使用pdfminer水平阅读pdf文件 [英] read pdf file horizontally with pdfminer

问题描述

我想用pdfminer(版本20140328)提取pdf.

I would like to extract a pdf with pdfminer (version 20140328).

这是提取pdf的代码:

This is the code to extract the pdf:

import sys
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.pdfpage import PDFPage
from pdfminer.converter import XMLConverter, HTMLConverter, TextConverter
from pdfminer.layout import LAParams
from cStringIO import StringIO
import urllib2

def pdf_to_string(data):
    fp = StringIO(data)
    rsrcmgr = PDFResourceManager()
    retstr = StringIO()
    codec = 'utf-8'
    laparams = LAParams()
    device = TextConverter(rsrcmgr, retstr, codec=codec, laparams=laparams)
    # Create a PDF interpreter object.
    interpreter = PDFPageInterpreter(rsrcmgr, device)

    # Process each page contained in the document.
    for page in PDFPage.get_pages(fp):
        interpreter.process_page(page)
        data =  retstr.getvalue()

    return data

pdf_url="http://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ecofin/140836.pdf"
file_object = urllib2.urlopen(urllib2.Request(pdf_url)).read()
string=pdf_to_string(file_object)

这是pdf的屏幕截图:

This is a screenshot of the pdf:

问题是pdfminer不会水平读取(先人然后位置)，而是按列读取(然后所有人都将其各自的位置读取):

The problem is that pdfminer doesn't read it horizontally (person then position) but in columns (all the persons then all their respective positions):

Belgium: 
Mr Koen GEENS 

Bulgaria: 
Mr Petar CHOBANOV 

Czech Republic: 
Mr Radek URBAN 


Minister for Finance, with responsibility for the Civil 
Service 

Minister for Finance 

Deputy Minister for Finance 

如何使pdfminer水平阅读文本?

推荐答案

我找到了 pdftotext 的可行解决方案:

I have found a working solution with pdftotext:

import tempfile, subprocess

def pdf_to_string(file_object):
    pdfData = file_object.read()
    tf = tempfile.NamedTemporaryFile()
    tf.write(pdfData)
    tf.seek(0)
    outputTf = tempfile.NamedTemporaryFile()

    if (len(pdfData) > 0) :
        out, err = subprocess.Popen(["pdftotext", "-layout", tf.name, outputTf.name ]).communicate()
        return outputTf.read()
    else :
        return None

pdf_file="files/2014_1.pdf"
file_object = file(pdf_file, 'rb')
print pdf_to_string(file_object)

这将产生正确的输出，然后是人员姓名，然后是位置:).

This produces the right output, person names then positions :).

已解决！

这篇关于使用pdfminer水平阅读pdf文件的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！