我试图使用Python提取包含在这个PDF文件中的文本。

我正在使用PyPDF2包(版本1.27.2),并有以下脚本:

import PyPDF2

with open("sample.pdf", "rb") as pdf_file:
    read_pdf = PyPDF2.PdfFileReader(pdf_file)
    number_of_pages = read_pdf.getNumPages()
    page = read_pdf.pages[0]
    page_content = page.extractText()
print(page_content)

当我运行代码时,我得到以下输出,这与PDF文档中包含的输出不同:

 ! " # $ % # $ % &% $ &' ( ) * % + , - % . / 0 1 ' * 2 3% 4
5
 ' % 1 $ # 2 6 % 3/ % 7 / ) ) / 8 % &) / 2 6 % 8 # 3" % 3" * % 31 3/ 9 # &)
%

如何提取PDF文档中的文本?


当前回答

一种更健壮的方法,假设有多个PDF或只有一个!

import os
from PyPDF2 import PdfFileWriter, PdfFileReader
from io import BytesIO

mydir = # specify path to your directory where PDF or PDF's are

for arch in os.listdir(mydir): 
    buffer = io.BytesIO()
    archpath = os.path.join(mydir, arch)
    with open(archpath) as f:
            pdfFileObj = open(archpath, 'rb')
            pdfReader = PyPDF2.PdfFileReader(pdfFileObj)
            pdfReader.numPages
            pageObj = pdfReader.getPage(0) 
            ley = pageObj.extractText()
            file1 = open("myfile.txt","w")
            file1.writelines(ley)
            file1.close()
            

其他回答

Pdftotext是最好和最简单的一个! Pdftotext也保留了结构。

我尝试了PyPDF2, PDFMiner和其他一些程序,但没有一个能给出令人满意的结果。

在尝试textract(似乎有太多依赖项)和pypdf2(无法从我测试的pdf中提取文本)和tika(太慢)后,我最终使用xpdf中的pdftotext(正如已经在另一个答案中建议的那样),并直接从python中调用二进制(您可能需要调整路径到pdftotext):

import os, subprocess
SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__))
args = ["/usr/local/bin/pdftotext",
        '-enc',
        'UTF-8',
        "{}/my-pdf.pdf".format(SCRIPT_DIR),
        '-']
res = subprocess.run(args, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
output = res.stdout.decode('utf-8')

有pdftotext,它基本上相同,但这假设pdftotext在/usr/local/bin中,而我在AWS lambda中使用这个,并希望从当前目录使用它。

顺便说一句:要在lambda上使用这个,你需要把二进制文件和依赖项放到libstdc++中。到函数中。我个人需要编译xpdf。由于这方面的说明会让这个答案变得更糟,我把它们放在了我的个人博客上。

Pdfplumber是一个更好的从pdf中读取和提取数据的库。它还提供了读取表数据的方法,在经历了大量这样的库之后,pdfplumber最适合我。

请注意,它最适合机器编写的pdf,而不是扫描的pdf。

import pdfplumber
with pdfplumber.open(r'D:\examplepdf.pdf') as pdf:
first_page = pdf.pages[0]
print(first_page.extract_text())

看看PyPDF2<=1.26.0的代码:

import PyPDF2
pdf_file = open('sample.pdf', 'rb')
read_pdf = PyPDF2.PdfFileReader(pdf_file)
page = read_pdf.getPage(0)
page_content = page.extractText()
print page_content.encode('utf-8')

输出结果为:

!"#$%#$%&%$&'()*%+,-%./01'*23%4
5'%1$#26%3/%7/))/8%&)/26%8#3"%3"*%313/9#&)
%

使用相同的代码从201308FCR.pdf读取pdf .输出正常。

它的文档解释了原因:

def extractText(self):
    """
    Locate all text drawing commands, in the order they are provided in the
    content stream, and extract the text.  This works well for some PDF
    files, but poorly for others, depending on the generator used.  This will
    be refined in the future.  Do not rely on the order of text coming out of
    this function, as it will change if this function is made more
    sophisticated.
    :return: a unicode string object.
    """

试试borb,一个纯python PDF库

import typing  
from borb.pdf.document import Document  
from borb.pdf.pdf import PDF  
from borb.toolkit.text.simple_text_extraction import SimpleTextExtraction  


def main():

    # variable to hold Document instance
    doc: typing.Optional[Document] = None  

    # this implementation of EventListener handles text-rendering instructions
    l: SimpleTextExtraction = SimpleTextExtraction()  

    # open the document, passing along the array of listeners
    with open("input.pdf", "rb") as in_file_handle:  
        doc = PDF.loads(in_file_handle, [l])  
  
    # were we able to read the document?
    assert doc is not None  

    # print the text on page 0
    print(l.get_text(0))  

if __name__ == "__main__":
    main()