def read_pdf_text(pdf_path):
#     # 打开 PDF 文件
#     document = fitz.open(pdf_path)
#
#     # 初始化一个空字符串来存储所有页面的文本
#     pdf_text = ''
#
#     # 遍历每一页
#     for page_num in range(len(document)):
#         page = document.load_page(page_num)  # 加载页面
#         text = page.get_text("text")  # 提取页面上的文本
#         pdf_text += text + '\n'  # 将文本添加到总文本中，每页之间用换行分隔
#
#     return pdf_text
#
#
# def clean_text(text):
#     # 1. 移除段内不必要的换行
#     # 将段落内的换行符替换为空格
#     text = re.sub(r'(\S)\n(\S)', r'\1 \2', text)
#
#     # 2. 保留段落之间的换行
#     # 合并连续的空行
#     cleaned_text = re.sub(r'\n\s*\n+', '\n\n', text)
#
#     # 3. 去除每行开头和结尾的空白字符
#     cleaned_text = re.sub(r'^\s+|\s+$', '', cleaned_text, flags=re.MULTILINE)
#
#     return cleaned_text
#
#
# def remove_unwanted_content(text):
#     # 1. 去除目录
#     text = re.sub(r'^目录.*?\n', '', text, flags=re.DOTALL | re.MULTILINE)
#
#     # 2. 去除页眉和页脚
#     # 假设页眉和页脚通常是每页的第一行和最后一行
#     lines = text.split('\n')
#     new_lines = []
#     for i in range(0, len(lines), 2):  # 每两行处理一次
#         if i + 1 < len(lines):
#             # 去除页眉和页脚
#             new_lines.append(lines[i + 1])
#     text = '\n'.join(new_lines)
#
#     # 3. 去除注释
#     text = re.sub(r'$.*?$', '', text)  # 去除括号内的注释
#     text = re.sub(r'\*.*?\*', '', text)  # 去除星号内的注释
#
#     # 4. 去除空行和多余的空白
#     text = re.sub(r'\n\s*\n+', '\n\n', text)
#
#     return text
#
#
# def format_special_terms(text):
#     # 识别图、表、Figure、Table等关键词，并在它们之前插入换行符
#     text = re.sub(r'(?<!\n)(图|表|Figure|Table)\s+', r'\n\g<1> ', text)
#     return text
#
#
# def save_text_as_utf8(text, output_path):
#     # 将文本以 UTF-8 格式保存到文件
#     with open(output_path, 'w', encoding='utf-8') as file:
#         file.write(text)
#
#
# def main(pdf_path):
#     # 提取 PDF 文件名
#     pdf_filename = os.path.basename(pdf_path)
#     pdf_name, _ = os.path.splitext(pdf_filename)
#
#     # 创建新文件夹
#     output_dir = 'pdf_output'
#     if not os.path.exists(output_dir):
#         os.makedirs(output_dir)
#
#     # 构建输出文件路径
#     output_path = os.path.join(output_dir, f"{pdf_name}.txt")
#
#     # 提取文本
#     pdf_text = read_pdf_text(pdf_path)
#
#     # 清理文本
#     cleaned_text = clean_text(pdf_text)
#
#     # 去除不需要的信息
#     cleaned_text = remove_unwanted_content(cleaned_text)
#
#     # 格式化特殊术语
#     final_text = format_special_terms(cleaned_text)
#
#     # 保存清理后的文本
#     save_text_as_utf8(final_text, output_path)
#
#     print(f"Text extracted, cleaned, and saved to {output_path} in UTF-8 format.")
#
#
# if __name__ == "__main__":
#     pdf_path = r'C:\Users\Machenike\Desktop\火工大\hgd240914\1图像目标智能识别软件设计方案.pdf'  # 请确保这里的路径正确指向您的 PDF 文件
#     main(pdf_path)

# import pdfplumber
#
# def read_chinese_pdf_text(pdf_path):
#     text = ""
#     with pdfplumber.open(pdf_path) as pdf:
#         for page in pdf.pages:
#             text += page.extract_text()
#         return text
# if __name__ == '__main__':
#     path =  r'C:\Users\Machenike\Desktop\火工大\hgd240914\1图像目标智能识别软件设计方案.pdf'
#     data = read_chinese_pdf_text(path)
#     print(data)
import os
import pdfplumber
import re


def read_chinese_pdf_text(pdf_path):
    project_root = os.getcwd()
    output_dir = os.path.join(project_root, 'output')
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)

    base_name = os.path.splitext(os.path.basename(pdf_path))[0]
    output_file = os.path.join(output_dir, f"{base_name}.txt")

    text = ""
    with pdfplumber.open(pdf_path) as pdf:
        for page in pdf.pages:
            height = page.height
            crop_area = (0, height * 0.1, page.width, height * 0.9)
            cropped_page = page.crop(crop_area)

            page_text = cropped_page.extract_text() or ''
            lines = page_text.split('\n')

            processed_lines = []
            current_line = ""
            for line in lines:
                line = line.strip()  # 去除每行首尾的空白字符
                if len(line) == 0:  # 如果是空行，则直接添加换行
                    if current_line:
                        processed_lines.append(current_line)
                        current_line = ""
                    processed_lines.append('')
                    continue

                # 检查是否是表格或图片的名称
                if re.match(r'(表|图)\s*\d+', line):
                    if current_line:
                        processed_lines.append(current_line)
                        current_line = ""
                    processed_lines.append(line)
                    processed_lines.append('')
                    continue

                # 判断是否需要合并行
                if (not current_line.endswith(('。', '！', '？', '.', '!', '?', ','))) and not line.startswith('  '):
                    # 当前行不是句子结尾，且下一行没有缩进，合并这两行
                    current_line += line
                else:
                    if current_line:
                        processed_lines.append(current_line)
                    current_line = line

            if current_line:  # 添加最后一行
                processed_lines.append(current_line)

            text += '\n'.join(processed_lines) + '\n'

    with open(output_file, 'w', encoding='utf-8') as file:
        file.write(text)

    return output_file


if __name__ == '__main__':
    path = r'C:\Users\Machenike\Desktop\火工大\hgd240914\1图像目标智能识别软件设计方案.pdf'
    result_file = read_chinese_pdf_text(path)
    print(f"Text has been extracted and saved to {result_file}")